阅读说明：本技术 电力变压器的健康度评价方法和装置 (Health degree evaluation method and device for power transformer ) 是由 王坚俊 孙林涛 邹晖 刘昌标 周国伟 彭晨光 于 2021-08-20 设计创作，主要内容包括：本申请公开了一种电力变压器的健康度评价方法和装置,具体为构建待评价电力变压器的包括多个故障类型和多个初始评价指标的状态评价因素集；计算所述初始评价指标的常权重系数；计算所述故障类型与所述初始评价指标的带权综合关联度；对所述带权综合关联度进行筛选,得到有效综合关联度；基于所述多个故障类型、所述有效综合关联度、所述常权重系数和变权重系数进行处理,得到所述待评价电力变压器的实时健康状态。通过实时健康状况可以及时发现潜伏性故障,从而能够制定科学合理的检修计划,避免了过检修或欠检修情况的出现。同时通过分析变压器故障得分,能够初步得到故障类型的倾向性信息,对于检修工作能够提供了决策支持。(The application discloses a method and a device for evaluating the health degree of a power transformer, which particularly comprises the steps of constructing a state evaluation factor set of the power transformer to be evaluated, wherein the state evaluation factor set comprises a plurality of fault types and a plurality of initial evaluation indexes; calculating a constant weight coefficient of the initial evaluation index; calculating the weighted comprehensive association degree of the fault type and the initial evaluation index; screening the weighted comprehensive association degree to obtain an effective comprehensive association degree; and processing the fault types, the effective comprehensive association degrees, the constant weight coefficient and the variable weight coefficient to obtain the real-time health state of the power transformer to be evaluated. Latent faults can be found in time through real-time health conditions, so that a scientific and reasonable maintenance plan can be formulated, and the condition of over-maintenance or under-maintenance is avoided. Meanwhile, by analyzing the transformer fault score, the tendency information of the fault type can be obtained preliminarily, and decision support can be provided for the maintenance work.)

1. A health degree evaluation method of a power transformer is characterized by comprising the following steps:

constructing a state evaluation factor set of the power transformer to be evaluated, wherein the state evaluation factor set comprises a plurality of fault types and a plurality of initial evaluation indexes;

calculating a constant weight coefficient of the initial evaluation index;

calculating the weighted comprehensive association degree of the fault type and the initial evaluation index;

screening the weighted comprehensive association degree to obtain an effective comprehensive association degree;

and processing the fault types, the effective comprehensive association degrees, the constant weight coefficient and the variable weight coefficient to obtain the real-time health state of the power transformer to be evaluated.

2. The method for evaluating health degree according to claim 1, wherein the calculating of the constant weight coefficient of the initial evaluation index includes the steps of:

determining an order relation of the plurality of initial evaluation indexes;

judging the relative importance among the plurality of initial evaluation indexes;

and calculating the constant weight coefficient.

3. The health evaluation method according to claim 1, wherein the calculating of the comprehensive degree of association of the fault type with the initial evaluation index includes the steps of:

and calculating the fault type and the initial evaluation index based on a TOPSIS improved grey correlation degree analysis method to obtain the comprehensive correlation degree.

4. The health evaluation method of claim 1, wherein the screening of the weighted comprehensive association to obtain the effective comprehensive association comprises the steps of:

sequencing all the weighted comprehensive association degrees;

and screening the weighted comprehensive association degree based on a preset threshold value to obtain the effective comprehensive association degree.

5. A health degree evaluation device for a power transformer, the health degree evaluation device comprising:

the system comprises a data set construction module, a state evaluation factor set and a state evaluation factor setting module, wherein the data set construction module is configured to construct the state evaluation factor set of the power transformer to be evaluated, and the state evaluation factor set comprises a plurality of fault types and a plurality of initial evaluation indexes;

a first calculation module configured to calculate a constant weight coefficient of the initial evaluation index;

the second calculation module is configured to calculate a weighted comprehensive association degree of the fault type and the initial evaluation index;

the data screening module is configured to screen the weighted comprehensive association degree to obtain an effective comprehensive association degree;

and the evaluation execution module is configured to perform processing based on the plurality of fault types, the effective comprehensive association degree, the constant weight coefficient and the variable weight coefficient to obtain the real-time health state of the power transformer to be evaluated.

