Distribution network fault diagnosis method, system and device based on fuzzy theory and HHT and storage medium

Abstract

The invention discloses a fuzzy theory and HHT-based distribution network fault diagnosis method, system and device, and a storage medium, and relates to the technical field of power system fault diagnosis, and the method comprises the steps: obtaining the motion information of a protection switch, and calculating the switching value fault degree of an element; acquiring fault voltage and current information, and calculating a current fault energy value; based on the element switching value fault degree and the current fault energy value, data fusion is carried out by establishing a multi-source information fusion model; based on the fused data, using a fuzzy C-means clustering method to establish a decision model and performing power grid fault diagnosis; the method solves the problem of fault diagnosis of the power distribution network in the high-altitude complex mountainous area, integrates multi-source information, realizes omnibearing and multi-dimensional diagnosis, and can adapt to severe natural conditions and environments with insufficient communication. And by adopting a distributed architecture and a modular design, the real-time performance, the robustness and the expansibility are enhanced, and the safety and the stability of the power distribution network are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system fault diagnosis, and in particular to a power distribution network fault diagnosis method, system and device based on fuzzy theory and HHT and a storage medium. BACKGROUND

[0002] Fault diagnosis of power distribution network in high-altitude complex mountainous area is an important research direction of power grid operation and maintenance. The environment in mountainous area is complex, the natural conditions are severe, the terrain is complex and the communication network coverage is insufficient, resulting in fast equipment aging, frequent faults and difficult rapid positioning.

[0003] The current mainstream method adopts intelligent fault diagnosis based on a single signal, and uses machine learning or deep learning algorithm to model a single signal (such as current or voltage signal) to identify fault types. For example, support vector machine can classify fault signal features, and neural network can identify nonlinear fault features. However, these methods usually rely on a single data source (electrical signal or environmental data), and it is difficult to fully reflect complex fault characteristics and causes. Moreover, it is good at identifying single fault modes (such as short circuit and grounding), but weak in identifying complex modes such as implicit faults and multiple fault superposition, and cannot meet the diagnosis requirements of high real-time and high accuracy. SUMMARY

[0004] In view of the above problems, the present application is proposed.

[0005] Therefore, the technical problem solved by the present application is that the existing power distribution network fault diagnosis technology has the problems of single data source, poor environmental adaptability, weak complex fault mode recognition ability, and low real-time and calculation efficiency, and realizes all-around, multi-dimensional perception and diagnosis of power distribution network faults, and automatic extraction and efficient classification of complex fault characteristics.

[0006] To solve the above technical problems, the present application provides the following technical scheme:

[0007] In a first aspect, the present application provides a power distribution network fault diagnosis method based on fuzzy theory and HHT, comprising:

[0008] Obtaining the action information of the protection switch, calculating the element switch quantity fault degree;

[0009] Obtaining the fault voltage and current information, calculating the current fault energy value;

[0010] Based on the element switch quantity fault degree and the current fault energy value, data fusion is performed through the establishment of a multi-source information fusion model;

[0011] Based on the fused data, a decision model is established using fuzzy C-means clustering method and power grid fault diagnosis is performed.

[0012] As an optimal scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT, wherein:

[0013] The action information of the protection switch is acquired, and the element switch quantity fault degree is calculated.

[0014] The action information of the protection switch is analyzed by using the Petri net model, the input matrix, the output matrix, the initial probability value of the place and the transition threshold value are added on the basis of the place set and the transition set of the basic Petri net.

[0015] According to the token value of the place in the initial state, it is judged whether the transition meets the triggering condition, the input matrix row vector corresponding to the target transition is extracted, the input strength is calculated in different ways according to the structure type of the Petri net, and the input strength is compared with the transition threshold value, if the input strength is greater than or equal to the transition threshold value, the transition meets the triggering condition.

[0016] As an optimal scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT, wherein:

[0017] The action information of the protection switch is acquired, and the element switch quantity fault degree is calculated.

[0018] When the transition meets the triggering condition, the transition triggering is performed to obtain the new place token value, and the token value is updated in a corresponding way according to different input structure types.

[0019] The transition triggering condition judgment and the token value updating operation are repeated until the network reaches a stable state, that is, all transitions do not meet the triggering condition, at this time, the token value of each place is the switch quantity fault degree of each element.

