CN106780108A - A kind of distribution transformer state evaluating method based on improvement evidential reasoning fusion - Google Patents

Abstract

The present invention relates to the technical field of power equipment, in particular to a distribution transformer state evaluation method based on improved evidence reasoning fusion, including: step 1, according to the data of distribution transformer test information, inspection information, and operation information to be evaluated, establish Distribution transformer status evaluation secondary index system; step 2, using fuzzy evaluation method to obtain the second-level index evidence of the distribution transformer to be evaluated about the comment set The initial basic probability distribution; step 3, use the improved analytic hierarchy process to determine the weight of the relative importance of indicators in each layer, and use the formula to calculate its confidence coefficient; step 4, modify the initial evidence source through the confidence coefficient to obtain the basic probability distribution ; Step 5, use the improved evidence reasoning synthesis rule to carry out evidence fusion, so as to obtain the final state of the distribution transformer. This evaluation method establishes a hierarchical and multi-level state evaluation model of distribution transformers by analyzing the state quantity information of distribution transformers.

Description

A Distribution Transformer Condition Assessment Method Based on Improved Evidence Inference Fusion

technical field

The invention belongs to the technical field of electric power equipment, and in particular relates to a distribution transformer state evaluation method based on improved evidence reasoning fusion.

Background technique

As the core equipment of the power distribution system, the distribution transformer directly affects the safety and stability of the power consumption of the entire power user. Evaluating the health status of distribution transformers can help power operators to keep abreast of the health status of distribution transformers in order to repair and maintain them. Therefore, it is of great research significance to correctly evaluate the status of distribution transformers.

Distribution transformers contain multiple components, and each component has different state information, so a scientific and effective information processing model is the core of distribution transformer state evaluation. However, there is a large amount of information that affects the final state of distribution transformers. These information reflect the state of the transformer from different aspects and levels, and have associated uncertainty and ambiguity. Using these state quantities directly will not allow accurate evaluation.

Application No. 201310136283.8 is an invention patent titled "Method for State Evaluation of Distribution Network Transformer Based on Fuzzy Theory". Although it improves work efficiency and the accuracy of state evaluation, it can evaluate the state of the transformer as a whole and each component. The final evaluation result is based on the degree of possibility It overcomes the one-sidedness of the traditional evaluation results to determine the status. However, gray association is used in the core information fusion part of the evaluation method, which cannot solve the uncertainty and conflict that affect the association of many state quantities of distribution transformers.

Contents of the invention

The purpose of the present invention is to establish a hierarchical evaluation system for multi-attribute decision-making by analyzing the state quantity information of distribution transformers. On the basis of fuzzy comprehensive evaluation and analytic hierarchy Layer multilevel state assessment model. It provides a correct evaluation method of distribution transformer status by selecting and processing it among numerous distribution transformer information.

In order to achieve the above purpose, the technical solution adopted by the present invention is: a distribution transformer state assessment method based on improved evidence reasoning fusion, including the following steps:

Step 1. According to the distribution transformer test information, inspection information, and operation information data to be evaluated, establish a secondary index system for distribution transformer status evaluation;

Step 2. Use the fuzzy evaluation method to obtain the initial basic probability distribution of the second-level index evidence of the distribution transformer to be evaluated about the comment set V;

Step 3, using the improved analytic hierarchy process to determine the weight of the relative importance of each layer index, and using the formula to calculate its confidence coefficient;

Step 4. Modify the initial evidence source through the confidence coefficient to obtain the basic probability distribution;

Step 5. Use the improved evidence reasoning synthesis rule to carry out evidence fusion, so as to obtain the final state of the distribution transformer.

In the above-mentioned distribution transformer state assessment method based on improved evidence reasoning fusion, the specific method of step 2 is as follows:

1) Divide the comment set V into four different states: good, general, attention, and abnormal, V={v ₁ , v ₂ , v ₃ , v ₄ }={good, general, attention, abnormal};

2) Determine the degree of membership of the evaluation index for the comment set; including the following steps:

a. Evaluation indicators that can be directly quantified are determined by relative deterioration degree;

b. The assessment indicators that are fuzzy and cannot be quantified adopt a detailed bonus table, and the evaluation indicators are added according to the bonus table to quantify their status.

In the above-mentioned distribution transformer state assessment method based on improved evidence reasoning fusion, the calculation formula of the confidence coefficient in step 3 is:

β _ij = w _ij /w _ik (1);

In formula (1), β _ij is the confidence coefficient, and the weights of the lower-level individual indicators {x _i1 , x _i2 ,..., x _ij ,..., x _iJ } of the first-level index i are {w _i1 , w _i2 ,...,w _ij ,...,w _iJ }, and w _ik is the maximum value of the above weight values.