6. The health evaluation apparatus according to claim 5, wherein the first calculation module includes:

an order relation determination unit configured to determine an order relation of the plurality of initial evaluation indexes;

a judgment execution unit for judging the relative importance among the plurality of initial evaluation indexes;

and the calculation execution unit is used for calculating the constant weight coefficient.

7. The health assessment apparatus of claim 5, wherein said second calculation module is configured to calculate said fault type and said initial assessment indicator based on a TOPSIS improved grey correlation analysis method, resulting in said comprehensive correlation.

8. The health evaluation apparatus of claim 5, wherein the data filtering module comprises:

the sequencing execution unit is used for sequencing all the weighted comprehensive association degrees;

and the screening execution module is used for screening the weighted comprehensive association degree based on a preset threshold value to obtain the effective comprehensive association degree.

Technical Field

The application relates to the technical field of electric equipment, in particular to a method and a device for evaluating health degree of a power transformer.

Background

The health condition of the power equipment directly influences the safe and stable operation of the power system. The power transformer is used as a hub device of a power transmission system, has a vital function on the safe operation of a power grid, and once the operation state of the power transformer is abnormal, the power transformer not only can damage the service life of the equipment, but also can affect the power grid and other equipment connected with the power transformer, and can cause large-area power failure in serious conditions, thereby causing immeasurable loss to social economy and people life.

The power transformer has a complex structure and a plurality of assemblies, and generally comprises a body, a sleeve, a tap switch, a non-electric quantity protection device, a cooler and other parts, and in the operation process of the power transformer, the health condition of any assembly is gradually deteriorated with probability due to the influence of internal and external acting forces, so that the whole abnormal or fault state of the transformer is caused.

For a long time, the health level and the stability of the operation state of many power transformers are mainly ensured at home and abroad by means of regular maintenance. However, in the actual operation process, even if the power transformers are power transformers of the same manufacturer and the same batch and have similar operation environments, the health conditions of the power transformers are not completely the same, so that the power transformers which do not need to be overhauled (overhauled), and some power transformers which need to be overhauled in advance cannot be overhauled in time (underoverhauled) easily occur in the overhaul period.

Disclosure of Invention

In view of this, the present application provides a method and an apparatus for evaluating health degree of a power transformer, which are used to obtain real-time health degree of the power transformer, so that a reasonable maintenance plan can be formulated according to the real-time health degree, and the occurrence of over-maintenance or under-maintenance situations is avoided.

In order to achieve the above object, the following solutions are proposed:

a health degree evaluation method of a power transformer comprises the following steps:

constructing a state evaluation factor set of the power transformer to be evaluated, wherein the state evaluation factor set comprises a plurality of fault types and a plurality of initial evaluation indexes;

calculating a constant weight coefficient of the initial evaluation index;

calculating the weighted comprehensive association degree of the fault type and the initial evaluation index;

screening the weighted comprehensive association degree to obtain an effective comprehensive association degree;

and processing the fault types, the effective comprehensive association degrees, the constant weight coefficient and the variable weight coefficient to obtain the real-time health state of the power transformer to be evaluated.

Optionally, the calculating a constant weighting factor of the initial evaluation index includes:

determining an order relation of the plurality of initial evaluation indexes;

judging the relative importance among the plurality of initial evaluation indexes;

and calculating the constant weight coefficient.

Optionally, the calculating a comprehensive association degree between the fault type and the initial evaluation index includes:

and calculating the fault type and the initial evaluation index based on a TOPSIS improved grey correlation degree analysis method to obtain the comprehensive correlation degree.

Optionally, the screening the weighted comprehensive association degree to obtain an effective comprehensive association degree includes:

sequencing all the weighted comprehensive association degrees;

and screening the weighted comprehensive association degree based on a preset threshold value to obtain the effective comprehensive association degree.