[0020] The beneficial effects of the preferred technical scheme are that the transition triggering and the token value updating are performed according to specific rules until the network reaches a stable state, so that the switch quantity fault degree of each element is obtained. This iterative updating method can fully consider the complex logical relationship between the protection switch action information, so that the calculated switch quantity fault degree can more accurately reflect the actual fault possibility of the element, and the accuracy of the fault diagnosis is improved.

[0021] As an optimal scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT, wherein:

[0022] The action information of the protection switch is acquired, and the element switch quantity fault degree is calculated.

[0023] The electrical quantity real-time recording signal after the fault of the distribution network is analyzed, the signal is subjected to Hilbert transform, and the signal instantaneous frequency is defined.

[0024] The signal is subjected to empirical mode decomposition, and the signal is decomposed into a plurality of IMF components and a residual signal.

[0025] The Hilbert transform result of the IMF component is summarized to obtain the Hilbert spectrum of the original signal, the marginal spectrum is obtained by integrating the Hilbert spectrum in the time domain, and the current fault energy value of each line is calculated based on the marginal spectrum.

[0026] As an optimal scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT, wherein:

[0027] The data fusion based on the element switching quantity fault degree and the current fault energy value comprises:

[0028] The evidence source focal element is determined based on the element switching quantity fault degree and the current fault energy value, threshold comparison is performed on the basic probability assignment of each evidence source focal element, a fault table with binary attributes is constructed, and the focal element conflict coefficient of each proposition in the fault table and the conflict distance between the evidence sources are calculated.

[0029] The beneficial effects of the preferred technical scheme are that: by determining the evidence source focal element, constructing the fault table, and calculating the focal element conflict coefficient and the conflict distance between the evidence sources, the focal element conflict in the same proposition and the conflict degree between the evidence sources can be intuitively judged, which provides a clear conflict analysis basis for subsequent reasonable classification and fusion of the evidence sources, and helps to improve the accuracy and reliability of multi-source information fusion.

[0030] As an optimal scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT, wherein:

[0031] The data fusion based on the element switching quantity fault degree and the current fault energy value by establishing a multi-source information fusion model further comprises:

[0032] The evidence virtual alliance is formed by classifying the evidence sources according to the threshold value of the conflict distance: when the conflict distance does not satisfy a specific threshold value, there is no strong conflict between the evidence sources and classification is not needed; when the conflict distance satisfies a specific threshold value, the minimum value of the non-diagonal elements in the distance matrix is found, a virtual alliance is formed by the corresponding evidence sources and replaces the original evidence sources, the distance between the virtual alliance and other evidence sources is calculated, and the distance matrix is updated.

[0033] The beneficial effects of the preferred technical scheme are that: according to the threshold value of the conflict distance, the evidence sources are classified to form the evidence virtual alliance, which can reasonably divide the evidence sources with different conflict degrees, avoids the adverse effects of the conflict between the evidence sources on the fusion result, makes the evidence sources with similar characteristics form an alliance, creates favorable conditions for more effective information fusion, and improves the stability and accuracy of fault diagnosis.

[0034] As an optimal scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT, wherein:

[0035] The multi-source information fusion module is used for data fusion based on the element switching quantity fault degree and the current fault energy value by establishing a multi-source information fusion model.

[0036] The credibility of each evidence source is determined as the weight of the evidence source, the weight of the virtual alliance is determined according to the credibility of each evidence source in the alliance, the improved D-S evidence theory weighted fusion rule is used to fuse each evidence source based on the weight of the virtual alliance, the evidence sources in each alliance are fused to obtain the alliance fusion result, and the fusion results of the alliances are weighted and fused to obtain the final multi-source information fusion result.

[0037] In a second aspect, the present application provides a distribution network fault diagnosis system based on fuzzy theory and HHT, comprising:

[0038] The switching quantity fault degree calculation module is used for acquiring the action information of the protection switch and calculating the element switching quantity fault degree.

[0039] The current fault energy value calculation module is used for acquiring the fault voltage and current information and calculating the current fault energy value.

[0040] The multi-source information fusion module is used for data fusion based on the element switching quantity fault degree and the current fault energy value by establishing a multi-source information fusion model.

[0041] The power grid fault diagnosis decision module is used for establishing a decision model by using the fuzzy C-means clustering method based on the fused data and performing power grid fault diagnosis.