In the above-mentioned distribution transformer state assessment method based on improved evidence reasoning fusion, the basic probability assignment formula in step 4 is:

m _ij (v _n )=β _ij μ _ij,n (2);

In formula (2), m _ij (v _n ) is the basic probability distribution, β _ij is the confidence coefficient, and μ _ij,n is the probability assignment.

In the above distribution transformer state assessment method based on improved evidence reasoning fusion, the improved evidence reasoning synthesis rule described in step 5 is:

m _i(j+1) (Θ)＝m _ij+ 1(Θ) m _i(j) (Θ)+K' δ _i (v _Θ ,m) (4);

In formulas (3) and (4), K' is the revised conflict factor, δ _i (v _Θ ,m) is the conflict allocation ratio, and m _i(j+1) (Θ) is the uncertain set probability allocation.

The beneficial effects of the present invention are: adopting the improved evidence reasoning fusion method solves the uncertainties and conflicts that affect the association among numerous state quantity information of distribution transformers, not only allows people to assign reliability to a single element of the hypothesis space, but also It can also be assigned to its subset, which is very similar to the evidence collection process of human beings at all levels of abstraction, avoiding a certain state quantity deviation and causing a huge deviation in the result. On the basis of fuzzy comprehensive evaluation and analytic hierarchy process, the fusion of evidence reasoning is improved, and a hierarchical and multi-level state evaluation model of distribution transformers is established, which can evaluate the state of distribution transformers as a whole and each component, overcoming the traditional evaluation The results are one-sided, and the anti-interference ability is strong and the convergence speed is fast.

Description of drawings

Fig. 1 is the flowchart of an embodiment of the present invention;

Fig. 2 is a distribution transformer status evaluation index system diagram of an embodiment of the present invention;

Fig. 3 is a graph of the membership function of an evaluation index according to an embodiment of the present invention.

detailed description

Embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings.

Examples of the described embodiments are shown in the drawings, wherein like or similar reference numerals designate like or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

The following disclosure provides many different embodiments or examples for implementing different structures of the present invention. To simplify the disclosure of the present invention, components and arrangements of specific examples are described below. They are examples only and are not intended to limit the invention. Furthermore, the present invention may repeat reference numerals and/or letters in different instances. This repetition is for the purpose of simplicity and clarity and does not in itself indicate a relationship between the various embodiments and/or arrangements discussed. In addition, various specific examples of processes and materials are provided herein, but one of ordinary skill in the art may recognize the applicability of other processes and/or the use of other materials. Additionally, configurations described below in which a first feature is "on" a second feature may include embodiments where the first and second features are formed in direct contact, and may include additional features formed between the first and second features. For example, such that the first and second features may not be in direct contact.

In the description of the present invention, it should be noted that, unless otherwise specified and limited, the terms "connected" and "connected" should be understood in a broad sense, for example, it can be a mechanical connection or an electrical connection, and it can also be the internal communication of two elements. It may be directly connected or indirectly connected through an intermediary. Those of ordinary skill in the related art can understand the specific meanings of the above terms according to specific situations.

This embodiment adopts the following technical solution: a distribution transformer state assessment method based on improved evidence reasoning fusion, including the following steps:

Step 1. According to the distribution transformer test information, inspection information, and operation information data to be evaluated, establish a secondary index system for distribution transformer status evaluation;

Step 2. Use the fuzzy evaluation method to obtain the initial basic probability distribution of the second-level index evidence of the distribution transformer to be evaluated about the comment set V;

Step 3, using the improved analytic hierarchy process to determine the weight of the relative importance of each layer index, and using the formula to calculate its confidence coefficient;

Step 4. Modify the initial evidence source through the confidence coefficient to obtain the basic probability distribution;

Step 5. Use the improved evidence reasoning synthesis rule to carry out evidence fusion, so as to obtain the final state of the distribution transformer.

Further, the specific method of the step 2 is as follows:

1) Divide the comment set V into four different states: good, general, attention, and abnormal, V={v ₁ , v ₂ , v3,v ₄ }={good, general, attention, abnormal};

2) Determine the degree of membership of the evaluation index for the comment set; including the following steps:

a. Evaluation indicators that can be directly quantified are determined by relative deterioration degree;

b. The assessment indicators that are fuzzy and cannot be quantified adopt a detailed bonus table, and the evaluation indicators are added according to the bonus table to quantify their status.

