Fuzzy logic-based transformer health degree evaluation method and system
Through the transformer health assessment method based on fuzzy logic, using feature quantities and fuzzy inference technology, the singleness and inaccuracy of transformer fault detection and health status assessment in the prior art are solved, and more efficient and accurate evaluation results are achieved .
Patent Information
- Application Number
- CN202411366343.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-29
- Publication Date
- 2024-12-17
AI Technical Summary
The existing technology has a single method, untimely and inaccurate problem in transformer fault detection and health status assessment, especially in bad weather conditions, and the application effect of high-dimensional data based on statistics, clustering, neural networks and other methods is not good. , easily cause false detection and missed detection.
Using the transformer health evaluation method based on fuzzy logic, the health index evaluation value is constructed by collecting the characteristic quantities of the transformer such as temperature, voltage, current, pressure, noise, displacement and oil furfural concentration of the transformer, and the priority relationship matrix is calculated using the hierarchical analysis method. Membership function and fuzzy rules are used to perform fuzzy reasoning, combined with anti-fuzzing processing to obtain health assessment results.
It improves the timely rate of transformer fault detection and the accuracy of health status evaluation, reduces manual interference, enhances the objectivity and accuracy of evaluation results, and reduces the occurrence of false detection and missed detection.
Smart Images
Figure CN119150690A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of transformer evaluation, and in particular to a transformer health evaluation method and system based on fuzzy logic. Background Art
[0002] A transformer is a device that uses the principle of electromagnetic induction to change AC voltage. Its main components are the primary coil, the secondary coil and the iron core. The transformer is a device that converts AC voltage, alternating current and impedance. When AC current flows through the primary coil, AC magnetic flux is generated in the iron core, inducing voltage (or current) in the secondary coil. Once a transformer fails, it is not only not conducive to people's daily convenient life, but also poses a threat to personal safety.
[0003] At present, the methods used to detect transformer faults in power systems are relatively simple, basically through manual inspections, which is time-consuming and laborious. Especially in bad weather such as rain and snow, manual inspections cannot promptly and accurately determine the occurrence of faults. Often, they go to the site for diagnosis after the fault occurs. The timely rate of fault identification is low, and this diagnosis often relies on manual experience. Alternatively, transformer faults are detected by methods based on statistics, clustering, neural network classification models, etc. However, the application effect of statistical methods on high-dimensional data is not ideal, because the actual data distribution law cannot be predicted and it is difficult to characterize it with a single distribution model. It is difficult to determine the initial number of clusters when detecting outliers based on clustering methods, the detection effect is not ideal, and the time complexity is high. Classification-based methods have high requirements on the data quality of the training set. In addition, the existing transformer health status assessment is relatively simple, and the given evaluation indicators or weight distribution are affected by human interference factors, which are prone to false detection, missed detection, etc., resulting in unreasonable transformer status assessment and poor engineering application. Summary of the invention
[0004] In view of the above problems, the present invention is proposed to provide a transformer health assessment method and system based on fuzzy logic that overcomes all or at least part of the above problems.
[0005] According to one aspect of the present invention, a transformer health assessment method based on fuzzy logic is provided, comprising: Collecting characteristic quantities characterizing the health status of the transformer according to a preset period, wherein the characteristic quantities include at least one of the following: temperature, voltage, current, pressure, noise, displacement, oil level, and furfural concentration; Constructing health index evaluation values of the transformer corresponding to different cycles according to the characteristic quantities, and calculating a priority relationship matrix of the health index evaluation values by using a hierarchical analysis method; The priority relationship matrix is selected as the input of fuzzy logic reasoning, and the membership function and fuzzy rules are designed. Fuzzy reasoning is performed on the input according to the fuzzy rules to obtain the fuzzy state result of the transformer, the fuzzy state result is defuzzified, and the fuzzy set with the largest membership is selected as the health assessment result of the transformer.
[0006] Furthermore, the temperature includes the inner winding temperature, the core temperature and the oil temperature of the transformer.
[0007] Furthermore, the specific formula for constructing the health index evaluation value of the transformer corresponding to different cycles according to the characteristic quantity is: ; in, is the health index evaluation value, is the i-th feature quantity, for The weight of m is the number of features. (1,2,...,m), e is a constant, e=2.718.
[0008] Furthermore, the priority relationship matrix is constructed as follows: ; in, is the scale of the i-th index and the j-th index, (1,2,...,m).
[0009] Furthermore, the weight The calculation formula is: ; in, , represented as an intermediate calculated value, for The weight of m is the number of features. is the scale of the i-th indicator and the j-th indicator.