[0042] In a third aspect, the present application provides an electronic device, comprising:

[0043] A memory and a processor.

[0044] The memory is used for storing computer executable instructions, and the processor is used for executing the computer executable instructions, so that the one or more processors implement the distribution network fault diagnosis method based on fuzzy theory and HHT as described in the present application.

[0045] In a fourth aspect, the present application provides a computer readable storage medium storing computer executable instructions, which are executed by a processor to implement the distribution network fault diagnosis method based on fuzzy theory and HHT.

[0046] The application has the advantages that the power distribution network fault diagnosis method based on fuzzy theory and HHT multi-source information fusion has obvious practical application value. The power distribution network in high-altitude complex mountainous areas faces complex natural conditions and terrain restrictions, and the traditional single data source fault diagnosis method is difficult to adapt. The application can perceive and diagnose the power distribution network fault in all directions and multiple dimensions by fusing electrical signals, switch information and other multi-source information. This makes it possible to obtain fault information more comprehensively and improve the accuracy of fault diagnosis in low temperature, strong wind, ice and snow and other harsh natural conditions and in mountainous areas with insufficient communication network coverage. The application improves the perception ability of the system to hidden faults and composite faults for complex fault modes. In the mountainous power distribution network, the fault forms are various and complex, and the hidden faults and multiple faults are common. The traditional method has weak recognition ability for such faults. The application can quickly and accurately locate and identify these complex faults by automatically extracting complex fault features and efficiently classifying them, thereby ensuring the safe and stable operation of the power distribution network. In addition, the application uses a distributed architecture and edge computing to improve the real-time diagnosis and uses modular design to enhance the robustness and scalability of the system. In practical application, the application can respond to faults more timely and adapt to power distribution networks of different scales and needs, thereby providing strong support for improving the operation efficiency and safety of the power grid. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can be obtained without creative labor under the premise of the drawings.

[0048] Figure 1 is the overall flowchart of the power distribution network fault diagnosis method based on fuzzy theory and HHT provided by the application. DETAILED DESCRIPTION

[0049] In order to make the above-mentioned purposes, features and advantages of the application more apparent and easy to understand, the specific embodiments of the application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor should be within the protection scope of the application.

[0050] Embodiment 1, refer to Figure 1 The first embodiment of the application provides a power distribution network fault diagnosis method based on fuzzy theory and HHT, which comprises:

[0051] S1: Obtain the action information of the protection switch, and calculate the component switch quantity fault degree;

[0052] S2: Obtain the fault voltage and current information, and calculate the current fault energy value;

[0053] S3: Based on the component switch quantity fault degree and the current fault energy value, data fusion is performed through the establishment of a multi-source information fusion model;

[0054] S4: Based on the fused data, a decision model is established using the fuzzy C-means clustering method and power grid fault diagnosis is performed.

[0055] It should be noted that through steps S1-S4, the protection switch action information and the fault voltage and current information are fully integrated, the component switch quantity fault degree and the current fault energy value are calculated respectively, different types of data are effectively fused using the multi-source information fusion model, and finally the decision model is established with the aid of the fuzzy C-means clustering method. The complementarity between multi-source information can be effectively utilized, the accuracy and reliability of the distribution network fault diagnosis are significantly improved, and strong technical support is provided for fast positioning and processing of distribution network faults and safe and stable operation of the power system.

[0056] Embodiment 2, refer to Figure 1 For an embodiment of the present application, a distribution network fault diagnosis method based on fuzzy theory and HHT is provided based on the previous embodiment, comprising:

[0057] In this embodiment, the step S1 of obtaining the action information of the protection switch and calculating the component switch quantity fault degree comprises:

[0058] The action information of the protection switch is analyzed using a Petri net (PN) model to obtain the component switch quantity fault degree;

[0059] It should be noted that the basic Petri net is a directed graph composed of places, transitions, token values and directed arcs, and has dynamic properties such as boundedness, safety, liveness and reachability. The place represents the possible local state of the system, the transition represents the event that modifies the local state of the system, the token represents the resource status of the system, and the directed arc represents the connection relationship between the local state and the event. The fuzzy Petri net is obtained by expanding the fuzzy processing capability based on the basic Petri net, and the token value in the place node is represented by a fuzzy number with a probability interval [0, 1], and the condition for starting the transition node.