A health degree evaluation device of a power transformer, the health degree evaluation device comprising:

the system comprises a data set construction module, a state evaluation factor set and a state evaluation factor setting module, wherein the data set construction module is configured to construct the state evaluation factor set of the power transformer to be evaluated, and the state evaluation factor set comprises a plurality of fault types and a plurality of initial evaluation indexes;

a first calculation module configured to calculate a constant weight coefficient of the initial evaluation index;

the second calculation module is configured to calculate a weighted comprehensive association degree of the fault type and the initial evaluation index;

the data screening module is configured to screen the weighted comprehensive association degree to obtain an effective comprehensive association degree;

and the evaluation execution module is configured to perform processing based on the plurality of fault types, the effective comprehensive association degree, the constant weight coefficient and the variable weight coefficient to obtain the real-time health state of the power transformer to be evaluated.

Optionally, the first computing module includes:

an order relation determination unit configured to determine an order relation of the plurality of initial evaluation indexes;

a judgment execution unit for judging the relative importance among the plurality of initial evaluation indexes;

and the calculation execution unit is used for calculating the constant weight coefficient.

Optionally, the second calculating module is configured to calculate the fault type and the initial evaluation index based on a TOPSIS improved grey correlation degree analysis method, so as to obtain the comprehensive correlation degree.

Optionally, the data screening module includes:

the sequencing execution unit is used for sequencing all the weighted comprehensive association degrees;

and the screening execution module is used for screening the weighted comprehensive association degree based on a preset threshold value to obtain the effective comprehensive association degree.

According to the technical scheme, the method and the device for evaluating the health degree of the power transformer are disclosed, and specifically a state evaluation factor set comprising a plurality of fault types and a plurality of initial evaluation indexes of the power transformer to be evaluated is constructed; calculating a constant weight coefficient of the initial evaluation index; calculating the weighted comprehensive association degree of the fault type and the initial evaluation index; screening the weighted comprehensive association degree to obtain an effective comprehensive association degree; and processing the fault types, the effective comprehensive association degrees, the constant weight coefficient and the variable weight coefficient to obtain the real-time health state of the power transformer to be evaluated. Latent faults can be found in time through real-time health conditions, so that a scientific and reasonable maintenance plan can be formulated, and the condition of over-maintenance or under-maintenance is avoided. Meanwhile, by analyzing the transformer fault score, the tendency information of the fault type can be obtained preliminarily, and decision support can be provided for the maintenance work.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a flowchart of a method for evaluating health of a power transformer according to an embodiment of the present disclosure;

FIG. 2 is a schematic diagram of improved gray correlation analysis distance;

fig. 3 is a block diagram of a health evaluation apparatus for a power transformer according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, 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 application.

Example one

Fig. 1 is a flowchart of a method for evaluating health of a power transformer according to an embodiment of the present application.

As shown in fig. 1, the health degree evaluation method provided in this embodiment is used for evaluating the real-time health degree of a power transformer to be evaluated, and specifically includes the following steps:

and S1, constructing a state evaluation factor set of the power transformer to be evaluated.

The state evaluation factor set includes a plurality of fault types and a plurality of initial evaluation indexes. Specifically, fault types and fault monitoring quantities of 9 power transformers are extracted according to fault types and fault monitoring quantities given by 'oil immersed transformer (reactor) state evaluation guide (Q/GDW 169-2008)' power equipment preventive experiment regulation 'and' analysis and judgment guide (GB/T7252-2001 ') in transformer oil' of national grid company, and test and monitoring single state quantities of 24 transformers are selected as initial evaluation indexes as shown in table 1.

TABLE 1 Power Transformer Fault types

TABLE 2 evaluation index of initial state of power transformer

And S2, calculating a constant weight coefficient of the initial evaluation index.

After determining a plurality of initial evaluation indexes, the calculation of the constant weight coefficient is realized by the following steps:

first, the order relationship of a plurality of initial evaluation indexes is determined.

Definition 1 for a certain evaluation criterion, if the evaluation index x_iGreater than x_jThen is marked as x_i>x_j。

Definition 2 evaluation index x₁,x₂,…,x_mFor the evaluation criteria are:

x₁*>x₂*>…>x_m*

then attribute x₁,x₂,…,x_mAnd establishing an order relation according to the descending importance degree.

