阅读说明：本技术 一种高压直流断路器运行状态推衍方法和系统 (High-voltage direct-current circuit breaker operation state derivation method and system ) 是由 曹楠 栾洪洲 施秀萍 李天琦 黄远超 陈翔宇 郭宁明 杜向楠 于 2021-05-28 设计创作，主要内容包括：本发明提供了一种高压直流断路器运行状态推衍方法和系统,包括：采集高压直流断路器的特征子集中各特征参数的特征值；基于所述特征值,分别计算高压直流断路器特征参量的相关系数矩阵和自增减趋势矩阵；利用所述相关系数矩阵和自增减趋势矩阵,对高压直流断路器运行状态进行推衍；其中,所述高压直流断路器的特征子集,是基于最大相关原则、最小冗余原则、多准则赋权排序算法和关联算法,对高压直流断路器的初始特征量集合中的特征参量进行筛选得到的。本发明为运行人员及时了解高压直流断路器运行状态、捕捉高压直流断路器早期缺陷、科学制定状态检修计划提供数据支撑,为高压直流断路器整体运行效率的提升和可靠性提高提供了有力的技术支撑。(The invention provides a method and a system for deriving the running state of a high-voltage direct-current circuit breaker, which comprise the following steps: collecting characteristic values of characteristic parameters in a characteristic subset of the high-voltage direct-current circuit breaker; respectively calculating a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameters of the high-voltage direct-current circuit breaker based on the characteristic values; deducing the running state of the high-voltage direct-current circuit breaker by utilizing the correlation coefficient matrix and the self-increasing and self-decreasing trend matrix; the characteristic subset of the high-voltage direct-current circuit breaker is obtained by screening characteristic parameters in an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion weighting sorting algorithm and a correlation algorithm. The invention provides data support for operators to know the operation state of the high-voltage direct-current circuit breaker in time, catch early defects of the high-voltage direct-current circuit breaker and scientifically make a state maintenance plan, and provides powerful technical support for improving the overall operation efficiency and reliability of the high-voltage direct-current circuit breaker.)

1. An operation state derivation method for a high-voltage direct current breaker, the method comprising:

collecting characteristic values of characteristic parameters in a characteristic subset of the high-voltage direct-current circuit breaker;

respectively calculating a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameters of the high-voltage direct-current circuit breaker based on the characteristic values;

deducing the running state of the high-voltage direct-current circuit breaker by utilizing the correlation coefficient matrix and the self-increasing and self-decreasing trend matrix;

the characteristic subset of the high-voltage direct-current circuit breaker is obtained by screening characteristic parameters in an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion weighting sorting algorithm and a correlation algorithm.

2. The method of claim 1, wherein the initial set of feature quantities consists of multidimensional features, each dimensional feature consisting of a plurality of feature quantities.

3. The method according to claim 2, wherein the process of screening the initial characteristic quantity set of the high voltage direct current breaker based on the maximum correlation principle, the minimum redundancy principle, the multi-criterion weighted ranking algorithm and the correlation algorithm comprises:

based on a maximum correlation principle, a minimum redundancy principle and a multi-criterion empowerment sorting algorithm, screening the characteristic parameters in the initial characteristic quantity set to obtain a set Q;

and (4) removing the characteristic parameters with the support degree and the confidence degree smaller than the threshold requirement in the set Q by using a correlation algorithm to obtain the characteristic subset.

4. The method as claimed in claim 3, wherein the screening the feature parameters in the initial feature parameter set to obtain a set Q based on a maximum correlation rule, a minimum redundancy rule, and a multi-criteria weighted ranking algorithm comprises:

step 1): making the set Q as an empty set;

step 2): based on the maximum correlation principle, selecting a characteristic parameter with the maximum mean weight of the category separability measurement from the initial characteristic parameter set, and putting the characteristic parameter into a set Q;

step 3): when the characteristic parameters in the initial characteristic quantity set are completely redundant with the characteristic parameters in the set Q, deleting the characteristic parameters redundant with the characteristic parameters in the initial characteristic quantity set and the set Q based on the minimum redundancy principle;

step 4): calculating the redundancy between each characteristic parameter in the initial characteristic quantity set and the set Q;

step 5): calculating the state representation degree of each characteristic parameter in the initial characteristic quantity set by using the redundancy between each characteristic parameter in the initial characteristic quantity set and the set Q, and selecting one characteristic parameter with the highest state representation degree from the initial characteristic quantity set to be put into the set Q based on a multi-criterion empowerment sorting algorithm;

step 6: if the characteristic parameters exist in the initial characteristic quantity set, returning to the step 3); otherwise, the set Q is output.

5. The method of claim 4, wherein the mean weight of the class separability measures of the feature quantities in the initial set of feature quantities is calculated as follows:

in the formula (I), the compound is shown in the specification,is the mean weight, avg (f) of the class separability measure of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set_ik) Is the mean value, avg, of the historical sampling value of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set_i(F_i) Is the average value, x, of the historical sampling values of all the characteristic parameters under the ith dimension characteristic in the initial characteristic quantity set_f(f_ik) Is the f-th historical sampling value of the k-th characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set, the total number of historical sampling values of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set is i belongs to (1-N), N is the total number of dimensions in the initial characteristic quantity set, and k belongs to (1-N)_i)，N_iIs the total number of the feature parameters included in the ith dimension feature in the initial feature quantity set.

6. The method of claim 4, wherein the identification process of the redundancy between the characteristic parameters in the initial characteristic quantity set and the characteristic parameters in the set Q comprises:

if the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), characteristic parameter g in set Q_zInformation entropy of (2) and feature parameter f in initial feature quantity set_jAnd the characteristic parameter g in the set Q_zThe joint entropy of the initial feature quantity set is equal, then the feature quantity f in the initial feature quantity set is equal_jAnd the characteristic parameter g in the set Q_zComplete redundancy, otherwise, the characteristic parameter f in the initial characteristic quantity set_jAnd the characteristic parameter g in the set Q_zThere is no redundancy;

wherein, the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), characteristic parameter g in set Q_zInformation entropy of (2) and feature parameter f in initial feature quantity set_jAnd the characteristic parameter g in the set Q_zIs based on the first characteristic parameter f in the initial characteristic quantity set_jAnd/or the characteristic quantity g in the set Q_zDetermined from historical sample values of g_z∈Q，f_jAnd e is U, and U is an initial characteristic quantity set.

