阅读说明：本技术 一种汽轮发电机组的故障确定方法及系统 (Fault determination method and system for steam turbine generator unit ) 是由 于信波 冷述文 赵峰 孙明 薛松 刘士方 张华东 张敬 刘磊 邵帅 于 2021-08-23 设计创作，主要内容包括：本发明提供了一种汽轮发电机组的故障确定方法及系统,方法包括：获取汽轮发电机组发生故障时的故障振动数据；故障振动数据包括多种振动因素数据；根据故障振动数据,确定汽轮发电机组的故障类型；获取汽轮发电机组中能够引起故障类型的多种状态参数；根据状态参数和故障振动数据,构建故障类型参数矩阵；根据故障类型参数矩阵,确定汽轮发电机组的故障原因和故障位置。本发明通过建立故障类型参数矩阵,能够确定故障的原因和位置,以便工作人员能够及时确定故障修复方案。(The invention provides a method and a system for determining faults of a steam turbine generator unit, wherein the method comprises the following steps: acquiring fault vibration data when the steam turbine generator unit is in fault; the fault vibration data comprises a plurality of vibration factor data; determining the fault type of the steam turbine generator unit according to the fault vibration data; acquiring various state parameters which can cause fault types in a steam turbine generator unit; constructing a fault type parameter matrix according to the state parameters and the fault vibration data; and determining the fault reason and the fault position of the steam turbine generator unit according to the fault type parameter matrix. According to the invention, the reason and the position of the fault can be determined by establishing the fault type parameter matrix, so that a worker can determine the fault repairing scheme in time.)

1. A method for determining faults of a steam turbine generator unit is characterized by comprising the following steps:

acquiring fault vibration data when the steam turbine generator unit is in fault; the fault vibration data comprises a plurality of vibration factor data;

determining the fault type of the steam turbine generator unit according to the fault vibration data;

acquiring various state parameters which can cause the fault type in the steam turbine generator unit;

constructing a fault type parameter matrix according to the state parameters and the fault vibration data;

and determining the fault reason and the fault position of the steam turbine generator unit according to the fault type parameter matrix.

2. The method for determining the fault of the steam turbine generator unit according to claim 1, wherein the obtaining of the fault vibration data when the steam turbine generator unit is in fault specifically includes:

acquiring current vibration data of a steam turbine generator unit;

judging whether the difference value between the current vibration data and the vibration reference value is smaller than a difference value threshold value or not to obtain a first judgment result;

and if the first judgment result is negative, determining that the current vibration data are fault vibration data.

3. The method for determining the fault of the steam turbine generator unit according to claim 1, wherein the determining the fault type of the steam turbine generator unit according to the fault vibration data specifically comprises:

determining any fault type in the fault type set as a current fault type; the fault type set comprises a plurality of fault types which can be caused by vibration of the steam turbine generator unit;

acquiring a plurality of rules of the current fault type and a plurality of symptom credibility corresponding to each rule respectively;

determining the product of the minimum value of the plurality of symptom credibility corresponding to each rule and the rule credibility coefficient as the rule credibility corresponding to each rule;

determining the maximum value of the rule credibility as the fault credibility of the current fault type;

traversing all fault types to obtain a plurality of fault credibility;

and determining the fault type corresponding to the maximum value in the multiple regular fault credibility as the fault type of the steam turbine generator unit.

4. The method for determining the fault of the steam turbine generator unit according to claim 1, wherein the fault type parameter matrix is:

wherein, X'₀Is a vibration factor matrix, x'₀(1)、x'₀(2) And x'₀(N) are respectively vibration factors of 1 st, 2 nd and N th, X'_iA parameter matrix of the ith vibration factor; x'_i＝[x′_i(1),x′_i(2),...，x′_i(N)]^T，i＝1,2,...,n，x′_i(k) A kth state parameter, k ═ 1,2,. and N, which affects the ith vibration factor; n is the type quantity of the state parameters influencing the ith vibration factor, and N is the type number of the vibration factors in the fault vibration data.

5. The method for determining the fault of the steam turbine generator unit according to claim 4, wherein the determining of the fault cause and the fault position of the steam turbine generator unit according to the fault type parameter matrix specifically comprises:

carrying out non-dimensionalization on the fault type parameter matrix to obtain a non-dimensionalized fault type parameter matrix;

determining the absolute value of the difference value between each element in the first column of elements in the non-dimensionalized fault type parameter matrix and each element in the row where the element is located, so as to obtain an absolute difference value matrix;

using formulasProcessing the absolute difference matrix to obtain a correlation coefficient matrix;

calculating an average value of each row of elements in the correlation coefficient matrix, and determining a state parameter corresponding to a maximum value in the average values as a fault state parameter;

determining the fault reason and the fault position of the steam turbine generator unit according to the fault state parameters;

wherein ξ_0i(k) The correlation coefficient of the kth state parameter of the ith vibration factor is Delta (min) is the minimum value of all elements in the absolute difference matrix, rho is the resolution coefficient, Delta (max) is the maximum value of all elements in the absolute difference matrix, and Delta (min) is the maximum value of all elements in the absolute difference matrix_0i(k) Is the element of the kth row of the ith column in the absolute difference matrix.