Establishing an evaluation index set { x }₁,x₂,……,x_mThe order relation steps are as follows:

(x) decision maker in index set { x)₁,x₂,……,x_mIn the above, one index which is most important for a certain evaluation target is selected and marked as x₁*；

Secondly, the decision maker selects the most important index relative to a certain evaluation target from the rest m-1 indexes and marks the index as x₂*；

Thirdly, repeating the steps, selecting for m-1 times, and marking the rest evaluation index as x_m*；

Determining an evaluation index set { x }₁,x₂,……,x_mThe order relationship of.

Without loss of generality, the above equation is written as:

x₁>x₂>…>x_m

then, the relative importance between the initial evaluation indexes is judged

Define 3 decision maker for index x_kAnd x_k-1Is judged by its weight w_k-1And w_kThe ratio of (A) to (B) is denoted as r_kAs follows:

r_kthe assigned values of (c) are referred to as shown in table 3.

TABLE 3r_kAssignment reference

Finally, the constant weight coefficients are calculated.

The calculation formula of the constant weight coefficient is as follows:

and obtaining the weight of other evaluation indexes, namely:

and finally obtaining a subjective weight coefficient determined by a sequence relation method, and recording the subjective weight coefficient as:

W＝[ω₁,ω₂，……,ω_m]

and S3, calculating the weighted comprehensive association degree.

Gray correlation analysis is one of the important branches of the Gray system theory proposed in 1982 by professor Dengdong. The basic idea is to judge whether the correlation between different data sequences is tight according to the similarity of the data sequence curve geometry. In the multi-index comprehensive evaluation problem, a forward ideal sequence is constructed, and the correlation degree of each evaluation object and the forward ideal sequence is calculated, wherein the greater the correlation degree is, the more similar the forward ideal sequence is, and the better the evaluation result is. The comprehensive evaluation method based on grey correlation analysis has the following advantages:

(1) the calculation is simple, and the practicability and operability are strong;

(2) the mathematical method is a non-statistical method, and has no requirements on the number of samples and the distribution of the samples;

(3) the situation that the quantitative result of the relevance is inconsistent with the qualitative analysis can not occur.

The grey correlation analysis procedure was as follows:

assuming n evaluation objects, each objectIf m initial evaluation indexes exist and t groups of historical fault data exist, the ideal evaluation object X is positively evaluated⁺Can be expressed as:

X⁺＝[X₁ ⁺，X₂ ⁺，…，X_n ⁺]^T

wherein X_i ⁺＝[X_i1 ⁺,X_i2 ⁺,…,X_im ⁺]I is 1,2, …, n, X which is the positive ideal evaluation object of the i-th fault_ik ⁺Which represents the maximum number of occurrences of fault i in the t sets of historical fault data, k is 1,2, …, m.

For the jth set of data (j ═ 1,2, …, t), the fault observation matrix X can be expressed as:

note X_iAnd X_i ⁺The grey correlation coefficient at the kth index is:

in the formula (I), the compound is shown in the specification,ΔX_ij＝|X_ij ⁺-x_ijin order to reduce the influence of extreme values on the evaluation result, a resolution coefficient rho is introduced, and rho belongs to [0, 1 ]]In the invention, rho is taken to be 0.5.

Note X_iAnd X_i+ gray degree of association xi_iThen, then

The method is used for calculating the weighted comprehensive relevance between the fault type and the initial evaluation index based on the improved grey relevance of TOPSIS.

The gray degree of association is essentially a geometric "distance", and the greater the degree of association, the closer the evaluation object X is geometrically to the ideal object X +, as shown in fig. 2. However, assuming that the evaluation indexes are simplified to 2 and have the same weight, the distance between the rating index and the ideal evaluation object is the same, and it is difficult to determine the importance degree only by the gray correlation degree. Therefore, the traditional grey correlation analysis is improved by combining the idea of the ideal improvement method (TOPSIS), and the intrinsic meaning of the distance between the rating index and an ideal evaluation object is deeply mined.