Further, the formula for calculating the confidence coefficient in step 3 is:

β _ij = w _ij /w _ik (1);

In formula (1), β _ij is the confidence coefficient, and the weights of the lower-level individual indicators {x _i1 , x _i2 ,..., x _ij ,..., x _iJ } of the first-level index i are {w _i1 , w _i2 ,...,w _ij ,...,w _iJ }, and w _ik is the maximum value of the above weight values.

Further, the basic probability distribution formula described in step 4 is:

m _ij (v _n )=β _ij μ _ij,n (2);

In formula (2), m _ij (v _n ) is the basic probability distribution, β _ij is the confidence coefficient, and μ _ij,n is the probability assignment.

Furthermore, the improved evidence reasoning synthesis rule described in step 5 is:

m _i(j+1) (Θ)=m _ij+1 (Θ) m _i(j) (Θ)+K' δ _i (v _Θ ,m) (4);

In formulas (3) and (4), K' is the revised conflict factor, δ _i (v _Θ ,m) is the conflict allocation ratio, and m _i(j+1) (Θ) is the uncertain set probability allocation.

Example 1

As shown in Figure 1, a distribution transformer status assessment model based on improved evidence inference fusion is established, including the following specific steps:

S1. Under the principle of combining comprehensiveness, unity and hierarchy, select the evaluation indicators from the components of the distribution transformer to be evaluated, and establish a multi-level evaluation system for the status of the distribution transformer.

Since the status evaluation of distribution transformers is affected by various indicators, the selection of hierarchical indicators should follow the principles of comprehensiveness, unity and combination of levels, and the qualitative and quantitative indicators that are clearly defined and available should be selected as much as possible. In this embodiment 1, the oil-immersed distribution transformer is taken as an example. According to the maintenance characteristics of the oil-immersed distribution transformer, the components such as windings, bushings, tap changers, oil tanks, insulating oil, grounding, non-electrical protection, and identification of the transformer are analyzed. Select the evaluation index quantity in and establish the distribution transformer status evaluation system, as shown in Figure 2.

S2. Apply the fuzzy evaluation method to obtain the initial basic probability distribution of the second-level index evidence of the distribution transformer to be evaluated with respect to the comment set V.

In order to facilitate inspection personnel to accurately judge the health status of power transformers and take appropriate measures reasonably, combined with relevant literature, the status of distribution transformers is divided into four categories: "good", "general", "attention" and "abnormal". state. which is:

V＝{v ₁ ,v ₂ ,v ₃ ,v ₄ }＝{good, common, attention, abnormal}

Before comprehensive evaluation, the membership degree of each index to the comment set should be determined first. In fuzzy mathematics, a fuzzy subset in the domain of discourse can be mapped to [0,1] through the function μ(x), and the function μ(x) is the membership function, which is mapped to the value on [0,1] is the degree of membership.

In the distribution transformer state evaluation model, the indicators that can be directly quantified can be determined by the concept of relative deterioration degree. For indicators that are bigger and better, such as grounding resistance:

For the smaller the better type indicators such as joint temperature:

In formulas (1) and (2): d _i is the relative deterioration degree of the evaluation index of the distribution transformer; y _max or y _min is the limit value of the index, and its value can be determined by referring to relevant literature; y ₀ is the index’s Initial value; y _i is the actual measured value of the indicator.

For indicators that have certain ambiguity and cannot be quantified, such as pollution level, distribution transformer oil tank, tap changer, etc. In this embodiment 1, a detailed bonus table is formulated for the evaluation indicators that cannot be quantified. Assuming that the ideal operating condition is 0 points, the objects to be evaluated are added points according to the bonus table, and their states are quantified and mapped to the interval [0,1] superior.

The membership function of the evaluation index adopts triangular distribution function and semi-trapezoidal distribution function, as shown in Figure 3.

Determination of membership function: Substituting the relative deterioration degree determined above into the membership function, the relative deterioration degree of the combination of semi-trapezoid and triangle corresponds to the fuzzy boundary interval of the four states. Thus, the membership function of an evaluation index for state V={v ₁ ,v ₂ ,v ₃ ,v ₄ } can be obtained:

In the formulas (3), (4), (5), and (6), x is the relative deterioration degree of the evaluation index, and v ₁ (x)~v ₄ (x) represent the evaluation index corresponding to the state v ₁ ~v ₄ Membership function; the fuzzy evaluation value μ _ij,n of the lower attribute index x _ij of the first-level index i relative to the evaluation level v _n .