7. The method of claim 4, wherein the redundancy between each feature in the initial set of features and the set Q is calculated as follows:

in the formula (I), the compound is shown in the specification,for the characteristic parameter f in the initial characteristic quantity set_jRedundancy with respect to set Q, I (f)_j,g_z) Characteristic parameters in an initial characteristic quantity set and characteristic parameters g in a set Q_zMutual information between, g_z∈Q，f_jE is U, and U is an initial characteristic quantity set;

wherein, the characteristics in the initial characteristic quantity set participate in the characteristics in the set QAmount g_zMutual information I (f) between_j,g_z) Is calculated as follows:

I(f_j,g_z)＝H(f_j)+H(g_z)-H(f_j,g_z)

in the formula, H (f)_j) For the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), H (g)_z) For the characteristic quantity g in the set Q_zEntropy of (d), H (f)_j,g_z) For the characteristic parameter f in the initial characteristic quantity set_jAnd the characteristic parameter g in the set Q_zThe joint entropy of (a).

8. The method according to claim 4, wherein the degree of state representation of each feature in the initial set of features is calculated as follows:

wherein J (f)_j) For the characteristic parameter f in the initial characteristic quantity set_jThe degree of the state characterization of (a),is the mean weight of the class separability measure of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set,for the characteristic parameter f in the initial characteristic quantity set_jAnd the set Q.

9. The method of claim 1, wherein the collecting the characteristic values of the characteristic parameters in the characteristic subset of the high voltage direct current circuit breaker comprises:

collecting the characteristic values of each characteristic parameter in the characteristic subset at the current moment and r-1 moments before the current moment;

wherein r is the total number of time instants contained in the sampling time window.

10. The method of claim 9, wherein calculating a correlation coefficient matrix of the characteristic parameter of the high voltage direct current breaker based on the characteristic value comprises:

and determining a correlation coefficient matrix of the characteristic parameters of the high-voltage direct-current circuit breaker at the current moment by using a correlation coefficient calculation method based on the characteristic values.

11. The method of claim 10, wherein calculating a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker based on the characteristic value comprises:

according to the characteristic values, characteristic value matrixes of the high-voltage direct-current circuit breaker at the current moment and h-1 moments before the current moment are respectively determined;

determining a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at h-1 moments before the current moment by using the characteristic value matrix;

estimating a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the current moment by adopting a least square method based on the self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at h-1 moments before the current moment;

the self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment is obtained by subtracting the characteristic value matrix of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment from the characteristic value matrix of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment;

the characteristic value matrixes of the high-voltage direct-current circuit breaker at the current moment and r-1 moments before the current moment are x row 1 column matrixes, alpha row elements of the matrix are characteristic values of alpha characteristic parameters in the characteristic subset at the current moment and r-1 moments before the current moment, x is the total number of the characteristic parameters in the characteristic subset, h is a positive integer smaller than r, w is a positive integer smaller than h-1, and h is the total number of moments contained in a derivation time window.

12. The method of claim 11, wherein the deriving the operating state of the high voltage direct current circuit breaker using the correlation coefficient matrix and the auto-increment and decrement trend matrix comprises:

taking the sum of the product of a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the current moment and the characteristic value matrix of the high-voltage direct-current circuit breaker at the current moment as a characteristic value matrix of the high-voltage direct-current circuit breaker at a moment after the current moment;

the characteristic value matrix of the high-voltage direct current breaker at a moment after the current moment represents the operation state of the high-voltage direct current breaker at a moment after the current moment.

13. The method of claim 2, wherein the feature dimensions of the initial set of feature quantities include, but are not limited to: thermal faults, voltage withstand breakdown faults, electrical circuit faults, transient faults, secondary system faults, water cooling system faults and energy supply system faults;

the characteristic parameters under thermal failure include but are not limited to: IGBT junction temperature, mechanical switch contact temperature, submodule temperature, MOV overheating protection action, overheating alarm and protection action information;

the characteristic parameters under the fault of the electric circuit include but are not limited to: the method comprises the following steps that sub-module voltage, sub-module current, main branch current, transfer branch current, lightning arrester leakage current, main branch abnormal state information, transfer branch abnormal state information and energy consumption branch abnormal state information are obtained;

the characteristic parameters under the transient fault include, but are not limited to: system voltage, system current, breaking time, dissipation energy, abnormal state of the rapid mechanical switch and conversion overtime;

the characteristic parameters under the secondary system fault include but are not limited to: communication abnormality information, protection action information, and control abnormality information;

the characteristic parameters under the fault of the water cooling system include but are not limited to: the water cooling system fault state and water leakage detection device gives an alarm;

the characteristic parameters under the fault of the energy supply system include but are not limited to: and energy supply system fault information.

14. An operating state deriving system for a high voltage direct current circuit breaker, the system comprising:

the acquisition module is used for acquiring the characteristic values of all characteristic parameters in the characteristic subset of the high-voltage direct-current circuit breaker;

the calculation module is used for respectively calculating a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameters of the high-voltage direct-current circuit breaker based on the characteristic values;

the derivation module is used for deriving the running state of the high-voltage direct-current circuit breaker by utilizing the correlation coefficient matrix and the self-increasing and self-decreasing trend matrix;

the characteristic subset of the high-voltage direct-current circuit breaker is obtained by screening characteristic parameters in an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion weighting sorting algorithm and a correlation algorithm.

Technical Field

The invention relates to the field of on-line monitoring of direct current equipment, in particular to a method and a system for deducing the running state of a high-voltage direct current breaker.

Background

The high-voltage direct-current circuit breaker is comprehensive electrical equipment formed by effectively integrating various devices with different electrical characteristics through a reasonable connection mode and set operation logics, and has the current-carrying and insulating capabilities of a mechanical switch and the breaking capability of a solid-state switch.