TOPSIS improved grey correlation degree analysis organically combines the distance defined by a traditional ideal solution and the grey correlation degree to construct a positive and negative bidirectional evaluation method. The basic idea can be described as that m fault types and n state monitoring indexes are set, and the index value is x_ijFor the ith scheme, an observation matrix Xi is constructed and an ideal evaluation object X is positively arranged⁺And negative ideal evaluation object X^-And obtaining the positive gray correlation degree and the negative gray correlation degree corresponding to the scheme, and comprehensively considering the positive gray correlation degree and the negative gray correlation degree to obtain the optimal judgment.

Let negative ideal evaluation object X-be expressed as:

X^-＝[X₁ ^-，X₂ ^-，…，X_n ^-]^T

similar to the calculation process of gray correlation analysis, the negative gray correlation calculation formula is:

in the formula

Analysis from a geometric perspectiveThe evaluation thought of the gray correlation analysis improved by the ideal solution method is as follows: if the object X is evaluated₁And X₂To the forward ideal object X⁺If the distances of (2) are equal, only the evaluation object to X is considered⁺Cannot distinguish X₁And X₂High or low. However, as can be seen in conjunction with the distance from the evaluation object to the negative ideal object, X₂And X^-Is farther away, and comprehensively considered, X can be made₂Is superior to X₁And (4) judging. To this end, the overall relevance of the different fault types to the positive and negative ideal solutions is calculated in this section:

in combination with the subjective weights calculated above, are determined by

Z′_i＝ω_iγ_i，i＝1，2，…，9

And calculating to obtain the weighted comprehensive association Zi'. To avoid loss of generality, Zi' is normalized to obtain Zi as follows:

and S4, obtaining the effective comprehensive relevance through screening the weighted comprehensive relevance.

In view of the fact that the number of initial evaluation indexes is large and the field operation is not facilitated to be carried out, the contribution of some indexes in the state evaluation process is small, and therefore the transformer state evaluation index system is excessively complex. Thus, the following steps are performed:

first, the weighted integrated relevance vector Z obtained above is compared_iThe values of the elements in the vector are sorted from big to small, and the sorted elements and indexes represented by the elements are updated to a vector Z_iPerforming the following steps; then, a threshold value T is set, and on the premise that the threshold value T is not exceeded, for Z_iThe weighted comprehensive association degree values of the inner indexes are accumulated according to the sequence from large to small, and when the value of the last index isAdding the evaluation index to enable the accumulated value to be larger than or equal to the threshold value, stopping adding the evaluation index, deleting the non-accumulated comprehensive association degree and the index represented by the comprehensive association degree, defining the remained evaluation index as a key index, and recording the key index as Z_i ^*，Z_i ^*The selection method of (2) is as follows:

and S5, calculating the real-time health degree of the power transformer.

According to the transformer state evaluation guide rule and the evaluation standard, a method for evaluating the health score of the transformer by using score production is provided, and the score expression is as follows:

in the formula, omega_fIs a fault weight value; x is the number of_f,jAnd ω_f,jRespectively scoring and weighting key indexes; and m is the number of key indexes corresponding to a certain fault.

Considering the problem that the normal weight coefficient cannot change due to the change of the operation degradation degree of the transformer, when the degradation degree of the state monitoring amount has a large difference, the weight coefficient cannot play a good balance role for different fault types, and a large deviation may occur in the equipment health state evaluation result. Therefore, the invention adopts a variable weight coefficient calculation method to determine the weight coefficient of each fault type:

in the formula is omega'_i，ω_iAnd y_iThe variable weight coefficient, the constant weight coefficient and the score value of the fault type Fi are respectively, n is 9, and the method takes omega_i1/9, take α as 0.

According to the technical scheme, the embodiment provides the method for evaluating the health degree of the power transformer, and specifically comprises the steps of constructing a state evaluation factor set of the power transformer to be evaluated, wherein the state evaluation factor set comprises a plurality of fault types and a plurality of initial evaluation indexes; calculating a constant weight coefficient of the initial evaluation index; calculating the weighted comprehensive association degree of the fault type and the initial evaluation index; screening the weighted comprehensive association degree to obtain an effective comprehensive association degree; and processing the fault types, the effective comprehensive association degrees, the constant weight coefficient and the variable weight coefficient to obtain the real-time health state of the power transformer to be evaluated. Latent faults can be found in time through real-time health conditions, so that a scientific and reasonable maintenance plan can be formulated, and the condition of over-maintenance or under-maintenance is avoided. Meanwhile, by analyzing the transformer fault score, the tendency information of the fault type can be obtained preliminarily, and decision support can be provided for the maintenance work.