6. A fault determination system for a steam turbine generator unit, the system comprising:

the fault vibration data acquisition module is used for acquiring fault vibration data when the steam turbine generator unit fails; the fault vibration data comprises a plurality of vibration factor data;

the fault type determining module is used for determining the fault type of the steam turbine generator unit according to the fault vibration data;

the state parameter acquisition module is used for acquiring various state parameters which can cause the fault type in the steam turbine generator unit;

the fault type parameter matrix construction module is used for constructing a fault type parameter matrix according to the state parameters and the fault vibration data;

and the fault diagnosis module is used for determining the fault reason and the fault position of the steam turbine generator unit according to the fault type parameter matrix.

7. The system for determining a fault of a steam turbine generator unit according to claim 5, wherein the fault vibration data acquisition module specifically includes:

the current vibration data acquisition unit is used for acquiring current vibration data of the steam turbine generator unit;

the judging unit is used for judging whether the difference value between the current vibration data and the vibration reference value is smaller than a difference value threshold value or not to obtain a first judging result; if the first judgment result is negative, calling a fault vibration data determination unit;

and the fault vibration data determining unit is used for determining that the current vibration data are fault vibration data.

8. The system for determining the fault of the steam turbine generator unit according to claim 5, wherein the fault type determination module specifically includes:

the current fault type determining unit is used for determining any fault type in the fault type set as a current fault type; the fault type set comprises a plurality of fault types which can be caused by vibration of the steam turbine generator unit;

the rule obtaining unit is used for obtaining a plurality of rules of the current fault type and a plurality of symptom credibility corresponding to each rule;

the rule credibility determining unit is used for determining the product of the minimum value of the plurality of symptom credibility corresponding to each rule and the rule credibility coefficient as the rule credibility corresponding to each rule;

the fault reliability determining unit is used for determining that the maximum value in the rule reliability is the fault reliability of the current fault type;

the fault type traversing unit is used for traversing all fault types to obtain a plurality of fault credibility;

and the fault type judging unit is used for determining the fault type corresponding to the maximum value in the fault credibility of the plurality of rules as the fault type of the steam turbine generator unit.

9. The system of claim 5, wherein the fault type parameter matrix is:

wherein, X'₀Is a vibration factor matrix, x'₀(1)、x'₀(2) And x'₀(N) are respectively vibration factors of 1 st, 2 nd and N th, X'_iA parameter matrix of the ith vibration factor; x'_i＝[x′_i(1),x′_i(2),...，x′_i(N)]^T，i＝1,2,...,n，x′_i(k) A kth state parameter, k ═ 1,2,. and N, which affects the ith vibration factor; n is the type quantity of the state parameters influencing the ith vibration factor, and N is the type number of the vibration factors in the fault vibration data.

10. The system for determining a fault of a steam turbine generator unit according to claim 9, wherein the fault diagnosis module specifically includes:

the non-dimensionalization unit is used for carrying out non-dimensionalization processing on the fault type parameter matrix to obtain a non-dimensionalized fault type parameter matrix;

an absolute difference matrix determining unit, configured to determine an absolute value of a difference between each element in the first column of elements in the dimensionless fault type parameter matrix and each element in the row of the dimensionless fault type parameter matrix, to obtain an absolute difference matrix;

a correlation coefficient matrix determination unit for using a formulaProcessing the absolute difference matrix to obtain a correlation coefficient matrix;

the fault state parameter determining unit is used for calculating the average value of each row of elements in the correlation coefficient matrix and determining the state parameter corresponding to the maximum value in the average values as the fault state parameter;

the fault diagnosis unit is used for determining the fault reason and the fault position of the steam turbine generator unit according to the fault state parameters;

wherein ξ_0i(k) The correlation coefficient of the kth state parameter of the ith vibration factor is Delta (min) is the minimum value of all elements in the absolute difference matrix, rho is the resolution coefficient, Delta (max) is the maximum value of all elements in the absolute difference matrix, and Delta (min) is the maximum value of all elements in the absolute difference matrix_0i(k) Is the element of the kth row of the ith column in the absolute difference matrix.

Technical Field

The invention relates to the technical field of monitoring of a steam turbine generator unit, in particular to a fault determination method and system of the steam turbine generator unit.