In the steady-state operation process of the high-voltage direct-current circuit breaker, analyzing the change trend of the whole operation state of the high-voltage direct-current circuit breaker, predicting equipment faults and analyzing the life cycle state are one of the important tasks of on-line monitoring of the high-voltage direct-current circuit breaker. Because the characteristic parameters of the power electronic device have the characteristics of slow state change, unobvious deviation characteristics, easy unknown disturbance and noise covering and the like in the degradation process, the degradation process of the power electronic device in the steady-state operation process of the high-voltage direct-current circuit breaker is difficult to depict by the conventional monitoring means, and the potential fault which endangers the safe operation of equipment can be formed due to the accumulation of the device degradation time.

The high-voltage direct-current circuit breaker is complex in topological structure design, and the vulnerability of equipment operation is increased due to the application of a large number of power electronic devices and switching devices. The incidence relation between power electronic devices is complex, any local small degradation can be transmitted and diffused through communication paths between subsystems, so that the degradation process can be suddenly changed in a short time, and the function of a high-voltage direct-current breaker system is failed.

Currently, monitoring research on the high-voltage direct-current circuit breaker mostly focuses on failure mechanisms, fault characteristic parameter analysis, electronic component fault diagnosis and positioning and the like of power electronic devices or modules, and related research on identification or auxiliary identification of tiny degradation of the power electronic devices and switching devices is lacked.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention provides a method for deriving the running state of a high-voltage direct-current circuit breaker, which comprises the following steps:

collecting characteristic values of characteristic parameters in a characteristic subset of the high-voltage direct-current circuit breaker;

respectively calculating a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameters of the high-voltage direct-current circuit breaker based on the characteristic values;

deducing the running state of the high-voltage direct-current circuit breaker by utilizing the correlation coefficient matrix and the self-increasing and self-decreasing trend matrix;

the characteristic subset of the high-voltage direct-current circuit breaker is obtained by screening characteristic parameters in an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion weighting sorting algorithm and a correlation algorithm.

Preferably, the initial feature quantity set is composed of multi-dimensional features, and each dimensional feature is composed of a plurality of feature quantities.

Further, the process of screening the initial characteristic quantity set of the high-voltage direct-current circuit breaker based on the maximum correlation principle, the minimum redundancy principle, the multi-criterion empowerment sorting algorithm and the association algorithm includes:

based on a maximum correlation principle, a minimum redundancy principle and a multi-criterion empowerment sorting algorithm, screening the characteristic parameters in the initial characteristic quantity set to obtain a set Q;

and (4) removing the characteristic parameters with the support degree and the confidence degree smaller than the threshold requirement in the set Q by using a correlation algorithm to obtain the characteristic subset.

Further, the screening the feature parameters in the initial feature parameter set based on a maximum correlation principle, a minimum redundancy principle, and a multi-criterion weighted sorting algorithm to obtain a set Q includes:

step 1): making the set Q as an empty set;

step 2): based on the maximum correlation principle, selecting a characteristic parameter with the maximum mean weight of the category separability measurement from the initial characteristic parameter set, and putting the characteristic parameter into a set Q;

step 3): when the characteristic parameters in the initial characteristic quantity set are completely redundant with the characteristic parameters in the set Q, deleting the characteristic parameters redundant with the characteristic parameters in the initial characteristic quantity set and the set Q based on the minimum redundancy principle;

step 4): calculating the redundancy between each characteristic parameter in the initial characteristic quantity set and the set Q;

step 5): calculating the state representation degree of each characteristic parameter in the initial characteristic quantity set by using the redundancy between each characteristic parameter in the initial characteristic quantity set and the set Q, and selecting one characteristic parameter with the highest state representation degree from the initial characteristic quantity set to be put into the set Q based on a multi-criterion empowerment sorting algorithm;

step 6: if the characteristic parameters exist in the initial characteristic quantity set, returning to the step 3); otherwise, the set Q is output.

Further, the calculation formula of the mean value weight of the class separability measure of the feature parameters in the initial feature quantity set is as follows:

in the formula (I), the compound is shown in the specification,is the mean weight, avg (f) of the class separability measure of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set_ik) The historical sampling values of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set are allValue avg_i(F_i) Is the average value, x, of the historical sampling values of all the characteristic parameters under the ith dimension characteristic in the initial characteristic quantity set_f(f_ik) Is the f-th historical sampling value of the k-th characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set, the total number of historical sampling values of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set is i belongs to (1-N), N is the total number of dimensions in the initial characteristic quantity set, and k belongs to (1-N)_i)，N_iIs the total number of the feature parameters included in the ith dimension feature in the initial feature quantity set.

Further, the identification process that the characteristic parameters in the initial characteristic quantity set and the characteristic parameters in the set Q have redundancy includes:

if the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), characteristic parameter g in set Q_zInformation entropy of (2) and feature parameter f in initial feature quantity set_jAnd the characteristic parameter g in the set Q_zThe joint entropy of the initial feature quantity set is equal, then the feature quantity f in the initial feature quantity set is equal_jAnd the characteristic parameter g in the set Q_zComplete redundancy, otherwise, the characteristic parameter f in the initial characteristic quantity set_jAnd the characteristic parameter g in the set Q_zThere is no redundancy;

wherein, the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), characteristic parameter g in set Q_zInformation entropy of (2) and feature parameter f in initial feature quantity set_jAnd the characteristic parameter g in the set Q_zIs based on the first characteristic parameter f in the initial characteristic quantity set_jAnd/or the characteristic quantity g in the set Q_zDetermined from historical sample values of g_z∈Q，f_jAnd e is U, and U is an initial characteristic quantity set.

Further, the redundancy between each feature parameter in the initial feature parameter set and the set Q is calculated as follows:

in the formula (I), the compound is shown in the specification,for the characteristic parameter f in the initial characteristic quantity set_jRedundancy with respect to set Q, I (f)_j,g_z) Characteristic parameters in an initial characteristic quantity set and characteristic parameters g in a set Q_zMutual information between, g_z∈Q，f_jAnd e is U, and U is an initial characteristic quantity set.