The following is a specific embodiment of the present application.

A220 kV oil-immersed power transformer (SFPS9.180000/220) of a certain transformer substation in Henan province is selected as a case for analysis.

The key index system of the transformer is evaluated according to the steps listed in the invention, different faults of the transformer are scored, the variable weight coefficient is calculated according to the degradation degree represented by the score value, and the result is shown in the table below. It can be seen that the lower the score of a fault type, the greater its weight coefficient, which means that more serious fault types are valued to a greater extent in this model.

And (4) synthesizing the fault score values of the 9 transformers in the table to obtain the integral operation state score of 0.3427 of the transformer, and looking up a table to know that the transformer operates in an abnormal state at present. And further analyzing the scores of the fault types, wherein the score of the fault type F1 is obviously abnormal, so that the transformer is known to have winding faults and needs to be scheduled for maintenance as soon as possible. The transformer is detected in detail after power failure, and the winding of the transformer is influenced by external force, so that the winding is deformed, and the distance between the two blocks is changed.

The above example analysis shows that the evaluation result of the health condition of the transformer obtained by the method provided by the invention is basically consistent with the actual condition, and the analysis shows that the evaluation method of the health degree of the transformer based on the TOPSIS improved grey correlation analysis can obtain the health score close to the actual condition, and can give the probability distribution of the fault type according to the calculation process, thereby providing preliminary guidance for the maintenance decision.

Example two

Fig. 3 is a block diagram of a health evaluation apparatus for a power transformer according to an embodiment of the present application.

As shown in fig. 3, the health degree evaluation device provided in this embodiment is used for evaluating the real-time health degree of a power transformer to be evaluated, and specifically includes a data set construction module 10, a first calculation module 20, a second calculation module 30, a data screening module 40, and an evaluation execution module 50.

The data set construction module is used for constructing a state evaluation factor set of the power transformer to be evaluated.

The state evaluation factor set includes a plurality of fault types and a plurality of initial evaluation indexes. Specifically, fault types and fault monitoring quantities of 9 power transformers are extracted according to fault types and fault monitoring quantities given by 'oil immersed transformer (reactor) state evaluation guide (Q/GDW 169-2008)' power equipment preventive experiment regulation 'and' analysis and judgment guide (GB/T7252-2001 ') in transformer oil' of national grid company, and test and monitoring single state quantities of 24 transformers are selected as initial evaluation indexes as shown in table 1.

TABLE 1 Power Transformer Fault types

TABLE 2 evaluation index of initial state of power transformer

The first calculation module is used for calculating the constant weight coefficient of the initial evaluation index

The module comprises an order relation determining unit, a judging execution unit and a calculating execution unit.

The order relation determining unit is used for determining order relations of the plurality of initial evaluation indexes.

Definition 1 for a certain evaluation criterion, if the evaluation index x_iGreater than x_jThen is marked as x_i>x_j。

Definition 2 evaluation index x₁,x₂,…,x_mFor the evaluation criteria are:

x₁*>x₂*>…>x_m*

then attribute x₁,x₂,…,x_mAnd establishing an order relation according to the descending importance degree.

Establishing an evaluation index set { x }₁,x₂,……,x_mThe order relation steps are as follows:

(x) decision maker in index set { x)₁,x₂,……,x_mIn the above, one index which is most important for a certain evaluation target is selected and marked as x₁*；

Secondly, the decision maker selects the most important index relative to a certain evaluation target from the rest m-1 indexes and marks the index as x₂*；

Thirdly, repeating the steps, selecting for m-1 times, and marking the rest evaluation index as x_m*；

Determining an evaluation index set { x }₁,x₂,……,x_mThe order relationship of.

Without loss of generality, the above equation is written as:

x₁>x₂>…>x_m

the judgment execution unit is used for judging the relative importance between the initial evaluation indexes.