[0060] Specifically, the Petri net is defined as a triple, i.e. PN=(P, T, C), then:

[0061]

[0062] Where P represents the state nodes (such as switches and lines) in the source data represented by the finite library set, and element p i For a location; T is a finite transition set representing the logical rules of state changes in the source data (such as protection actions triggering fault propagation), and the element Tt j For transitions; elements in C are arcs representing causal relationships between source data (such as changes in switch states affecting the probability of line faults), and are p in the PN library. i With change t j The flow relationship between them; "×" represents the Cartesian product; This means that there is no isolated relationship.

[0063] Extending the triplet structure of the PN to a six-tuple structure yields the fuzzy Petri net (FPN), represented as:

[0064] FPN={P,T,I,O,α,T λ}

[0065] Where P and T are defined the same as in a basic Petri net; I is the input matrix, reflecting the mapping from the input place to the transition; O is the output matrix, representing the mapping weights from the transition to the output place; α is the initial probability value (Token value) of the place, α(p i )∈[0,1];T λ T is the transition threshold. λ ∈[0,1].

[0066] Based on the initial place token values, determine whether the transition satisfies the activation condition: In the initial state, the place set P = {P1, P2, ..., P} n The token value of} is α = [α1, α2, ..., α]. n For the target transition T, extract the corresponding row vector I of the input matrix. j =[I 1j ,I 2j ,…,I nj ], I ij >0 indicates that the place P is a library. i It is T j The input is calculated using different methods depending on the Petri net's structure type (simple, AND, OR): if transition T j only The given conditions are sufficient; the input strength is the maximum weighted value. Compared with the transition threshold, if the input intensity S j ≥T λ Then the transition T j The activation conditions are met;

[0067] If the transition satisfies the excitation condition, then the transition is excited to obtain a new Coustoken value: if the transition T jOnly by a single library P i Input (I ij =1), the input intensity is S j =α i And S j ≥T λ , the new token value is

[0068] α new =S j ·O jk ;

[0069] Repeat the series of operations "determine whether the transition meets the firing condition according to the initial state of the library token value, if it meets the transition firing to obtain the new library token value", until the network reaches a stable state; When all transitions do not meet the firing condition (S < T λ ), the switch quantity fault degree of each element can be obtained, and at this time the token value of each library is the switch quantity fault degree of the element.

[0070] In the embodiment, the step S2 of obtaining the fault voltage and current information and calculating the current fault energy value includes:

[0071] The HHT (Hilbert-Huang Transform) is used to analyze the fault voltage and current to obtain the current and energy fault indicators.

[0072] Specifically, the current, voltage and other electrical quantities real-time recording signals after the fault of the power distribution network are analyzed, the Hilbert transform is performed, and the signal instantaneous frequency is defined.

[0073]

[0074] The empirical mode decomposition is performed, and after the EMD decomposition of the signal x(t), it is expressed as:

[0075]

[0076] Wherein, c i (t) represents the i-th IMF component, and r n (t) represents the residual signal after decomposition.

[0077] The Hilbert transform results of the IMF components are summarized to obtain the Hilbert spectrum of the original signal, which is expressed as:

[0078]

[0079] Wherein, H(ω,t) represents the Hilbert spectrum of the original signal, which describes the change of signal amplitude with time and frequency on the frequency axis.

[0080] The marginal spectrum h(ω) is obtained by integrating the Hilbert spectrum in the time domain, and it reflects the amplitude (energy) of the signal in the frequency axis with the frequency, and the formula is as follows:

[0081]

[0082] The marginal spectrum reflects the amplitude (energy) of the signal in the frequency axis with the frequency, and since the instantaneous frequency defined at this point is a function of time, the existence of energy at a certain frequency value in the marginal spectrum represents that the wave of the frequency component must exist in the original signal.

[0083] According to the Hilbert marginal energy spectrum of the fault line current signal, the fault line is identified.

[0084] Further, after the power grid fails, the transient component is generated by the fault current flowing through the fault line, so that the fault line current signal has strong energy, and the current fault energy value is defined to represent the total energy of the line current after the fault. The HHT transformation is performed on the current signals of the fault line and the non-fault line.

[0085] After the power grid fails, the total energy of the current signal of the fault line is greater than that of the non-fault line, and according to this feature, the fault line can be identified.