[0010] Furthermore, the membership function includes a triangle membership function and a quadrilateral membership function, and the health assessment results include safe, dangerous and serious.
[0011] Furthermore, the fuzzy rules are specifically: When conducting health assessment, each characteristic quantity corresponds to a fuzzy logic reasoning rule; The basis for constructing each fuzzy logic reasoning rule is specifically that the health state corresponding to a smaller feature quantity is safe, the health state corresponding to a medium feature quantity is dangerous, and the health state corresponding to a larger feature quantity is serious.
[0012] Furthermore, the initial expression of the membership function is: ; in, is a constant, E∈[0,1] , which controls the amplitude of the membership function, t is the value range of the membership function, and b is the translation amount.
[0013] Furthermore, the defuzzification of the state fuzzy result further includes: The state fuzzy result is defuzzified using the center of gravity method, and the specific formula of the center of gravity method is: ; ; in, is the x-coordinate of the center of gravity, is the y coordinate of the center of gravity, is the x-coordinate of the i-th fuzzy rule, is the y coordinate of the i-th fuzzy rule, The triangle membership function or quadrilateral membership function for the fuzzification input.
[0014] According to another aspect of the present invention, a transformer health assessment system based on fuzzy logic is provided, comprising: A collection module, used for collecting characteristic quantities representing the health status of the transformer according to a preset period, wherein the characteristic quantities include at least one of the following: temperature, voltage, current, pressure, noise, displacement, oil level and furfural concentration; A processing module, used to construct health index evaluation values corresponding to the transformer in different cycles according to the characteristic quantity, and calculate a priority relationship matrix of the health index evaluation values by using a hierarchical analysis method; An evaluation module is used to select the priority relationship matrix as the input of fuzzy logic reasoning, design membership functions and fuzzy rules, perform fuzzy reasoning on the input according to the fuzzy rules to obtain the fuzzy state result of the transformer, perform defuzzification on the fuzzy state result, and select the fuzzy set with the largest membership as the health evaluation result of the transformer.
[0015] According to the scheme provided by the present invention, characteristic quantities characterizing the health status of the transformer are collected according to a preset period, wherein the characteristic quantities include at least one of the following: temperature, voltage, current, pressure, noise, displacement, oil level and furfural concentration; the health index evaluation values corresponding to the transformer in different periods are constructed according to the characteristic quantities, and the priority relationship matrix of the health index evaluation values is calculated by the hierarchical analysis method; the priority relationship matrix is selected as the input quantity of fuzzy logic reasoning, the membership function and fuzzy rules are designed, and the input quantity is fuzzy reasoned according to the fuzzy rules to obtain the fuzzy result of the state of the transformer, the fuzzy result of the state is defuzzified, and the fuzzy set with the largest membership is selected as the health evaluation result of the transformer. The present invention evaluates the health of the transformer through the fuzzy logic method, uses fuzzy sets and fuzzy systems to process some uncertain characteristic quantities, and further improves the accuracy of the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings: Figure 1 A schematic diagram of a flow chart of a transformer health assessment method based on fuzzy logic according to an embodiment of the present invention is shown; Figure 2 A structural schematic diagram of a transformer health assessment system based on fuzzy logic according to an embodiment of the present invention is shown. DETAILED DESCRIPTION
[0017] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.
[0018] Figure 1 A schematic flow chart of a transformer health assessment method based on fuzzy logic according to an embodiment of the present invention is shown. Figure 2 The structure diagram of the transformer health assessment system based on fuzzy logic of an embodiment of the present invention is shown, including an acquisition module 210, a processing module 220, and an assessment module 230. The method evaluates the health of the transformer by using a fuzzy logic method, uses fuzzy sets and fuzzy systems to process some uncertain feature quantities, and further improves the accuracy of the assessment result. Specifically, the following steps are included: Step S101 : collecting characteristic quantities representing the health status of the transformer according to a preset period.
[0019] The reasons for transformer failure or damage are relatively complex. First, the temperature of the internal components of the transformer is too high, which causes the insulation material's withstand voltage and mechanical strength to decrease. For example, winding overheating in the primary coil or secondary coil, iron core overheating, and oil overheating. Secondly, when some components of the transformer are loose or the transformer has ferromagnetic resonance, vibration will occur inside the transformer and different degrees of noise will occur. Thirdly, due to oil leakage in the transformer, the oil level in the transformer is generally within the specified range. When the transformer leaks oil, the oil level in the transformer is low, causing the transformer's internal bushing and decomposition switch to be exposed to the air, resulting in a decrease in its insulation level, which is prone to breakdown discharge and accidents. In addition, when the operating current or operating voltage of the transformer exceeds the rated value, it will also cause damage to the transformer.