Wherein, the characteristic parameter g in the characteristic participation set Q in the initial characteristic quantity set_zMutual information I (f) between_j,g_z) Is calculated as follows:

I(f_j,g_z)＝H(f_j)+H(g_z)-H(f_j,g_z)

in the formula, H (f)_j) For the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), H (g)_z) For the characteristic quantity g in the set Q_zEntropy of (d), H (f)_j,g_z) For the characteristic parameter f in the initial characteristic quantity set_jAnd the characteristic parameter g in the set Q_zThe joint entropy of (a).

Further, the calculation formula of the state characterization degree of each feature parameter in the initial feature quantity set is as follows:

wherein J (f)_j) For the characteristic parameter f in the initial characteristic quantity set_jThe degree of the state characterization of (a),is the mean weight of the class separability measure of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set,for the characteristic parameter f in the initial characteristic quantity set_jThe redundancy with respect to the set Q,

preferably, the acquiring the characteristic value of each characteristic parameter in the characteristic subset of the high-voltage direct-current circuit breaker includes:

collecting the characteristic values of each characteristic parameter in the characteristic subset at the current moment and r-1 moments before the current moment;

wherein r is the total number of time instants contained in the sampling time window.

Further, the calculating a correlation coefficient matrix of the characteristic parameter of the high-voltage direct-current circuit breaker based on the characteristic value includes:

and determining a correlation coefficient matrix of the characteristic parameters of the high-voltage direct-current circuit breaker at the current moment by using a correlation coefficient calculation method based on the characteristic values.

Further, the calculating a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker based on the characteristic value comprises:

according to the characteristic values, characteristic value matrixes of the high-voltage direct-current circuit breaker at the current moment and h-1 moments before the current moment are respectively determined;

determining a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at h-1 moments before the current moment by using the characteristic value matrix;

estimating a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the current moment by adopting a least square method based on the self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at h-1 moments before the current moment;

the self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment is obtained by subtracting the characteristic value matrix of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment from the characteristic value matrix of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment;

the characteristic value matrixes of the high-voltage direct-current circuit breaker at the current moment and r-1 moments before the current moment are x row 1 column matrixes, alpha row elements of the matrix are characteristic values of alpha characteristic parameters in the characteristic subset at the current moment and r-1 moments before the current moment, x is the total number of the characteristic parameters in the characteristic subset, h is a positive integer smaller than r, w is a positive integer smaller than h-1, and h is the total number of moments contained in a derivation time window.

Further, the deriving the operating state of the high-voltage direct-current circuit breaker by using the correlation coefficient matrix and the auto-increment and decrement trend matrix includes:

taking the sum of the product of a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the current moment and the characteristic value matrix of the high-voltage direct-current circuit breaker at the current moment as a characteristic value matrix of the high-voltage direct-current circuit breaker at a moment after the current moment;

the characteristic value matrix of the high-voltage direct current breaker at a moment after the current moment represents the operation state of the high-voltage direct current breaker at a moment after the current moment.

Further, the feature dimensions of the initial feature quantity set include, but are not limited to: thermal faults, voltage withstand breakdown faults, electrical circuit faults, transient faults, secondary system faults, water cooling system faults and energy supply system faults;

the characteristic parameters under thermal failure include but are not limited to: IGBT junction temperature, mechanical switch contact temperature, submodule temperature, MOV overheating protection action, overheating alarm and protection action information;

the characteristic parameters under the fault of the electric circuit include but are not limited to: the method comprises the following steps that sub-module voltage, sub-module current, main branch current, transfer branch current, lightning arrester leakage current, main branch abnormal state information, transfer branch abnormal state information and energy consumption branch abnormal state information are obtained;

the characteristic parameters under the transient fault include, but are not limited to: system voltage, system current, breaking time, dissipation energy, abnormal state of the rapid mechanical switch and conversion overtime;

the characteristic parameters under the secondary system fault include but are not limited to: communication abnormality information, protection action information, and control abnormality information;

the characteristic parameters under the fault of the water cooling system include but are not limited to: the water cooling system fault state and water leakage detection device gives an alarm;

the characteristic parameters under the fault of the energy supply system include but are not limited to: and energy supply system fault information.

Based on the same inventive concept, the invention also provides a high-voltage direct current breaker operation state derivation system, which comprises:

the acquisition module is used for acquiring the characteristic values of all characteristic parameters in the characteristic subset of the high-voltage direct-current circuit breaker;

the calculation module is used for respectively calculating a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameters of the high-voltage direct-current circuit breaker based on the characteristic values;

the derivation module is used for deriving the running state of the high-voltage direct-current circuit breaker by utilizing the correlation coefficient matrix and the self-increasing and self-decreasing trend matrix;

the characteristic subset of the high-voltage direct-current circuit breaker is obtained by screening characteristic parameters in an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion weighting sorting algorithm and a correlation algorithm.

Compared with the closest prior art, the invention has the following beneficial effects:

the invention provides a method and a system for deriving the running state of a high-voltage direct-current circuit breaker, which comprise the following steps: collecting characteristic values of characteristic parameters in a characteristic subset of the high-voltage direct-current circuit breaker; respectively calculating a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameters of the high-voltage direct-current circuit breaker based on the characteristic values; deducing the running state of the high-voltage direct-current circuit breaker by utilizing the correlation coefficient matrix and the self-increasing and self-decreasing trend matrix; the characteristic subset of the high-voltage direct-current circuit breaker is obtained by screening characteristic parameters in an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion weighting sorting algorithm and a correlation algorithm. On the basis of the high-voltage direct-current breaker characteristic parameter subset, the correlation and the change trend of the high-voltage direct-current breaker characteristic parameters are analyzed on line by means of statistics, linear regression algorithm and the like, and then the on-line derivation of the running state of the high-voltage direct-current breaker is realized; the method provides data support for operators to know the operation state of the high-voltage direct-current circuit breaker in time, catch early defects of the high-voltage direct-current circuit breaker and scientifically make a state maintenance plan, and provides powerful technical support for improving the overall operation efficiency and reliability of the high-voltage direct-current circuit breaker.

According to the method, the similarity and the incidence relation among the characteristic parameters of each module in the high-voltage direct-current circuit breaker are utilized to screen the characteristic parameters of the existing high-voltage direct-current circuit breaker, and the extracted characteristic parameter subset of the high-voltage direct-current circuit breaker can reflect various fault conditions of the high-voltage direct-current circuit breaker comprehensively.