[0086] After the fault occurs, the current signal of the i-th (i=0, 1, …, 2) line is x i (t), and the corresponding Hilbert spectrum H(ω, t) is obtained after the Hilbert-Huang transformation is performed, and the Hilbert marginal energy spectrum S(ω) of the current of each line within two cycles after the fault is calculated.

[0087] The current fault energy value e i of the i-th line is calculated, which represents the sum of the energy of the signal within two cycles after the fault in all frequency bands, and is represented as:

[0088]

[0089] In another possible implementation, in the process of identifying the fault line according to the Hilbert marginal energy spectrum of the fault line current signal, the threshold comparison method can also be used. Specifically, according to the Hilbert marginal energy spectrum of each line during normal operation, the energy distribution range at each characteristic frequency is statistically obtained, and the energy threshold is set based on this. The Hilbert marginal energy spectrum of each line is compared with the set threshold. If the energy of a certain line at the characteristic frequency exceeds the threshold, it is considered that the line may have failed.

[0090] In the embodiment, the data fusion based on the element switching value fault degree and the current fault energy value in step S3 is performed by establishing a multi-source information fusion model, which includes:

[0091] The data fusion is performed by establishing a multi-source information fusion model based on the element switching value fault degree and the current fault energy value through improving the D-S (Dempster-Shafer Evidence Theory) evidence theory;

[0092] Specifically, the switching value data is calculated to obtain the switching value fault degree of the element by the fuzzy Petri net, and the corresponding focal elements are "element i fault" and "element i non-fault".

[0093] The electrical quantity data is analyzed by the HHT to obtain the current fault energy value and the marginal energy spectrum, and the corresponding focal elements are "line j exists fault energy anomaly"; the protection device action signal is converted into the focal elements "protection k correct action" and "protection k misoperation".

[0094] The basic probability assignment of each evidence source focal element is compared with a threshold value to obtain a fault table FT with a binary attribute, and the conflict of the focal elements in the same proposition can be directly judged.

[0095] The conflict coefficients of the focal elements in each proposition in the fault table and the conflict distances J between the evidence sources are calculated, reflecting the conflict degree between the evidence sources.

[0096] The conflict distances are compared with a threshold value, and each evidence source is classified to form an evidence virtual alliance.

[0097] In another possible implementation, in the process of comparing the conflict distances with a threshold value and classifying each evidence source to form an evidence virtual alliance, the conflict distance threshold value J a is set, and then it is judged whether the non-diagonal elements (i.e. i≠j) in the conflict distance matrix J satisfy the classification condition, and then the multiple evidence sources are classified to form a virtual alliance. If max(J ij )≤J α , it indicates that there is no strong conflict between the evidences, and no classification and division are needed; if max(J ij )>J α , further division is needed. The minimum value J ij (i≠j) of the non-diagonal elements in the distance matrix J is found, a virtual alliance G is formed by the evidence source i and the evidence source j, and i and j are replaced; the maximum value of the distance between the evidence source i and the evidence source j and other evidences is calculated to form the distance between the virtual alliance G and other evidences, i.e. J Gf =max(J if ,J jf), update the distance matrix J; if the minimum value of the non-diagonal elements of the updated distance matrix J is not greater than the threshold value J ij )>J α , the operation of forming a virtual alliance and updating the distance matrix is repeatedly performed until min(J g , which indicates that the distance between each alliance meets the classification requirements, and the classification of the evidence sources is completed. The h evidence sources that meet the conditions are divided into g virtual alliances G1, G2,..., G k , where any alliance G k is composed of e evidence sources, that is:

[0098] G e}

[0099] The credibility Crd(i) of the evidence source i represents the weight of the evidence source i, and the weight of the virtual alliance depends on the credibility Crd(i) of each evidence source in the alliance. Therefore, the weight W k of the virtual alliance G k is:

[0100] W k =Crd(1)+Crd(2)+…+Crd(e)

[0101] The improved weighted fusion rule of D-S evidence theory based on virtual alliance can be obtained as:

[0102]

[0103] Wherein, (1) represents the fusion of the e evidence sources in the formed alliance G k , and the fusion result of the alliance is obtained, (2) represents the weighted fusion of the g alliances, and the final result is obtained.

[0104] The credibility Crd of the evidence sources in the virtual alliance G k is calculated, and the fusion weight W k of the virtual alliance G k is represented, and the improved weighted fusion of D-S evidence theory is performed to obtain the fusion result.