Define 3 decision maker for index x_kAnd x_k-1Is judged by its weight w_k-1And w_kThe ratio of (A) to (B) is denoted as r_kAs follows:

r_kthe assigned values of (c) are referred to as shown in table 3.

TABLE 3r_kAssignment reference

The calculation execution unit is used for calculating the constant weight coefficient.

The calculation formula of the constant weight coefficient is as follows:

and obtaining the weight of other evaluation indexes, namely:

and finally obtaining a subjective weight coefficient determined by a sequence relation method, and recording the subjective weight coefficient as:

W＝[ω₁,ω₂，……,ω_m]

the second calculation module is used for calculating the weighted comprehensive association degree.

Gray correlation analysis is one of the important branches of the Gray system theory proposed in 1982 by professor Dengdong. The basic idea is to judge whether the correlation between different data sequences is tight according to the similarity of the data sequence curve geometry. In the multi-index comprehensive evaluation problem, a forward ideal sequence is constructed, and the correlation degree of each evaluation object and the forward ideal sequence is calculated, wherein the greater the correlation degree is, the more similar the forward ideal sequence is, and the better the evaluation result is. The comprehensive evaluation method based on grey correlation analysis has the following advantages:

(1) the calculation is simple, and the practicability and operability are strong;

(2) the mathematical method is a non-statistical method, and has no requirements on the number of samples and the distribution of the samples;

(3) the situation that the quantitative result of the relevance is inconsistent with the qualitative analysis can not occur.

The grey correlation analysis procedure was as follows:

assuming n evaluation objects, each object has m initial evaluation indexes, and t groups of historical fault data exist, the forward ideal evaluation object X is⁺Can be expressed as:

X⁺＝[X₁ ⁺，X₂ ⁺，…，X_n ⁺]^T

wherein X_i ⁺＝[X_i1 ⁺,X_i2 ⁺,…,X_im ⁺]I is 1,2, …, n, X which is the positive ideal evaluation object of the i-th fault_ik ⁺Which represents the maximum number of occurrences of fault i in the t sets of historical fault data, k is 1,2, …, m.

For the jth set of data (j ═ 1,2, …, t), the fault observation matrix X can be expressed as:

note X_iAnd X_i ⁺The grey correlation coefficient at the kth index is:

in the formula (I), the compound is shown in the specification,ΔX_ij＝|X_ij ⁺-X_ijin order to reduce the influence of extreme values on the evaluation result, a resolution coefficient rho is introduced, and rho belongs to [0, 1 ]]In the invention, rho is taken to be 0.5.

Note X_iAnd X_i+ gray degree of association xi_iThen, then

The method is used for calculating the weighted comprehensive relevance between the fault type and the initial evaluation index based on the improved grey relevance of TOPSIS.

The gray degree of association is essentially a geometric "distance", and the greater the degree of association, the closer the evaluation object X is geometrically to the ideal object X +, as shown in fig. 2. However, assuming that the evaluation indexes are simplified to 2 and have the same weight, the distance between the rating index and the ideal evaluation object is the same, and it is difficult to determine the importance degree only by the gray correlation degree. Therefore, the traditional grey correlation analysis is improved by combining the idea of the ideal improvement method (TOPSIS), and the intrinsic meaning of the distance between the rating index and an ideal evaluation object is deeply mined.

TOPSIS improved grey correlation degree analysis organically combines the distance defined by a traditional ideal solution and the grey correlation degree to construct a positive and negative bidirectional evaluation method. The basic idea can be described as that m fault types and n state monitoring indexes are set, and the index value is x_ijFor the ith scheme, an observation matrix Xi is constructed and an ideal evaluation object X is positively arranged⁺And obtaining a positive gray correlation degree and a negative gray correlation degree corresponding to the scheme according to the negative ideal evaluation object X-, and comprehensively considering the positive gray correlation degree and the negative gray correlation degree to obtain optimal judgment.