The invention provides a data prediction means for the prediction of the running state of the high-voltage direct-current circuit breaker and the evaluation of the fault risk, realizes the fine description of the running characteristic parameters of the high-voltage direct-current circuit breaker by adjusting the sampling time window and the derivation time window, and provides data support for the identification of the micro fault of the high-voltage direct-current circuit breaker and the evaluation of the associated state of the micro fault.

Drawings

Fig. 1 is a flowchart of a method for deriving an operating state of a high-voltage dc circuit breaker according to the present invention;

FIG. 2 is a diagram of a general concept of a state derivation method according to an embodiment of the present invention;

FIG. 3 is a flow chart of feature subset extraction in an embodiment of the present invention;

FIG. 4 is a flow chart illustrating a state derivation implementation according to an embodiment of the present invention;

fig. 5 is a structural diagram of an operation state derivation system of a high-voltage direct-current circuit breaker according to the present invention.

Detailed Description

The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The invention provides a method for deriving the running state of a high-voltage direct-current circuit breaker, which comprises the following steps of:

step 1, collecting characteristic values of characteristic parameters in a characteristic subset of the high-voltage direct-current circuit breaker;

step 2, respectively calculating a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameters of the high-voltage direct-current circuit breaker based on the characteristic values;

step 3, deriving the running state of the high-voltage direct-current circuit breaker by utilizing the correlation coefficient matrix and the self-increasing and self-decreasing trend matrix;

the characteristic subset of the high-voltage direct-current circuit breaker is obtained by screening characteristic parameters in an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion weighting sorting algorithm and a correlation algorithm.

In this embodiment, before performing the derivation of the operating state of the high-voltage direct-current circuit breaker, the feature subset of the high-voltage direct-current circuit breaker needs to be extracted in advance, and after the feature subset is extracted, the derivation of the operating state of the high-voltage direct-current circuit breaker provided by the present invention is performed according to the general idea diagram of the state derivation method shown in fig. 2;

the method comprises the steps that a characteristic subset of the high-voltage direct-current circuit breaker is extracted in advance, and the characteristic subset can be called as a data preparation stage or an off-line analysis stage, and is extracted by screening characteristic parameters of the high-voltage direct-current circuit breaker by utilizing original accumulated data of the high-voltage direct-current circuit breaker;

the characteristic subset extraction flow chart is shown in fig. 3, and considering that the current high-voltage direct-current circuit breaker is small in operation quantity and lack of historical data accumulation samples, characteristic parameters in the high-voltage direct-current circuit breaker characteristic subset can be selected by referring to two parts of data, namely historical operation database records and high-voltage direct-current circuit breaker preventive test data;

the corresponding steps are as follows:

s1: primarily screening the characteristic parameters of the high-voltage direct-current circuit breaker, and recording the initial characteristic quantity set after screening as U;

the screening principle of the characteristic parameters of the high-voltage direct-current circuit breaker is as follows:

1) characteristic parameters which can be acquired on line and can reflect the current operating situation and faults of the high-voltage direct-current circuit breaker;

2) the statistical data which are stored in a historical database and can reflect the historical operation rule of the high-voltage direct-current circuit breaker;

3) recording preventive test data which can reflect the failure condition of the high-voltage direct-current circuit breaker in a historical database;

4) external data capable of influencing the development trend of the running state of the high-voltage direct-current circuit breaker;

after screening, the initial characteristic quantity set U of the high-voltage direct-current circuit breaker comprises n-dimensional characteristics, namely U ═ F₁…F_i…F_n) The ith dimension feature in the initial feature quantity set U comprises N_iA characteristic parameter, i.e.

Wherein f is_ikIs the k characteristic parameter, N, in the i-dimension characteristic in the initial characteristic quantity set U_iThe total number of the characteristic parameters contained in the ith dimension characteristic in the initial characteristic quantity set U is obtained;

the dimensions of the features in the initial feature quantity set U include, but are not limited to:

thermal faults, voltage withstand breakdown faults, electrical circuit faults, transient faults, secondary system faults, water cooling system faults and energy supply system faults;

the feature parameters included in the features of each dimension are shown in table 1:

TABLE 1

S2: further screening the characteristic parameters in the initial characteristic quantity set U based on a maximum correlation principle, a minimum redundancy principle and a multi-criterion empowerment sorting algorithm to obtain an initial characteristic subset Q of the high-voltage direct-current circuit breaker, and the method comprises the following steps:

s2-1, initializing an initial characteristic subset Q as an empty set;

s2-2, based on the maximum correlation principle, selecting a characteristic parameter with the maximum mean weight of the category separability measurement from the initial characteristic quantity set U, and putting the characteristic parameter into the initial characteristic subset Q as a first characteristic parameter of the initial characteristic subset Q;

is formulated as follows:

in the formula, g₁The characteristic parameter with the largest mean weight of the category separability measures in the initial characteristic quantity set U is also the first characteristic parameter in the initial characteristic subset Q,the mean weight of the category separability measurement of the kth feature parameter in the ith-dimension feature in the initial feature quantity set U is represented by a symbol arg max f (x), wherein the symbol arg max f (x) represents the value of x when f (x) takes the maximum value;

wherein the content of the first and second substances,is calculated as follows:

in the formula, avg (f)_ik) Is the average value, avg, of the history sampling values of the kth characteristic parameter in the ith dimension characteristic in the initial characteristic quantity set U_i(F_i) Is the average value, x, of the historical sample values of all the characteristic parameters in the ith dimension characteristic in the initial characteristic quantity set U_f(f_ik) Is the f-th historical sampling value of the k-th characteristic parameter in the ith dimension characteristic in the initial characteristic quantity set U, the total number of the history sampling values of the kth characteristic parameter in the ith dimension characteristic in the initial characteristic quantity set U.