[0105] In this embodiment, the establishment of the decision model and the power grid fault diagnosis using the fuzzy C-means clustering method based on the fused data in the above step S4 include:

[0106] After the improved D-S evidence theory is used to synthesize the evidence body, the probability characteristics need to be analyzed to determine the accident device, and the fuzzy C-means algorithm is used to divide several fault probability expressions into two categories: fault and non-fault.

[0107] Specifically, let the failure probabilities of the n components be represented as m(F1), m(F2), ..., m(F... n All components are initially divided into two categories, and the Gamma function is selected. Calculate the Gamma function value for the failure probability characterization of each component. If the following conditions are met:

[0108]

[0109] Then, the device is considered as a fault candidate class Γ1, and the remaining devices are classified as non-fault candidate class Γ2, where ε = 12;

[0110] If the fault candidate class contains N1 devices, the corresponding Gamma function values ​​are E1, E2, ..., E N1 .

[0111] Let E satisfy E = min{E1, E2, ..., E} N1 The device in question is a device that is bound to fail.

[0112] Calculate the mean of the original categories (Γ1,Γ2) for all devices, and finally obtain the sum of squared errors:

[0113]

[0114] From Γ i Selecting sample m(F) j );

[0115] If Ni = 1, then repeat from Γ i Selecting sample m(F) j Otherwise, continue;

[0116] calculate:

[0117]

[0118] When ρ k ≤ρ i At that time, Γ i m(F) j ) transferred to Γ k In the middle, recalculate m i and m k The value of J, then for J e Reassign;

[0119] When J e If no change occurs after N consecutive iterations, the rotation ends; otherwise, the process repeats from Γ. i Selecting sample m(F) j ).

[0120] The preset threshold value ε needs to be adjusted according to power grid scales with different sizes, so as to meet the diagnosis requirements of power systems of various levels.

[0121] Embodiment 3, the above is a schematic scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT of the embodiment. It should be noted that the technical scheme of the distribution network fault diagnosis system based on fuzzy theory and HHT belongs to the same concept as the technical scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT. The technical details of the distribution network fault diagnosis system based on fuzzy theory and HHT in this embodiment are not described in detail, and can be referred to the description of the technical scheme of the distribution network fault diagnosis method based on fuzzy theory and HHT.

[0122] The embodiment also provides a distribution network fault diagnosis system based on fuzzy theory and HHT, comprising:

[0123] The switch quantity fault degree calculation module is configured to obtain the action information of the protection switch and calculate the element switch quantity fault degree.

[0124] The current fault energy value calculation module is configured to obtain the fault voltage and current information and calculate the current fault energy value.

[0125] The multi-source information fusion module is configured to perform data fusion by establishing a multi-source information fusion model based on the element switch quantity fault degree and the current fault energy value.

[0126] The power grid fault diagnosis decision module is configured to establish a decision model using the fuzzy C-means clustering method and perform power grid fault diagnosis based on the fused data.

[0127] The embodiment also provides an electronic device suitable for the distribution network fault diagnosis method based on fuzzy theory and HHT, comprising:

[0128] The memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the distribution network fault diagnosis method based on fuzzy theory and HHT as proposed in the above embodiment.

[0129] The embodiment also provides a storage medium having a computer program stored thereon, and the program is executed by a processor to implement the distribution network fault diagnosis method based on fuzzy theory and HHT as proposed in the above embodiment.

[0130] The storage medium proposed in the embodiment and the distribution network fault diagnosis method based on fuzzy theory and HHT proposed in the above embodiment belong to the same inventive concept, and the technical details not described in detail in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0131] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A distribution network fault diagnosis method based on fuzzy theory and HHT, characterized in that, include: Obtain the action information of the protective switch and calculate the fault degree of the component switching quantity; Obtain fault voltage and current information, and calculate the current fault energy value; Based on the fault degree of component switching quantity and the energy value of current fault, data fusion is carried out by establishing a multi-source information fusion model; Based on the fused data, a decision model is established using fuzzy C-means clustering and power grid fault diagnosis is performed.