Background

The fault diagnosis and treatment technology for the steam turbine generator unit is an applied engineering subject and relates to the fields of vibration mechanics, rotor dynamics, vibration measurement, vibration fault diagnosis, equipment design, equipment operation and maintenance and the like. At present, a shafting fault diagnosis system takes analysis vibration information as a main technical means, can judge whether a fault of a turbo generator unit occurs and determine the type of the fault, and the fault reason is determined by staff maintenance, along with the continuous maximization and complicated development of unit equipment, the staff maintenance accuracy and efficiency cannot meet the actual requirements, and a large amount of manpower and material resources are consumed, so that a technology capable of determining the vibration fault reason and the fault position is urgently needed.

Disclosure of Invention

The invention aims to provide a method and a system for determining faults of a steam turbine generator unit, which can determine the reasons and the positions of the faults so that workers can determine a fault repair scheme in time.

In order to achieve the purpose, the invention provides the following scheme:

a fault determination method for a steam turbine generator unit comprises the following steps:

acquiring fault vibration data when the steam turbine generator unit is in fault; the fault vibration data comprises a plurality of vibration factor data;

determining the fault type of the steam turbine generator unit according to the fault vibration data;

acquiring various state parameters which can cause the fault type in the steam turbine generator unit;

constructing a fault type parameter matrix according to the state parameters and the fault vibration data;

and determining the fault reason and the fault position of the steam turbine generator unit according to the fault type parameter matrix.

Optionally, the obtaining of the fault vibration data when the steam turbine generator unit fails specifically includes:

acquiring current vibration data of a steam turbine generator unit;

judging whether the difference value between the current vibration data and the vibration reference value is smaller than a difference value threshold value or not to obtain a first judgment result;

and if the first judgment result is negative, determining that the current vibration data are fault vibration data.

Optionally, determining the fault type of the steam turbine generator unit according to the fault vibration data specifically includes:

determining any fault type in the fault type set as a current fault type; the fault type set comprises a plurality of fault types which can be caused by vibration of the steam turbine generator unit;

acquiring a plurality of rules of the current fault type and a plurality of symptom credibility corresponding to each rule respectively;

determining the product of the minimum value of the plurality of symptom credibility corresponding to each rule and the rule credibility coefficient as the rule credibility corresponding to each rule;

determining the maximum value of the rule credibility as the fault credibility of the current fault type;

traversing all fault types to obtain a plurality of fault credibility;

and determining the fault type corresponding to the maximum value in the multiple regular fault credibility as the fault type of the steam turbine generator unit.

Optionally, the fault type parameter matrix is:

wherein, X'₀Is a vibration factor matrix, x'₀(1)、x'₀(2) And x'₀(N) are respectively vibration factors of 1 st, 2 nd and N th, X'_iA parameter matrix of the ith vibration factor; x'_i＝[x'_i(1),x'_i(2),...，x'_i(N)]^T，i＝1,2,...,n，x'_i(k) A kth state parameter, k ═ 1,2,. and N, which affects the ith vibration factor; n is the type quantity of the state parameters influencing the ith vibration factor, and N is the type number of the vibration factors in the fault vibration data.

Optionally, the determining, according to the fault type parameter matrix, a fault cause and a fault location of the steam turbine generator unit specifically includes:

carrying out non-dimensionalization on the fault type parameter matrix to obtain a non-dimensionalized fault type parameter matrix;

determining the absolute value of the difference value between each element in the first column of elements in the non-dimensionalized fault type parameter matrix and each element in the row where the element is located, so as to obtain an absolute difference value matrix;

using formulasProcessing the absolute difference matrix to obtain a correlation coefficient matrix;

calculating an average value of each row of elements in the correlation coefficient matrix, and determining a state parameter corresponding to a maximum value in the average values as a fault state parameter;

determining the fault reason and the fault position of the steam turbine generator unit according to the fault state parameters;

wherein ξ_0i(k) The correlation coefficient of the kth state parameter of the ith vibration factor is Delta (min) is the minimum value of all elements in the absolute difference matrix, rho is the resolution coefficient, Delta (max) is the maximum value of all elements in the absolute difference matrix, and Delta (min) is the maximum value of all elements in the absolute difference matrix_0i(k) Is the element of the kth row of the ith column in the absolute difference matrix.

A fault determination system for a steam turbine generator unit, comprising:

the fault vibration data acquisition module is used for acquiring fault vibration data when the steam turbine generator unit fails; the fault vibration data comprises a plurality of vibration factor data;

the fault type determining module is used for determining the fault type of the steam turbine generator unit according to the fault vibration data;

the state parameter acquisition module is used for acquiring various state parameters which can cause the fault type in the steam turbine generator unit;

the fault type parameter matrix construction module is used for constructing a fault type parameter matrix according to the state parameters and the fault vibration data;

and the fault diagnosis module is used for determining the fault reason and the fault position of the steam turbine generator unit according to the fault type parameter matrix.

Optionally, the obtaining module of the fault vibration data specifically includes:

the current vibration data acquisition unit is used for acquiring current vibration data of the steam turbine generator unit;

the judging unit is used for judging whether the difference value between the current vibration data and the vibration reference value is smaller than a difference value threshold value or not to obtain a first judging result; if the first judgment result is negative, calling a fault vibration data determination unit;

and the fault vibration data determining unit is used for determining that the current vibration data are fault vibration data.