[0020] In this embodiment, the working condition information of each transformer is collected according to a preset period (such as every 10 seconds), wherein the characteristic quantity includes at least one of the following: temperature, voltage, current, pressure, noise, displacement, oil level and furfural concentration. Various transformer working condition detection devices are used to monitor the working condition of the transformer in real time, for example, the temperature sensor is used to detect the core temperature in the transformer, the oil level sensor is used to detect the oil level in the transformer, the noise sensor is used to detect the noise intensity generated in the transformer, the current sensor is used to detect the working current of the transformer, and the voltage sensor is used to detect the working voltage of the transformer. According to the above working state information of the transformer obtained, the transformer is analyzed for the winding overheating fault, core overheating fault, oil temperature overheating, low oil level, overload fault, overvoltage fault, loose device fault and ferromagnetic resonance fault.
[0021] Furthermore, the temperature includes the inner winding temperature, the core temperature and the oil temperature of the transformer.
[0022] For example, a temperature sensor is used to simultaneously detect the winding temperature, core temperature, and oil temperature of the transformer. Generally, when the internal winding temperature is greater than 50°C to 60°C, the core temperature is greater than 70°C to 80°C, or the oil temperature is greater than 50°C to 60°C, the transformer has a winding overheating fault, a core overheating fault, or an oil overheating fault.
[0023] Step S102: construct health index evaluation values corresponding to the transformer in different cycles according to the characteristic quantities, and calculate a priority relationship matrix of the health index evaluation values by using a hierarchical analysis method.
[0024] When a transformer fails, it may be a single cycle failure or a cycle offset. Therefore, in this example, based on the characteristics of periodic changes, the health index evaluation values corresponding to the transformer in different cycles are constructed according to the characteristic quantities, and the priority relationship matrix of the health index evaluation values is calculated by the hierarchical analysis method. The hierarchical analysis method combines qualitative and quantitative methods, increases the effectiveness of decision-making, and is simple to calculate, and the results are intuitive and clear.
[0025] Furthermore, the priority relationship matrix is constructed as follows: ; in, is the priority relation matrix, is the scale of the i-th index and the j-th index, (1,2,...,m).
[0026] The meaning of the scale is shown in Table 1, reflecting the importance of the evaluation factors.
[0027] Table 1: Matrix scale meaning table Scale meaning 1 Both indicators have equal importance 3 The former indicator is slightly more important than the latter indicator 5 The former indicator is obviously more important than the latter indicator 7 The former indicator is more important than the latter indicator 9 The previous indicator is more important than the later indicator
[0028] Furthermore, the specific formula for constructing the health index evaluation value of the transformer corresponding to different cycles according to the characteristic quantity is: ; in, is the health index evaluation value, is the i-th feature quantity, for The weight of , m is the number of features, e is a constant, e=2.718.
[0029] For example, when m=3, assuming is the oil breakdown voltage of the transformer, is the furfural concentration of transformer oil, is the hot spot temperature of the transformer winding, and the transformer health index evaluation model is: ; Among them, oil breakdown voltage is an important indicator for assessing the electrical strength of insulating oil. It is a measure of the ability of insulating oil to withstand voltage inside the transformer. The concentration of oil furfural can accurately reflect the aging degree of the transformer insulation paper. The remaining life of the transformer can be accurately predicted based on the aging degree of the insulation paper. The hot spot temperature of the winding is exponentially related to the aging degree of the insulation material. The potential failure risk of the transformer can be accurately detected based on the aging degree of the insulation material.
[0030] Furthermore, the weight The calculation formula is: ; in, , represented as an intermediate calculated value, for The weight of m is the number of features. is the scale of the ith index and the jth index, for example, the weight vector .
[0031] In this example, by scaling Calculate the weight vector Compared with expert experience, it is more objective, reduces some uncertainties or subjective factors, and further improves the accuracy of the evaluation results.
[0032] Step S103, select the priority relationship matrix as the input of fuzzy logic reasoning, design the membership function and fuzzy rules, perform fuzzy reasoning on the input according to the fuzzy rules to obtain the fuzzy result of the state of the transformer, defuzzify the fuzzy result of the state, and select the fuzzy set with the largest membership as the health assessment result of the transformer.
[0033] The health of the transformer is evaluated by fuzzy logic method, and fuzzy sets and fuzzy systems are used to deal with some uncertain characteristic quantities, which further improves the accuracy of the evaluation results.