2. The distribution network fault diagnosis method based on fuzzy theory and HHT as described in claim 1, characterized in that, The process of acquiring the action information of the protective switch and calculating the fault degree of the component switching quantity includes: The action information of the protection switch is analyzed using the Petri net model. Based on the basic Petri net's place set and transition set, the input matrix, output matrix, initial probability value of the place, and transition threshold are added. Based on the token value of the place in the initial state, it is determined whether the transition meets the excitation condition. For the target transition, the corresponding input matrix row vector is extracted. The input intensity is calculated in different ways according to the structure type of Petri net. The input intensity is compared with the transition threshold. If the input intensity is greater than or equal to the transition threshold, the transition meets the excitation condition.

3. The distribution network fault diagnosis method based on fuzzy theory and HHT as described in claim 2, characterized in that, The process of obtaining the action information of the protective switch and calculating the fault degree of the component switching quantity also includes: When a transition meets the excitation condition, the transition is excited to obtain a new Coulomb value. The Coulomb value is updated in an appropriate manner according to different input structure types. Repeat the transition trigger condition judgment and token value update operation until the network reaches a stable state, that is, all transitions no longer meet the trigger conditions. At this time, the token value of each place is the switching fault degree of each element.

4. The distribution network fault diagnosis method based on fuzzy theory and HHT as described in claim 3, characterized in that, The steps of acquiring fault voltage and current information and calculating the current fault energy value include: Analyze the real-time recorded electrical signals after a power distribution network fault, perform Hilbert transform on the signals, and define the instantaneous frequency of the signals; Empirical mode decomposition (EMD) is performed on the signal, which is decomposed into multiple IMF components and residual signal. The Hilbert transform results of the IMF components are summarized to obtain the Hilbert spectrum of the original signal. The marginal spectrum is obtained by integrating the Hilbert spectrum in the time domain. The current fault energy value of each line is calculated based on the marginal spectrum.

5. The distribution network fault diagnosis method based on fuzzy theory and HHT as described in claim 4, characterized in that, The data fusion based on component switching fault degree and current fault energy value, through the establishment of a multi-source information fusion model, includes: The focal elements of evidence sources are determined based on the fault degree of component switching quantity and the energy value of current fault. The basic probability assignment of each focal element of evidence sources is compared with the threshold to construct a fault table with binary attributes. Based on the fault table, the focal element conflict coefficient of each proposition in the fault table and the conflict distance between evidence sources are calculated respectively.

6. The distribution network fault diagnosis method based on fuzzy theory and HHT as described in claim 5, characterized in that, The data fusion based on component switching fault degree and current fault energy value, through the establishment of a multi-source information fusion model, also includes: A threshold comparison is performed on the conflict distances to classify the evidence sources into virtual alliances: when the conflict distances do not meet a specific threshold, there is no strong conflict between the evidence sources, and no classification is required; when the conflict distances meet a specific threshold, the minimum value of the off-diagonal elements in the distance matrix is ​​found, and the corresponding evidence sources form virtual alliances and replace the original evidence sources. The distances between the virtual alliances and other evidence sources are calculated, and the distance matrix is ​​updated.

7. The distribution network fault diagnosis method based on fuzzy theory and HHT as described in claim 6, characterized in that, The data fusion based on component switching fault degree and current fault energy value, through the establishment of a multi-source information fusion model, also includes: The credibility of each evidence source is determined and used as the weight of the evidence source. The weight of the virtual alliance is determined according to the credibility of each evidence source within the alliance. Based on the weight of the virtual alliance, the improved DS evidence theory weighted fusion rule is used to fuse each evidence source. First, the evidence sources within each alliance are fused to obtain the alliance fusion result. Then, the fusion results of each alliance are weighted and fused to obtain the final multi-source information fusion result.

8. A distribution network fault diagnosis system based on fuzzy theory and HHT, using the method described in any one of claims 1 to 7, characterized in that, include: The switch quantity fault degree calculation module is used to obtain the action information of the protection switch and calculate the switch quantity fault degree of the component. The current fault energy value calculation module is used to obtain fault voltage and current information and calculate the current fault energy value. The multi-source information fusion module is used to perform data fusion by establishing a multi-source information fusion model based on the fault degree of component switching quantity and the energy value of current fault. The power grid fault diagnosis decision module is used to establish a decision model and perform power grid fault diagnosis based on the fused data using the fuzzy C-means clustering method.

9. An electronic device, characterized in that, include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores computer-executable instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images