Optionally, the fault type determining module specifically includes:

the current fault type determining unit is used for determining any fault type in the fault type set as a current fault type; the fault type set comprises a plurality of fault types which can be caused by vibration of the steam turbine generator unit;

the rule obtaining unit is used for obtaining a plurality of rules of the current fault type and a plurality of symptom credibility corresponding to each rule;

the rule credibility determining unit is used for determining the product of the minimum value of the plurality of symptom credibility corresponding to each rule and the rule credibility coefficient as the rule credibility corresponding to each rule;

the fault reliability determining unit is used for determining that the maximum value in the rule reliability is the fault reliability of the current fault type;

the fault type traversing unit is used for traversing all fault types to obtain a plurality of fault credibility;

and the fault type judging unit is used for determining the fault type corresponding to the maximum value in the fault credibility of the plurality of rules as the fault type of the steam turbine generator unit.

Optionally, the fault type parameter matrix is:

wherein, X'₀Is a vibration factor matrix, x'₀(1)、x'₀(2) And x'₀(N) are respectively vibration factors of 1 st, 2 nd and N th, X'_iA parameter matrix of the ith vibration factor; x'_i＝[x'_i(1),x'_i(2),...，x'_i(N)]^T，i＝1,2,...,n，x'_i(k) A kth state parameter, k ═ 1,2,. and N, which affects the ith vibration factor; n is the type quantity of the state parameters influencing the ith vibration factor, and N is the type number of the vibration factors in the fault vibration data.

Optionally, the fault diagnosis module specifically includes:

the non-dimensionalization unit is used for carrying out non-dimensionalization processing on the fault type parameter matrix to obtain a non-dimensionalized fault type parameter matrix;

an absolute difference matrix determining unit, configured to determine an absolute value of a difference between each element in the first column of elements in the dimensionless fault type parameter matrix and each element in the row of the dimensionless fault type parameter matrix, to obtain an absolute difference matrix;

a correlation coefficient matrix determination unit for using a formulaProcessing the absolute difference matrix to obtain a correlation coefficient matrix;

the fault state parameter determining unit is used for calculating the average value of each row of elements in the correlation coefficient matrix and determining the state parameter corresponding to the maximum value in the average values as the fault state parameter;

the fault diagnosis unit is used for determining the fault reason and the fault position of the steam turbine generator unit according to the fault state parameters;

wherein ξ_0i(k) The correlation coefficient of the kth state parameter of the ith vibration factor is Delta (min) is the minimum value of all elements in the absolute difference matrix, rho is the resolution coefficient, Delta (max) is the maximum value of all elements in the absolute difference matrix, and Delta (min) is the maximum value of all elements in the absolute difference matrix_0i(k) Is the element of the kth row of the ith column in the absolute difference matrix.

According to the specific embodiment provided by the invention, the invention discloses the following technical effects:

the invention provides a method and a system for determining faults of a steam turbine generator unit, wherein the method comprises the following steps: acquiring fault vibration data when the steam turbine generator unit is in fault; the fault vibration data comprises a plurality of vibration factor data; determining the fault type of the steam turbine generator unit according to the fault vibration data; acquiring various state parameters which can cause fault types in a steam turbine generator unit; constructing a fault type parameter matrix according to the state parameters and the fault vibration data; and determining the fault reason and the fault position of the steam turbine generator unit according to the fault type parameter matrix. According to the invention, the reason and the position of the fault can be determined by establishing the fault type parameter matrix, so that a worker can determine the fault repairing scheme in time.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.

FIG. 1 is a flow chart of a method for determining a fault of a steam turbine generator unit according to an embodiment of the present invention;

FIG. 2 is a flow chart of the fault determination of the turbo generator set according to the embodiment of the present invention;

fig. 3 is a diagram of a fault determination system of a turbo generator set according to an embodiment of the present invention.

Detailed Description

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 only a part of the embodiments of the present invention, and not all of the embodiments. 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 aims to provide a method and a system for determining faults of a steam turbine generator unit, which can determine the reasons and the positions of the faults so that workers can determine a fault repair scheme in time.

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.

Fig. 1 is a flowchart of a method for determining a fault of a steam turbine generator unit according to an embodiment of the present invention, and as shown in fig. 1, the present invention provides a method for determining a fault of a steam turbine generator unit, including:

step 101: acquiring fault vibration data when the steam turbine generator unit is in fault; the fault vibration data comprises a plurality of vibration factor data;

step 102: determining the fault type of the steam turbine generator unit according to the fault vibration data;

step 103: acquiring various state parameters which can cause fault types in a steam turbine generator unit;

step 104: constructing a fault type parameter matrix according to the state parameters and the fault vibration data;

step 105: and determining the fault reason and the fault position of the steam turbine generator unit according to the fault type parameter matrix.