[0034] Furthermore, the membership function includes a triangle membership function and a quadrilateral membership function, and the health assessment results include safe, dangerous and serious.
[0035] Membership is an effective multi-factor decision-making method, which is characterized by the fact that the evaluation result is not an absolute affirmation or negation, but is represented by a fuzzy set. At present, there is no mature and effective method for establishing the membership function. For the same fuzzy concept, different systems will establish different membership functions. In this embodiment, the membership function includes a triangular membership function and a quadrilateral membership function. Among them, the shape of the triangular membership function is similar to a triangle, and its values at both ends of the evaluation level are lower, and the values in the middle are higher, showing a trend of gradually increasing and then gradually decreasing. The quadrilateral membership function can be regarded as an arbitrary membership function. The health assessment results include safe, dangerous and serious. Optionally, the health assessment results may also include healthy, early warning, etc.
[0036] Furthermore, the fuzzy rules are specifically: When conducting health assessment, each characteristic quantity corresponds to a fuzzy logic reasoning rule; The basis for constructing each fuzzy logic reasoning rule is as follows: the health state corresponding to a smaller feature quantity is safe, the health state corresponding to a medium feature quantity is dangerous, and the health state corresponding to a larger feature quantity is serious; Among them, fuzzy inference rules are generally defined as: Rule j :IF x1 is a j1 …and x n is a jn； THEN y j =b ji ,j=1,2,…,N; i=1,2,…,n; Among them, x i (i=1, 2, ..., n) is the input quantity, y j For the output result, a ji and b ji is a fuzzy subset, and N is the total number of rules.
[0037] Furthermore, the membership function The initial expression of is: ; in, is a constant, E∈[0,1] , which controls the amplitude of the membership function, t is the value range of the membership function, and b is the translation amount.
[0038] In this embodiment, the fuzzy membership function Use a triangular membership function or a bell-shaped membership function.
[0039] Furthermore, the defuzzification of the state fuzzy result further includes: The state fuzzy result is defuzzified using the centroid method, and the specific formula of the centroid method is: ; ; in, is the x-coordinate of the center of gravity, is the y coordinate of the center of gravity, is the x-coordinate of the i-th fuzzy rule, is the y coordinate of the i-th fuzzy rule, The triangle membership function or quadrilateral membership function for the fuzzification input.
[0040] Defuzzification is an important step in the fuzzy inference engine, also known as defuzzification. There are many methods of defuzzification, the most commonly used ones are the maximum membership method, the center of gravity method and the weighted average method. In this embodiment, the center of gravity method is used to defuzzify the state fuzzy result. In order to obtain accurate control quantity, the fuzzy method is required to be able to express the calculation result of the output membership function well. The center of gravity method takes the center of gravity of the area enclosed by the membership function curve and the horizontal coordinate as the final output value of the fuzzy reasoning. Compared with the maximum membership method, the center of gravity method has a smoother output reasoning control. Even in response to a slight change in the input signal, the output will change.
[0041] According to the scheme provided by the present invention, characteristic quantities characterizing the health status of the transformer are collected according to a preset period, wherein the characteristic quantities include at least one of the following: temperature, voltage, current, pressure, noise, displacement, oil level and furfural concentration; the health index evaluation values corresponding to the transformer in different periods are constructed according to the characteristic quantities, and the priority relationship matrix of the health index evaluation values is calculated by the hierarchical analysis method; the priority relationship matrix is selected as the input quantity of fuzzy logic reasoning, the membership function and fuzzy rules are designed, and the input quantity is fuzzy reasoned according to the fuzzy rules to obtain the fuzzy result of the state of the transformer, the fuzzy result of the state is defuzzified, and the fuzzy set with the largest membership is selected as the health evaluation result of the transformer. The present invention evaluates the health of the transformer through the fuzzy logic method, uses fuzzy sets and fuzzy systems to process some uncertain characteristic quantities, and further improves the accuracy of the evaluation results.
[0042] The present invention also provides a transformer health assessment system based on fuzzy logic, comprising: The acquisition module 210 is used to acquire characteristic quantities representing the health status of the transformer according to a preset period, wherein the characteristic quantities include at least one of the following: temperature, voltage, current, pressure, noise, displacement, oil level, and furfural concentration; A processing module 220 is used to construct health index evaluation values corresponding to the transformer in different cycles according to the characteristic quantity, and calculate a priority relationship matrix of the health index evaluation values by using a hierarchical analysis method; The evaluation module 230 is used to select the priority relationship matrix as the input of fuzzy logic reasoning, design the membership function and fuzzy rules, perform fuzzy reasoning on the input according to the fuzzy rules to obtain the fuzzy result of the state of the transformer, defuzzify the fuzzy result of the state, and select the fuzzy set with the largest membership as the health evaluation result of the transformer.