Let negative ideal evaluation object X-be expressed as:

X^-＝[X₁ ^-，X₂ ^-，…，X_n ^-]^T

similar to the calculation process of gray correlation analysis, the negative gray correlation calculation formula is:

in the formula

From the analysis of a geometric angle, the evaluation thought of the gray correlation analysis improved by an ideal solution method is as follows: if the object X is evaluated₁And X₂To the forward ideal object X⁺If the distances of (2) are equal, only the evaluation object to X is considered⁺Cannot distinguish X₁And X₂High or low. However, as can be seen in conjunction with the distance from the evaluation object to the negative ideal object, X₂The farther the distance from X-, the comprehensive consideration can be made into X₂Is superior to X₁And (4) judging. To this end, the overall relevance of the different fault types to the positive and negative ideal solutions is calculated in this section:

in combination with the subjective weights calculated above, are determined by

Z′_i＝ω_iγ_i，i＝1，2，…，9

And calculating to obtain the weighted comprehensive association Zi'. To avoid loss of generality, Zi' is normalized to obtain Zi as follows:

and the data screening module is used for screening the weighted comprehensive association degree to obtain the effective comprehensive association degree.

In view of the fact that the number of initial evaluation indexes is large and the field operation is not facilitated to be carried out, the contribution of some indexes in the state evaluation process is small, and therefore the transformer state evaluation index system is excessively complex. The module includes a sorting execution unit and a screening execution unit.

The sequencing execution unit is used for carrying out the weighted comprehensive relevance vector Z obtained above_iThe values of the elements in the vector are sorted from big to small, and the sorted elements and indexes represented by the elements are updated to a vector Z_iPerforming the following steps; the screening execution unit is used for screening Z on the premise of not exceeding a preset threshold value T_iAccumulating the weighted comprehensive association degree values of the inner indexes in the descending order, and when the addition of the last index causes the accumulated value to be larger than or equal to the threshold value, stopping the addition, deleting the comprehensive association degree which is not accumulated and the indexes represented by the comprehensive association degree, defining the remained evaluation indexes as key indexes, and recording the key indexes as Z_i ^*，Z_i ^*The selection method of (2) is as follows:

and the evaluation execution module is used for calculating the real-time health degree of the power transformer.

According to the transformer state evaluation guide rule and the evaluation standard, a method for evaluating the health score of the transformer by using score production is provided, and the score expression is as follows:

in the formula, omega_fIs a fault weight value; x is the number of_f,jAnd ω_f,jRespectively scoring and weighting key indexes; and m is the number of key indexes corresponding to a certain fault.

Considering the problem that the normal weight coefficient cannot change due to the change of the operation degradation degree of the transformer, when the degradation degree of the state monitoring amount has a large difference, the weight coefficient cannot play a good balance role for different fault types, and a large deviation may occur in the equipment health state evaluation result. Therefore, the invention adopts a variable weight coefficient calculation method to determine the weight coefficient of each fault type:

in the formula is omega'_i，ω_iAnd y_iThe variable weight coefficient, the constant weight coefficient and the score value of the fault type Fi are respectively, n is 9, and the method takes omega_i1/9, take α as 0.

As can be seen from the above technical solutions, the present embodiment provides a health degree evaluation device for a power transformer, which is specifically configured to construct a state evaluation factor set including a plurality of fault types and a plurality of initial evaluation indexes for the power transformer to be evaluated; calculating a constant weight coefficient of the initial evaluation index; calculating the weighted comprehensive association degree of the fault type and the initial evaluation index; screening the weighted comprehensive association degree to obtain an effective comprehensive association degree; and processing the fault types, the effective comprehensive association degrees, the constant weight coefficient and the variable weight coefficient to obtain the real-time health state of the power transformer to be evaluated. Latent faults can be found in time through real-time health conditions, so that a scientific and reasonable maintenance plan can be formulated, and the condition of over-maintenance or under-maintenance is avoided. Meanwhile, by analyzing the transformer fault score, the tendency information of the fault type can be obtained preliminarily, and decision support can be provided for the maintenance work.

The embodiments in the present specification 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.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of 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, embodiments of 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.

Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (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 terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, 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 terminal 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 terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.

The technical solutions provided by the present invention are described in detail above, and the principle and the implementation of the present invention are explained in this document by applying specific examples, and the descriptions of the above examples are only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.