Step 101, specifically comprising:

acquiring current vibration data of a steam turbine generator unit;

judging whether the difference value between the current vibration data and the vibration reference value is smaller than a difference value threshold value or not to obtain a first judgment result;

and if the first judgment result is negative, determining that the current vibration data is fault vibration data.

Step 102, specifically comprising:

determining any fault type in the fault type set as a current fault type; the fault type set comprises a plurality of fault types which can be caused by vibration of the steam turbine generator unit;

acquiring a plurality of rules of the current fault type and a plurality of symptom credibility corresponding to each rule respectively;

determining the product of the minimum value of the plurality of symptom credibility corresponding to each rule and the rule credibility coefficient as the rule credibility corresponding to each rule;

determining the maximum value of the reliability of the plurality of rules as the reliability of the current fault type;

traversing all fault types to obtain a plurality of fault credibility;

and determining the fault type corresponding to the maximum value in the multiple regular fault credibility as the fault type of the steam turbine generator unit.

Specifically, the fault type parameter matrix is:

wherein, X'₀Is a vibration factor matrix, x'₀(1)、x'₀(2) And x'₀(N) are respectively vibration factors of 1 st, 2 nd and N th, X'_iA parameter matrix of the ith vibration factor; x'_i＝[x'_i(1),x'_i(2),...，x'_i(N)]^T，i＝1,2,...,n，x'_i(k) A kth state parameter, k ═ 1,2,. and N, which affects the ith vibration factor; n is the type quantity of the state parameters influencing the ith vibration factor, and N is the type number of the vibration factors in the fault vibration data.

Step 105, specifically comprising:

carrying out non-dimensionalization on the fault type parameter matrix to obtain a non-dimensionalized fault type parameter matrix;

determining the absolute value of the difference value between each element in the first column of elements in the dimensionless fault type parameter matrix and each element in the row to obtain an absolute difference value matrix;

using formulasProcessing the absolute difference matrix to obtain a correlation coefficient matrix;

calculating the average value of each row of elements in the correlation coefficient matrix, and determining the state parameter corresponding to the maximum value in the average values as a fault state parameter;

determining the fault reason and the fault position of the steam turbine generator unit according to the fault state parameters;

wherein ξ_0i(k) The correlation coefficient of the kth state parameter of the ith vibration factor is Delta (min) is the minimum value of all elements in the absolute difference matrix, rho is the resolution coefficient, Delta (max) is the maximum value of all elements in the absolute difference matrix, and Delta (min) is the maximum value of all elements in the absolute difference matrix_0i(k) Is the element of the kth row of the ith column in the absolute difference matrix.

The vibration symptom only reflects part of the information of the fault, so the vibration fault property of the steam turbine generator unit can not be determined by one symptom, and other symptoms such as steam temperature, steam pressure, exciting current, maintenance data and the like must be introduced for judgment. Because the relation between the unit fault and the symptom and between the symptom and the symptom is complex, although the same symptom corresponds to a plurality of faults, the sensitivity degrees of different faults are not necessarily the same; one fault corresponds to multiple symptoms, but different symptoms play different roles in identifying the fault; how to embody the difference of the significance degree of the symptom in the rule and how to determine the logical relationship between the multiple symptoms before the rule is the main problem to be considered when organizing the rule. When a diagnosis rule is constructed, the rule premise generally comprises a plurality of different types of fault symptoms, information of different aspects of the fault is reflected, and the rule embodies the comprehensive action of the various symptoms, so that the fault identification and verification are accurately carried out by utilizing a plurality of rules.

The invention is based on a conventional vibration diagnosis mechanism, integrates design, operation and maintenance data of a steam turbine generator unit, and emphasizes on providing a multi-dimensional big data diagnosis method of the steam turbine generator unit suitable for engineering practice from the aspect of solving the actual engineering.

Fig. 2 is a flow chart of fault determination of a turbo generator set according to an embodiment of the present invention, and as shown in fig. 2, the present invention technically specifically solves the following problems:

1. data acquisition: the operating parameters (state parameters) of the steam Turbine generator unit are taken from a power plant SIS (Safety Instrumented System) and a TDM (rotating machine diagnosis, detection and management) System, the operating parameters comprise displacement, speed, acceleration vibration, steam temperature, steam pressure and bearing position, vibration data comprise vibration amplitude, phase and frequency, rotor position parameters in the bearing and the like, and in addition, the invention also uses maintenance parameters. An ETL (Extract-Transform-Load) tool extracts data (such as relational data, plane data and the like) in a data source to a temporary intermediate layer, then performs cleaning, conversion and integration, and finally loads the data to a data warehouse or a data set to become a basis for online analysis processing and data mining.