S3. Applying the improved analytic hierarchy process to calculate the weights of the relative importance of the indicators in each layer, and using the formula to calculate the confidence coefficient.

In this embodiment 1, the confidence coefficient is introduced to correct the initial evidence source, and the weight of each index is determined based on the improved analytic hierarchy process, assuming that the lower level single index {x _i1 ,x _i2 ,...,x _ij , The weights of ..., x _iJ } are respectively {w _i1 , w _i2 ,...,w _ij ,...,w _iJ }, let wi _ik be the maximum value of the above weights, then the confidence coefficient is:

β _ij =w _ij /w _ik

S4. Modify the initial evidence source through the confidence coefficient to obtain the basic probability distribution.

In the state evaluation system of distribution transformers, using the fuzzy membership function, the initial basic probability distribution of the second-level index evidence can be obtained, and its probability assignment is μ _ij,n , and the initial probability distribution can be obtained by modifying the initial probability distribution through the confidence coefficient Basic probability assignment:

m _ij (v _n )=β _ij μ _ij,n

S5. Use the improved evidence reasoning synthesis rule to carry out evidence fusion, so as to obtain the final state of the distribution transformer.

For the revised evidence source, the evidence conflict is assigned according to the confidence of the evidence. For the combination of the two basic probability assignments, there are the following rules:

m _i(2) (v _n )＝m _i1 (v _n )·m _i2 (v _n )+

m _i1 (v _n )·m _i2 (Θ)+m _i1 (Θ)·m _i2 (v _n )

+K'·δ _i (v _n ,m)

m _i(2) (Θ)＝m _i2 (Θ)·m _i1 (Θ)+K'·δ _i (v _Θ ,m)

in:

In the formula, K' is the corrected conflict factor, δ _i (v _n ,m) is the conflict distribution ratio, and m _i(2) (Θ) is the probability distribution of uncertain sets.

Similarly, by synthesizing evidence sources x _i1 , x _i2 ,...,x _ij+1 , the following recursive formula can be obtained:

m _i(j+1) (v _n )＝m _i(j) (v _n )·m _ij+1 (v _n )+

m _i(j) (v _n ) m _i(j) (Θ)+m _i(j) (Θ) m _ij+1 (v _n )

+K'·δ _i (v _n ,m)

m _i(j+1) (Θ)＝m _ij+1 (Θ) m _i(j) (Θ)+K' δ _i (v _Θ ,m)

The initial basic probability distribution of the first level can be obtained through the above synthesis rules, and the basic probability distribution of the first-level comprehensive index on the comment set V can be obtained by referring to the relevant knowledge about the conflict distribution of evidence sources. Then the distribution transformer evaluation results can be obtained through the above synthesis rules.

Example 2

The state evaluation of distribution transformers in a certain area is carried out, and the test information, inspection information, operation information and relative deterioration degree of each component are shown in Table 1.

Table 1 Relevant data of distribution transformer

Apply the fuzzy evaluation method to obtain the initial basic probability distribution of the comment set V of the second-level index evidence of the distribution transformer to be evaluated, see Table 2.

Table 2 Initial basic probability distribution of each index

Apply the improved analytic hierarchy process to calculate the weight of the relative importance of indicators in each layer, and use formula (10) to calculate its confidence coefficient, see Table 3

Table 3 Index weight and confidence coefficient

Taking the synthesis of evidence sources related to grounding devices as an example: the confidence coefficients of grounding resistance and grounding downconductor appearance are 1 and 0.580, respectively. After correcting the initial evidence sources, m ₅₁ (v ₁ )=0, m ₅₁ ( v ₂ )=0.607, m ₅₁ (v ₂ )=0.393, m ₅₁ (v ₁ )=0; m ₅₂ (v ₁ )=0.580, m ₅₂ (v ₂ )=0, m ₅₂ (v ₃ )=0 , m ₅₂ (v ₄ )=0. From m ₅₍₂₎ (v ₁ )=0.124, m ₅₍₂₎ (v ₂ )=0.478, m ₅₍₂₎ (v ₃ )=0.165, m ₅₍₂₎ (v ₄ )=0 , m ₅₍₂₎ (Θ)=0.091.