Claims
1. A transformer health assessment method based on fuzzy logic, characterized in that: include: Collecting characteristic quantities characterizing the health status of the transformer according to a preset period, wherein the characteristic quantities include at least one of the following: temperature, voltage, current, pressure, noise, displacement, oil level, and furfural concentration; Constructing health index evaluation values of the transformer corresponding to different cycles according to the characteristic quantities, and calculating a priority relationship matrix of the health index evaluation values by using a hierarchical analysis method; The priority relationship matrix is selected as the input of fuzzy logic reasoning, and the membership function and fuzzy rules are designed. Fuzzy reasoning is performed on the input according to the fuzzy rules to obtain the fuzzy state result of the transformer, the fuzzy state result is defuzzified, and the fuzzy set with the largest membership is selected as the health assessment result of the transformer.
2. The transformer health assessment method based on fuzzy logic according to claim 1 is characterized in that: The temperatures include inner winding temperature, core temperature and oil temperature of the transformer.
3. The transformer health assessment method based on fuzzy logic according to claim 1 or 2, characterized in that: The specific formula for constructing the health index evaluation value of the transformer corresponding to different cycles according to the characteristic quantity is: ； in, is the health index evaluation value, is the i-th feature quantity, (1,2,...,m), for The weight of m is the number of features, e is a constant, e=2.718, It is the sum of the products of all features and their weights.
4. The transformer health assessment method based on fuzzy logic according to claim 3 is characterized in that: The priority relationship matrix is constructed as follows: ； in, is the priority relation matrix, is the scale of the i-th index and the j-th index, (1,2,...,m).
5. The transformer health assessment method based on fuzzy logic according to claim 4 is characterized in that: The weight The calculation formula is: ； in, , represented as an intermediate calculated value, for The weight of m is the number of features. is the scale of the i-th indicator and the j-th indicator.
6. The transformer health assessment method based on fuzzy logic according to claim 1 is characterized in that: The membership function includes a triangle membership function and a quadrilateral membership function, and the health evaluation results include safe, dangerous and serious.
7. The transformer health assessment method based on fuzzy logic according to claim 6 is characterized in that: The fuzzy rules are specifically: When conducting health assessment, each characteristic quantity corresponds to a fuzzy logic reasoning rule; The basis for constructing each fuzzy logic reasoning rule is specifically as follows: the health state corresponding to the feature value being less than or equal to the first preset threshold is safe, the health state corresponding to the feature value being greater than the first preset threshold and less than or equal to the second preset threshold is dangerous, and the health state corresponding to the feature value being greater than the second preset threshold is serious.
8. The transformer health assessment method based on fuzzy logic according to claim 7 is characterized in that: The initial expression of the membership function is: ； in, is a constant, , which controls the amplitude of the membership function, t is the value range of the membership function, and b is the translation amount.
9. The transformer health assessment method based on fuzzy logic according to claim 8 is characterized in that: The defuzzification of the state fuzzy result further comprises: The state fuzzy result is defuzzified using the centroid method, and the specific formula of the centroid method is: ； ； in, is the x-coordinate of the center of gravity, is the y coordinate of the center of gravity, is the x-coordinate of the i-th fuzzy rule, is the y coordinate of the i-th fuzzy rule, The triangle membership function or quadrilateral membership function for the fuzzification input.
10. A transformer health assessment system based on fuzzy logic, the system is based on the transformer health assessment method based on fuzzy logic according to any one of claims 1 to 9, characterized in that: include: A collection module, used for collecting characteristic quantities representing the health status of the transformer according to a preset period, wherein the characteristic quantities include at least one of the following: temperature, voltage, current, pressure, noise, displacement, oil level and furfural concentration; A processing module, used to construct health index evaluation values corresponding to the transformer in different cycles according to the characteristic quantity, and calculate a priority relationship matrix of the health index evaluation values by using a hierarchical analysis method; An evaluation module is used to select the priority relationship matrix as the input of fuzzy logic reasoning, design membership functions and fuzzy rules, perform fuzzy reasoning on the input according to the fuzzy rules to obtain the fuzzy state result of the transformer, perform defuzzification on the fuzzy state result, and select the fuzzy set with the largest membership as the health evaluation result of the transformer.