The device database completes effective analysis of mass data, and imports the data from the front end into a centralized large distributed database or a distributed storage cluster, and performs data cleaning and preprocessing work on the basis of importation.

2. Data processing: selecting a single device which normally works as an object, and selecting the same measuring point under the same working condition for repeated measurement for multiple times. In order to obtain a certain statistical accuracy, generally, the number of times of measurement N is 20-25, a numerical characteristic value is obtained from the measurement result, and big data statistical calculation is performed on the vibration and various relation parameters to obtain a reference value of the vibration and relation data of the unit under various working conditions. Calculating the arithmetic mean value and the standard deviation of the data repeatedly measured for multiple times at the same measuring point under the same working condition, wherein the formula is as follows:

in the formula (I), the compound is shown in the specification,is an arithmetic mean value, x₁、x₂、Respectively the 1 st, 2 nd and nth measured values of the same measuring point under the same working condition₁＝1,2,3,…,N₁，N₁The number of measurements; sigma_nIs the standard deviation.

Determining an attention point value M for vibration data_a＝M_n+2σ_nAnd a hazard point value M_d＝M_n+3σ_nWherein the arithmetic mean is the reference value.

3. The diagnosis method comprises the following steps: in order to determine the vibration fault property and reason, the diagnosis system calculates the reliability of vibration symptom by using a rule diagnosis mode, and deduces the fault property according to the reliability, wherein the fault with the highest reliability is regarded as the fault property existing in the unit; and performing grey correlation degree calculation on the operation data, the overhaul data and the vibration data by using a grey correlation analysis method. And determining the factor with the highest fault association degree as a specific cause causing the vibration fault, and pushing out a targeted fault treatment measure.

3.1 vibration symptom confidence calculation

The system monitors the on-line state of the vibration of the unit in real time, and can acquire all symptom characteristic data of the vibration such as rotating speed, vibration waveform, frequency spectrum, frequency doubling amplitude, phase and the like from the system and calculate all symptom credibility.

The reliability calculation of the symptom is mainly divided into two methods, namely an instantaneous value method and a change rate method:

3.1.1 transient type Condition confidence calculation

The instantaneous value type condition is determined by data at a certain moment, and the reliability is any value in [0,1] when the symptoms are that the rotating speed is more than twice of the first-order critical rotating speed, the amplitude of a first frequency multiplication in a vibration frequency spectrum is large, and the unit operates under load.

For example, taking the symptom reliability indicating "a frequency component has a large amplitude in the vibration spectrum" as an example, a method of comparing the amplitude of the frequency component with the amplitudes of other frequency components may be adopted, as shown in the following equation:

CF (1X) is the symptom credibility of the symptom that the amplitude of a certain frequency component in the vibration spectrum is large; a (X), A (TX), A (LX) and A (HX) respectively have the amplitudes of the frequency, the pass frequency, the low frequency and the high frequency, a1, a2 and a3 are proportionality coefficients smaller than 1 and are related to fault properties, a1, a2 and a3 are determined mainly according to experience, and the numerical values reflect the influence degrees of different frequency components on the components. The same condition describes that the calculation method is not the same in different faults.

3.1.2 Rate of Change type Condition confidence calculation

The change rate type condition is determined by data at different time, and when calculating the confidence that the amplitude fluctuation is large when the rotating speed is not changed, the change amount of two sets of data can be compared with the set threshold value when the rotating speed is changed by less than 10rpm, as shown in the following formula:

wherein cf (X) is the confidence of symptom X; a. the₁、A₀、A_bRepresenting a real-time value, an initial value and a threshold value of the amplitude, respectively.

The reliability coefficients of different frequency bands in instantaneous value type calculation and the threshold value of amplitude change in change rate type calculation have close relation with the type, fault mechanism, unit load and process parameters of the unit, and the key and difficulty of fault mode identification are accurately determined.

Specifically, when the fault is diagnosed, an influence factor if (impact factor) needs to be further introduced to measure the influence degree of the fault on the unit vibration. For example, for a vibration failure of a rotating part, for example, when the vibration amplitude of the unit is 60 μm at 3000 rpm, the vibration amplitude of the unit is 120 μm after 2 seconds, the main influencing factors of the vibration failure of the rotating part are time, rotating speed and load, and based on 3000 rpm, the influencing factor IF of 3000 rpm after 2 seconds is 120/60-2.

The severity of the fault is denoted by SF, which is the product of the fault confidence and the fault impact factor, i.e., SF ═ CF × IF.

The SF determination criteria are as follows:

when SF is more than or equal to 0 and less than 0.3, the unit is normal in operation;

SF is less than or equal to 0.3 and less than 0.5, so that the normal operation of the unit is not influenced;

when SF is more than or equal to 0.5 and less than 0.8, warning is given, and the unit can operate in a short period;

if SF is more than or equal to 0.8, the machine is dangerous and should be stopped for processing as soon as possible.