S2-3, based on the principle of minimum redundancy, for each feature parameter in the initial feature parameter set U, if one feature parameter in the initial feature subset Q meets the condition that the information entropy of the feature parameter, the information entropy of the feature parameter and the joint entropy of the feature parameter and the information entropy are equal, rejecting the feature parameter in the initial feature parameter set U, otherwise, calculating the mutual information between the feature parameter and each feature parameter in the initial feature subset Q, and taking the maximum value as the redundancy between the feature parameter and the initial feature subset Q;

suppose that: for the characteristic parameter f in the initial characteristic quantity set U_jIf a characteristic quantity g exists in the initial characteristic subset Q_zSatisfies the condition of H (f)_j)＝H(g_z)＝H(f_j,g_z) Then explain f_jAnd g_zComplete redundancy, then f in the initial feature quantity set U_j(ii) a Otherwise, f is calculated_jMutual information with the characteristic parameters of the initial characteristic subset Q, and taking the maximum value as f_jRedundancy with the initial feature subset Q, denoted I_max(f_j,g_z),g_z∈Q；

Wherein, H (f)_j)＝-p(f_j)lgp(f_j)，H(g_z)＝-p(g_z)lgp(g_z)，H(f_j,g_z)＝-p(f_j,g_z)lgp(f_j,g_z)，I(f_j,g_z)＝H(f_j)+H(g_z)-H(f_j,g_z)；

In the formula, H (f)_j) Is a characteristic parameter f_jEntropy of information of (1), p (f)_j) Is a characteristic parameter f_jOut-of-limit probability/occurrence probability of, H (g)_z) Is a characteristic parameter g_zEntropy of information of (1), p (g)_z) Is a characteristic parameter g_zIs out of limit probabilityProbability of occurrence, H (f)_j,g_z) Is a characteristic parameter f_jAnd a characteristic parameter g_zJoint entropy of p (f)_j,g_z) Is a characteristic parameter f_jOut-of-limit/occurrence and characteristic parameter g_zProbability of occurrence of out-of-limit/occurrence colleagues; the out-of-limit probability is for the analog quantity and the occurrence probability is for the state quantity.

The p (f)_j) Based on the characteristic parameter f of the acquisition_jIs determined by the characteristic value of p (g)_z) Is based on the acquired characteristic parameters g_zIs determined by the characteristic value of p (f)_j,g_z) Characteristic parameter f based on acquisition_jCharacteristic value of (g) and acquired characteristic parameter g_zIs determined.

S2-4, based on a multi-criterion weighted sorting algorithm, selecting a feature parameter with the highest state representation degree from the initial feature quantity set U, putting the feature parameter into the initial feature subset Q, and returning to the step S2-3 until no feature parameter exists in the initial feature quantity set U;

is formulated as follows:

in the formula, g_lA characteristic parameter with the highest state characterization degree in the initial characteristic quantity set U is also the characteristic parameter currently put into the initial characteristic subset Q, J (f)_j) Is the characteristic parameter f in the initial characteristic quantity set U_jDegree of state characterization of;

wherein, the J (f)_j) The expression of (a) is:

in the formula I_max(f_j,g_z) Is f_jThe redundancy with respect to the initial feature subset Q,is the characteristic parameter f in the initial characteristic quantity set U_jThe mean weight of the class separability measure of (1).

S3: and verifying the relevance of the characteristic parameters in the initial characteristic subset Q through a correlation algorithm, and eliminating the characteristic parameters with the support degree and the confidence degree smaller than the threshold value requirement to obtain a characteristic subset S of the high-voltage direct-current circuit breaker.

The state derivation process of the invention can be called as an online analysis part, and the online state derivation of the high-voltage direct-current circuit breaker is realized by calculating the incidence matrix and the self-increasing and self-decreasing trend by utilizing the initial state determined by the characteristic subset.

Wherein, the state derivation implementation flowchart is shown in fig. 4, and the step 1 includes:

data preparation, namely determining a derivation time window h for state derivation of the high-voltage direct-current circuit breaker and a sampling time window r of each characteristic parameter in the characteristic subset S, and carrying out online acquisition on each characteristic parameter in the characteristic subset S;

setting the current time as t, and collecting the characteristic values of all characteristic parameters in the characteristic subset S from the t-r +1 time to the current time t on line.

The step 2 includes:

step 2-1: calculating a correlation coefficient matrix P (t) of each characteristic parameter in the characteristic subset S at the current moment t, namely calculating a cross correlation coefficient between an autocorrelation coefficient and the characteristic parameter of each characteristic parameter by using characteristic values of each characteristic parameter in the characteristic subset S from the moment t-r +1 acquired on line to the current moment t, and constructing a correlation coefficient matrix P (t) of each characteristic parameter in the characteristic subset S at the current moment t;

wherein the expression of P (t) is as follows:

wherein χ is the total number of parameters in the feature subset S, and p (t) is the α -row and β -column element p_αβ(t) is a correlation coefficient between the alpha-th characteristic parameter and the beta-th characteristic parameter in the characteristic subset S at the current moment t;

said p is_αβThe calculation formula of (t) is as follows:

in the formula, Cov (x)_α(t),x_β(t)) is the covariance of the a-th and ss-th characteristic in the subset S of characteristics at the current time t, Var [ X (t))]The variance of a characteristic parameter matrix X (t) of the high-voltage direct-current circuit breaker at the current moment t is shown, wherein X (t) is x₁(t)…x_α(t)…x_χ(t)]^T，x_α(t) is the characteristic value of the alpha-th characteristic parameter in the characteristic subset S at the current moment t;

the Cov (x)_α(t),x_β(t)) and Var [ X (t))]The method is obtained by respectively utilizing a covariance calculation formula and a variance calculation formula to calculate the characteristic values of all characteristic parameters in a characteristic subset S from a t-r +1 moment to a current moment t, which are acquired on line.

Step 2-2: estimating a self-increasing and self-decreasing trend matrix delta (t) of characteristic parameters in a characteristic subset S of the current moment t by utilizing a least square regression algorithm based on a characteristic parameter matrix of the high-voltage direct-current circuit breaker from the moment t-r +1 to the current moment t;

wherein the estimated value of delta (t) is usedExpressed, obtained by unary linear regression analysis. Extending Δ (t) forward to Δ (t-1), Δ (t-2), … …, Δ (t-h), where Δ (t- σ) ═ X (t- σ +1) -X (t- σ), σ ═ 1,2, …, h;

using least squares regression algorithm to saidThe formula for estimating is as follows:

solving the formula to obtain the characteristic parameters in the characteristic subset SThe minimum tendency of self-increment and self-decrement of the,equal to the sum of the minimum auto-decreasing trend and Δ (t-1).