According to the same steps and methods, the indicators of each layer are synthesized to finally obtain the basic probability distribution of distribution transformers as m(v ₁ )=0.213, m(v ₂ )=0.562, m(v ₃ )=0.121, m(v ₄ )=0, m(Θ)=0.104, so it is judged that the operating state of the distribution transformer is in the "normal" state. The analysis of the data of the transformer shows that only the grounding resistance and the corrosion of the oil tank are near the attention value, and the other parts are in the normal state. better condition. After using the basic information and operating data of the distribution transformer to evaluate its status, the monitoring of the low-scoring parts of the distribution transformer was strengthened. down, but still functioning normally.

It should be understood that the parts not described in detail in this specification belong to the prior art.

Although the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, those of ordinary skill in the art should understand that these are only examples, and various variations or modifications can be made to these embodiments without departing from the principles and principles of the present invention. substance. The scope of the invention is limited only by the appended claims.

With reference to the following description and drawings, some specific implementation modes in the embodiments of the present invention are specifically disclosed to show some ways of implementing the principles of the embodiments of the present invention, but it should be understood that the scope of the embodiments of the present invention is not limited by this limit. On the contrary, the embodiments of the present invention include all changes, modifications and equivalents coming within the spirit and scope of the appended claims.

Claims (5)

1. it is a kind of based on the distribution transformer state evaluating method for improving evidential reasoning fusion, it is characterised in that including following step Suddenly：

Step 1, according to distribution transformer Test Information to be assessed, patrol and examine information, the data of operation information, set up distribution transformer Device state estimation two-level index system；

Step 2, distribution transformer second level index evidence to be assessed is obtained using THE FUZZY EVALUATING METHOD on the initial of Comment gathers V Basic probability assignment；

Step 3, determine the weight of each layer index relative importance using improved H, and it calculated using formula to put Coefficient of reliability；

Step 4, by confidence level coefficient to initial evidence source change obtain basic probability assignment；

Step 5, evidence fusion is carried out using improved evidential reasoning composition rule, so as to obtain the final shape of distribution transformer State.

2. as claimed in claim 1 based on the distribution transformer state evaluating method for improving evidential reasoning fusion, its feature exists In the specific method of the step 2 is as follows：

1) the Comment gathers V is divided into good, general, attention, abnormal four kinds of different conditions, V={ v₁,v₂,v₃,v₄}={ is good Good, general, attention, exception }；

2) degree of membership of the evaluation index for Comment gathers is determined；Comprise the following steps：

A. the evaluation index that can directly quantify is determined using relative inferiority degree；

B. ambiguity, it is impossible to which, using detailed bonus point table is formulated, the evaluation index is according to bonus point table for the evaluation index of quantification treatment Bonus point, its state is quantified.

3. as claimed in claim 1 based on the distribution transformer state evaluating method for improving evidential reasoning fusion, its feature exists In the computing formula of confidence level coefficient is described in step 3：

β_ij=w_ij/w_ik(1)；

β in formula (1)_ijIt is confidence level coefficient, the lower floor single index { x of 1 grade of index i_i1,x_i2,...,x_ij..., x_iJPower { w is respectively again_i1,w_i2,...,w_ij,...,w_iJ, w_ikIt is then the maximum of above-mentioned weighted value.

4. as claimed in claim 1 based on the distribution transformer state evaluating method for improving evidential reasoning fusion, its feature exists In basic probability assignment formula is described in step 4：

m_ij(v_n)=β_ijμ_ij,n(2)；

M in formula (2)_ij(v_n) it is basic probability assignment, β_ijIt is confidence level coefficient, μ_ij,nIt is probability assignment.

5. as claimed in claim 1 based on the distribution transformer state evaluating method for improving evidential reasoning fusion, its feature exists In improved evidential reasoning composition rule is described in step 5：

$\begin{array}{c}{m}_{i(j+1)}\left(v_n\right)=m_{i\left(j\right)}\left(v_n\right)\·m_{ij+1}\left(v_n\right)+m_{i\left(j\right)}\left(v_n\right)\·m_{i\left(j\right)}\left(\mathrm{\Θ}\right)+\\ m_{i\left(j\right)}\left(\mathrm{\Θ}\right)\·m_{ij+1}\left(v_n\right)+K^{\′}\·{\mathrm{\δ}}_i(v_n,m)\end{array}---\left(3\right);$

m_i(j+1)(Θ)=m_ij+1(Θ)·m_i(j)(Θ)+K'·δ_i(v_Θ,m) (4)；

K' is the revised conflict factor, δ in formula (3), (4)_i(v_Θ, m) it is conflict allocation proportion, m_i(j+1)(Θ) is uncertain Set probability assignments.