When the SF value reaches the warning value, the system outputs a diagnosis result. The diagnostic results are shown in table 1:

TABLE 1 diagnostic results

Serial number	Name of failure	Degree of confidence
			1	Dropping of rotating parts	1
2	Mass unbalance	0.6
			3	Rotor thermal bending	0.46

In order to further confirm the specific factors of the fault, the system analyzes and processes the input operation parameters and vibration data, calculates the reliability of the instantaneous value type condition, and can confirm that the abnormal vibration component is mainly composed of power frequency components; calculating according to the change rate type condition reliability, obtaining the change amplitude of the power frequency vibration vector, entering the next diagnosis when the change amplitude is larger than a threshold value, and calculating the vibration mutation time difference (1-3 seconds); and (4) calculating according to the change rate type condition reliability, and comparing with a set vibration stability threshold value to judge whether the vibration is kept stable after the vibration mutation. According to the vibration characteristics and the time characteristics, the highest credibility of the diagnosis conclusion of the falling of the rotating part can be obtained through diagnosis by utilizing a forward and reverse reasoning rule, namely the falling of the rotating part with the fault name and the credibility of 1 are output. And the falling position of the rotating part can be further deduced by combining the variation amplitude of the power frequency vibration vector of the measuring point and the unbalanced response characteristic of the rotor at the position of the measuring point.

3.2 relational data relevancy calculation

Calculating the association degree of each operation parameter (such as overhigh or overlow vacuum, deviation of the unit from rated cycle operation, too fast load increase and decrease and the like), the position parameter of the bearing and the abnormal vibration time, and diagnosing the high association degree as a specific factor causing the fault. And performing grey correlation degree calculation on the operation data, the bearing position data and the vibration data input by the system by using a grey correlation analysis method.

3.2.1 determining the sequence of the assay

In order to realize qualitative analysis of the vibration fault cause, firstly, a dependent variable factor (vibration factor, such as vibration vector variation of a certain measuring point, frequency component amplitude and the like) and a plurality of independent variable factors (state parameters, such as steam temperature, steam pressure, exciting current, reactive power and the like) are determined. Let dependent variable data constitute a reference sequence (X'₀) The respective variable data constitute the comparison sequence (X'_i(i ═ 1, 2.., n)), and the fault type parameter matrix is obtained as:

wherein, X'₀Is a vibration factor matrix, x'₀(1)、x'₀(2) And x'₀(N) are respectively vibration factors of 1 st, 2 nd and N th, X'_iA parameter matrix of the ith vibration factor; x'_i＝[x'_i(1),x'_i(2),...，x'_i(N)]^T，i＝1,2,...,n，x'_i(k) A kth state parameter, k ═ 1,2,. and N, which affects the ith vibration factor; n is the type quantity of the state parameters influencing the ith vibration factor, and N is the type number of the vibration factors in the fault vibration data.

3.2.2 non-dimensionalization of the Fault type parameter matrix

The original fault type parameter matrix has different dimensions or orders of magnitude, in order to ensure the reliability of the analysis result, the fault type parameter matrix needs to be subjected to non-dimensionalization, an averaging method is adopted, namely, each row of data is averaged, and each value of each row is divided by the average value of the data of the row to obtain the non-dimensionalized fault type parameter matrix:

wherein, X₀For a dimensionless vibration factor matrix, x₀(1)、x₀(2) And x₀(N) vibration factors of 1 st, 2 nd and N 'respectively, X'_iA parameter matrix of the ith vibration factor which is non-dimensionalized; x_i＝[x_i(1),x_i(2),...，x_i(N)]^T，i＝1,2,...,n，x_i(k) For a k-th state parameter affecting the i-th vibration factor in a non-dimensionalized manner, k is 1, 2.

For example,represents a dimensionless value of the row 1, column 1 vibration data.

3.2.3 sequence of differencing, maximum and minimum differences

Calculating the absolute difference value of the corresponding period of the first column (reference sequence: vibration data dimensionless value) and the rest columns (comparison sequence: operation data dimensionless value) in the dimensionless fault type parameter matrix to form the following absolute difference value matrix:

wherein Δ_0i(k)＝|x₀(k)-x_i(k) I ═ 0,1,. n; k is 1, 2., N, i.e., the absolute difference between the dimensionless value of the kth row 1 column vibration data and the dimensionless value of the kth row i column operating data.

The maximum number and the minimum number in the absolute difference matrix are the maximum difference and the minimum difference:

3.2.4 calculating the correlation coefficient

And transforming the data in the absolute difference matrix as follows:obtaining a correlation coefficient matrix:

wherein ξ_0i(k) The correlation coefficient of the kth state parameter of the ith vibration factor is Delta (min) is the minimum value of all elements in the absolute difference matrix, rho is the resolution coefficient, Delta (max) is the maximum value of all elements in the absolute difference matrix, and Delta (min) is the maximum value of all elements in the absolute difference matrix_0i(k) Is the element of the kth row of the ith column in the absolute difference matrix. The value of the resolution coefficient rho is within (0,1), and according to experience, the value of rho is preferably less than or equal to 0.5, and the correlation coefficient xi is_0i(k) Is a positive number not exceeding 1, which reflects the ith comparison sequence X_iWith reference sequence X₀The degree of association at the kth stage.