The step 3 comprises the following steps:

state derivation, i.e. using estimated values of the auto-decreasing trend matrix Delta (t) of the characteristic quantities in the characteristic subset S at the current time tAnd calculating an increase and decrease quantity matrix of the characteristic parameters of the high-voltage direct-current circuit breaker at the current moment t by using the correlation coefficient matrix P (t) of each characteristic parameter in the characteristic subset S at the current moment t, and deriving the characteristic values of each characteristic parameter in the characteristic subset S at the moment t +1 by using the increase and decrease quantity matrix.

Namely: epsilon (t) ═ P (t) delta (t), X (t +1) ═ X (t) + epsilon (t), wherein epsilon (t) is an increase and decrease matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the current moment t.

It should be noted that in the process of deriving the state of the high-voltage direct-current circuit breaker, a derivation time window h and a sampling time window r in the step S need to be considered, and the calculation amount is increased and unnecessary calculation cost is paid due to the fact that the derivation time window h and the sampling time window r are too large; the undersize of the derivation time window h and the sampling time window r causes the reduction of the information granularity, so that the estimation of the self-increasing and self-decreasing trend is inaccurate.

In the specific embodiment of the invention, the method provided by the invention utilizes abundant electrical information of the high-voltage direct-current circuit breaker to predict and epitaxially derive the operation characteristic parameters of the high-voltage direct-current circuit breaker in a short time scale, and provides auxiliary analysis information for evaluating the degradation process of a power electronic device, so that operation and maintenance personnel are assisted to quickly find and capture early defects of the high-voltage direct-current circuit breaker, a data basis is provided for online identification, positioning and diagnosis of tiny faults of the high-voltage direct-current circuit breaker, and the method has important significance for ensuring safe and reliable operation of high-voltage direct-current circuit breaker equipment.

In a specific embodiment of the present invention, the derivation of the operating state of the high-voltage dc circuit breaker is implemented on the premise that:

1. the high-voltage direct-current circuit breaker is stable in operation state and in a non-fault state, all electrical parameters are not subjected to external impact to generate jumping change, and the acquired data are continuous;

2. the current characteristic parameter value of the high-voltage direct-current circuit breaker is influenced by the change of the past characteristic parameter value and the peripheral characteristic parameter value of the high-voltage direct-current circuit breaker, and the characteristic parameter values have relevance.

According to the invention, the fine state change process of the high-voltage direct-current circuit breaker in a short time scale is described through data derivation, so that data support is provided for operating personnel to know the operating state of the high-voltage direct-current circuit breaker in time, catch early defects of the high-voltage direct-current circuit breaker and scientifically make a state maintenance plan, and powerful technical support is provided for improving the overall operating efficiency and reliability of the high-voltage direct-current circuit breaker.

The method is based on the information acquired by the existing high-voltage direct-current circuit breaker, an additional measuring device is not needed, the self change trend of each characteristic parameter in the operation process of the high-voltage direct-current circuit breaker is considered on the basis of representing the operation loss accumulation of components, and the mutual influence relation among the characteristic parameters is considered from the viewpoint of system structure, so that the derivation of the operation state of the direct-current circuit breaker can be comprehensively realized.

Example 2:

the present invention also provides a system for deriving an operating state of a high voltage dc circuit breaker, as shown in fig. 5, including:

the acquisition module is used for acquiring the characteristic values of all characteristic parameters in the characteristic subset of the high-voltage direct-current circuit breaker;

the calculation module is used for respectively calculating a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameters of the high-voltage direct-current circuit breaker based on the characteristic values;

the derivation module is used for deriving the running state of the high-voltage direct-current circuit breaker by utilizing the correlation coefficient matrix and the self-increasing and self-decreasing trend matrix;

the characteristic subset of the high-voltage direct-current circuit breaker is obtained by screening characteristic parameters in an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion weighting sorting algorithm and a correlation algorithm.

Specifically, the initial feature set is composed of multidimensional features, and each dimensional feature is composed of a plurality of feature parameters.

Specifically, the system further includes a screening module for screening an initial characteristic quantity set of the high-voltage direct-current circuit breaker based on a maximum correlation principle, a minimum redundancy principle, a multi-criterion empowerment sorting algorithm and an association algorithm, where the screening module includes:

the screening unit is used for screening the characteristic parameters in the initial characteristic quantity set to obtain a set Q based on a maximum correlation principle, a minimum redundancy principle and a multi-criterion empowerment sorting algorithm;

and the eliminating unit is used for eliminating the characteristic parameters of which the support degree and the confidence degree are less than the threshold value requirements in the set Q by utilizing a correlation algorithm to obtain the characteristic subset.

Further, the screening unit includes:

setting a submodule for making the set Q an empty set;

the first selection subunit is used for selecting one characteristic parameter with the largest mean weight of the category separability metrics from the initial characteristic parameter set based on the maximum correlation principle and placing the characteristic parameter into a set Q;

the deleting subunit is used for deleting the characteristic parameters redundant in the initial characteristic quantity set and the characteristic parameters redundant in the set Q based on the minimum redundancy principle when the characteristic parameters in the initial characteristic quantity set and the characteristic parameters in the set Q are completely redundant;

the calculating subunit is used for calculating the redundancy between each characteristic parameter in the initial characteristic quantity set and the set Q;

the second selection subunit is used for calculating the state representation degree of each characteristic parameter in the initial characteristic quantity set by using the redundancy between each characteristic parameter in the initial characteristic quantity set and the set Q, and selecting one characteristic parameter with the highest state representation degree from the initial characteristic quantity set to be placed in the set Q based on a multi-criterion empowerment sorting algorithm;

the output subunit is used for returning to the deletion subunit if the characteristic parameters exist in the initial characteristic quantity set; otherwise, the set Q is output.

Specifically, the calculation formula of the mean value weight of the category separability measure of the feature parameters in the initial feature quantity set is as follows:

in the formula (I), the compound is shown in the specification,is the mean weight, avg (f) of the class separability measure of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set_ik) Is the mean value, avg, of the historical sampling value of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set_i(F_i) Is the average value, x, of the historical sampling values of all the characteristic parameters under the ith dimension characteristic in the initial characteristic quantity set_f(f_ik) Is the f-th historical sampling value of the k-th characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set,the total number of historical sampling values of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set is i belongs to (1-N), N is the total number of dimensions in the initial characteristic quantity set, and k belongs to (1-N)_i)，N_iIs the total number of the feature parameters included in the ith dimension feature in the initial feature quantity set.

Specifically, the identification process that the characteristic parameters in the initial characteristic quantity set and the characteristic parameters in the set Q have redundancy includes:

if the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), characteristic parameter g in set Q_zInformation entropy of (2) and feature parameter f in initial feature quantity set_jAnd the characteristic parameter g in the set Q_zThe joint entropy of the initial feature quantity set is equal, then the feature quantity f in the initial feature quantity set is equal_jAnd the characteristic parameter g in the set Q_zComplete redundancy, otherwise, the characteristic parameter f in the initial characteristic quantity set_jAnd the characteristic parameter g in the set Q_zThere is no redundancy;

wherein, the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), characteristic parameter g in set Q_zInformation entropy of (2) and feature parameter f in initial feature quantity set_jAnd the characteristic parameter g in the set Q_zIs based on the first characteristic parameter f in the initial characteristic quantity set_jAnd/or the characteristic quantity g in the set Q_zDetermined from historical sample values of g_z∈Q，f_jAnd e is U, and U is an initial characteristic quantity set.

Specifically, the redundancy between each feature parameter in the initial feature parameter set and the set Q is calculated as follows:

in the formula (I), the compound is shown in the specification,for the characteristic parameter f in the initial characteristic quantity set_jRedundancy with respect to set Q, I (f)_j,g_z) Characteristic parameters in an initial characteristic quantity set and characteristic parameters g in a set Q_zMutual information between, g_z∈Q，f_jAnd e is U, and U is an initial characteristic quantity set.

Wherein, the characteristic parameter g in the characteristic participation set Q in the initial characteristic quantity set_zMutual information I (f) between_j,g_z) Is calculated as follows:

I(f_j,g_z)＝H(f_j)+H(g_z)-H(f_j,g_z)

in the formula, H (f)_j) For the characteristic parameter f in the initial characteristic quantity set_jEntropy of information of (1), H (g)_z) For the characteristic quantity g in the set Q_zEntropy of (d), H (f)_j,g_z) For the characteristic parameter f in the initial characteristic quantity set_jAnd the characteristic parameter g in the set Q_zThe joint entropy of (a).

Specifically, the calculation formula of the state representation degree of each feature parameter in the initial feature quantity set is as follows:

wherein J (f)_j) For the characteristic parameter f in the initial characteristic quantity set_jThe degree of the state characterization of (a),is the mean weight of the class separability measure of the kth characteristic parameter under the ith dimension characteristic in the initial characteristic quantity set,for the characteristic parameter f in the initial characteristic quantity set_jThe redundancy with respect to the set Q,

specifically, the acquisition module is configured to:

collecting the characteristic values of each characteristic parameter in the characteristic subset at the current moment and r-1 moments before the current moment;

wherein r is the total number of time instants contained in the sampling time window.

Specifically, the calculation unit includes:

and the first determining submodule is used for determining a correlation coefficient matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the current moment by using a correlation coefficient calculation method based on the characteristic value.

Specifically, the computing unit further includes:

the second determining submodule is used for respectively determining the eigenvalue matrixes of the high-voltage direct-current circuit breaker at the current moment and h-1 moments before the current moment according to the eigenvalues;

the third determining submodule is used for determining a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at h-1 moments before the current moment by using the characteristic value matrix;

the estimation submodule is used for estimating a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the current moment by adopting a least square method based on the self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at h-1 moments before the current moment;

the self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment is obtained by subtracting the characteristic value matrix of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment from the characteristic value matrix of the high-voltage direct-current circuit breaker at the w-1 th moment before the current moment;

the characteristic value matrixes of the high-voltage direct-current circuit breaker at the current moment and r-1 moments before the current moment are x row 1 column matrixes, alpha row elements of the matrix are characteristic values of alpha characteristic parameters in the characteristic subset at the current moment and r-1 moments before the current moment, x is the total number of the characteristic parameters in the characteristic subset, h is a positive integer smaller than r, w is a positive integer smaller than h-1, and h is the total number of moments contained in a derivation time window.

Specifically, the derivation unit is configured to:

taking the sum of the product of a correlation coefficient matrix and a self-increasing and self-decreasing trend matrix of the characteristic parameter of the high-voltage direct-current circuit breaker at the current moment and the characteristic value matrix of the high-voltage direct-current circuit breaker at the current moment as a characteristic value matrix of the high-voltage direct-current circuit breaker at a moment after the current moment;

the characteristic value matrix of the high-voltage direct current breaker at a moment after the current moment represents the operation state of the high-voltage direct current breaker at a moment after the current moment.

Specifically, the feature dimensions of the initial feature quantity set include, but are not limited to: thermal faults, voltage withstand breakdown faults, electrical circuit faults, transient faults, secondary system faults, water cooling system faults and energy supply system faults;

the characteristic parameters under thermal failure include but are not limited to: IGBT junction temperature, mechanical switch contact temperature, submodule temperature, MOV overheating protection action, overheating alarm and protection action information;

the characteristic parameters under the fault of the electric circuit include but are not limited to: the method comprises the following steps that sub-module voltage, sub-module current, main branch current, transfer branch current, lightning arrester leakage current, main branch abnormal state information, transfer branch abnormal state information and energy consumption branch abnormal state information are obtained;

the characteristic parameters under the transient fault include, but are not limited to: system voltage, system current, breaking time, dissipation energy, abnormal state of the rapid mechanical switch and conversion overtime;

the characteristic parameters under the secondary system fault include but are not limited to: communication abnormality information, protection action information, and control abnormality information;

the characteristic parameters under the fault of the water cooling system include but are not limited to: the water cooling system fault state and water leakage detection device gives an alarm;

the characteristic parameters under the fault of the energy supply system include but are not limited to: and energy supply system fault information.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.