3.2.5 calculating the degree of association

Comparison of sequences X_i(operating data) with reference sequence X₀The degree of correlation of (vibration data) is reflected by N correlation coefficients, and X is obtained by averaging_i(operating data) with X₀Degree of association of (vibration data):

3.2.6 ordering by relevance: and (3) sequencing the association degree of each comparison sequence and the reference sequence in a descending order, wherein the greater the association degree is, the more consistent the change situation of the comparison sequence and the reference sequence is.

Fig. 3 is a structural diagram of a fault determination system of a steam turbine generator unit according to an embodiment of the present invention, and as shown in fig. 3, the present invention provides a fault determination system of a steam turbine generator unit, including:

the fault vibration data acquisition module 301 is used for acquiring fault vibration data when the steam turbine generator unit fails; the fault vibration data comprises a plurality of vibration factor data;

a fault type determination module 302, configured to determine a fault type of the steam turbine generator unit according to the fault vibration data;

a state parameter obtaining module 303, configured to obtain multiple state parameters that can cause a fault type in the steam turbine generator unit;

a fault type parameter matrix construction module 304, configured to construct a fault type parameter matrix according to the state parameters and the fault vibration data;

and the fault diagnosis module 305 is configured to determine a fault reason and a fault position of the steam turbine generator unit according to the fault type parameter matrix.

The fault vibration data acquisition module 301 specifically includes:

the current vibration data acquisition unit is used for acquiring current vibration data of the steam turbine generator unit;

the judging unit is used for judging whether the difference value between the current vibration data and the vibration reference value is smaller than a difference value threshold value or not to obtain a first judging result; if the first judgment result is negative, calling a fault vibration data determination unit;

and the fault vibration data determining unit is used for determining that the current vibration data are fault vibration data.

The fault type determining module 302 specifically includes:

the current fault type determining unit is used for determining any fault type in the fault type set as a current fault type; the fault type set comprises a plurality of fault types which can be caused by vibration of the steam turbine generator unit;

the rule obtaining unit is used for obtaining a plurality of rules of the current fault type and a plurality of symptom credibility corresponding to each rule;

the rule credibility determining unit is used for determining the product of the minimum value of the plurality of symptom credibility corresponding to each rule and the rule credibility coefficient as the rule credibility corresponding to each rule;

the fault reliability determining unit is used for determining that the maximum value in the reliability of the plurality of rules is the fault reliability of the current fault type;

the fault type traversing unit is used for traversing all fault types to obtain a plurality of fault credibility;

and the fault type judging unit is used for determining the fault type corresponding to the maximum value in the fault credibility of the plurality of rules as the fault type of the steam turbine generator unit.

The fault type parameter matrix is:

wherein, X'₀Is a vibration factor matrix, x'₀(1)、x'₀(2) And x'₀(N) are respectively vibration factors of 1 st, 2 nd and N th, X'_iA parameter matrix of the ith vibration factor; x'_i＝[x'_i(1),x'_i(2),...，x'_i(N)]^T，i＝1,2,...,n，x'_i(k) A kth state parameter, k ═ 1,2,. and N, which affects the ith vibration factor; n is the type quantity of the state parameters influencing the ith vibration factor, and N is the type number of the vibration factors in the fault vibration data.

The fault diagnosis module 305 specifically includes:

the non-dimensionalization unit is used for carrying out non-dimensionalization processing on the fault type parameter matrix to obtain a non-dimensionalized fault type parameter matrix;

the absolute difference matrix determining unit is used for determining the absolute value of the difference between each element in the first column of elements in the dimensionless fault type parameter matrix and each element in the corresponding row to obtain an absolute difference matrix;

a correlation coefficient matrix determination unit for using a formulaFor absolute difference matrixProcessing to obtain a correlation coefficient matrix;

the fault state parameter determining unit is used for calculating the average value of each row of elements in the correlation coefficient matrix and determining the state parameter corresponding to the maximum value in the average values as the fault state parameter;

the fault diagnosis unit is used for determining the fault reason and the fault position of the steam turbine generator unit according to the fault state parameters;

wherein ξ_0i(k) The correlation coefficient of the kth state parameter of the ith vibration factor is Delta (min) is the minimum value of all elements in the absolute difference matrix, rho is the resolution coefficient, Delta (max) is the maximum value of all elements in the absolute difference matrix, and Delta (min) is the maximum value of all elements in the absolute difference matrix_0i(k) Is the element of the kth row of the ith column in the absolute difference matrix.

The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.

The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.