Multi-source power failure information fusion method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of power distribution network fault recognition, in particular to a multi-source power failure information fusion method and device, computer equipment and a storage medium, and the method comprises the steps: collecting power failure information, and extracting power failure feature information as an evidence source; calculating a probability distribution function value of each evidence source, and generating a fault table with binary attributes; calculating a conflict coefficient and a conflict distance between the same evidence sources through the fault table, and obtaining a conflict distance matrix between the evidence sources; whether off-diagonal elements in the conflict distance matrix meet classification conditions or not is judged through a set conflict distance threshold value, classification is carried out, and evidence source classification is formed; and fusing the evidence sources in the evidence source classification by using an improved D-S fusion rule to obtain a fusion result. According to the method, the conflict coefficient and the conflict distance between the evidences are measured, so that the accuracy of the fusion result is not influenced when the evidences conflict, and the power failure reason can be accurately identified.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network fault identification, and in particular to a multi-source power outage information fusion method, device, computer equipment and storage medium. Background Art

[0002] In modern power systems and smart grids, in order to ensure the stable and safe operation of the power grid, a variety of monitoring, sensing, communication and analysis equipment are usually deployed. When a power outage or other fault occurs in the power grid, these devices and systems will generate a large amount of alarm information and data. Since the information from these different sources may be different or conflicting, information fusion is needed to provide more accurate and comprehensive power outage or fault diagnosis. Multi-source information fusion is gaining more and more attention in modern power.

[0003] The basic principle of information fusion is to reasonably allocate and use the information provided by multiple information sources, and process the complementary and redundant information between multiple information sources according to specific rules to reduce the ambiguity of information, and obtain a fusion result with a consistent interpretation of the measured object, which can improve the fault tolerance and reliability of the system and make the fusion system more superior than its subsystems. In multi-sensor information systems, the advantages of using information fusion technology in solving problems such as detection, tracking and identification are mainly reflected in the ability to increase the survivability of the system, expand the time-space coverage, reduce the uncertainty of event results, and improve the correctness of system decisions. At present, a series of studies are mainly conducted on the fusion mode of multi-information, system structure, and information representation and conversion in the system. Information fusion mainly has serial, parallel and hybrid fusion structures, and each sensor forms an information acquisition unit. According to the level of data processing, it can be divided into data layer, feature layer, and decision layer information fusion.

[0004] At present, the main information fusion technology algorithms include: dividing the information related to the target into different sets and applying the information correlation quantitative analysis method, such as the nearest neighbor rule, optimal difference, joint statistical association, etc.; applying the estimation theory method to multi-sensor data, such as multi-scale estimation, maximum likelihood estimation, Kalman filtering, weighted least squares method, etc.; using recognition technology to match the observed data, such as expert system, neural network, template method, etc.; using probability interval and determination interval to determine the DS evidence reasoning algorithm of multivariate hypothesis likelihood; building on a solid mathematical foundation and using prior probability to find the probability of occurrence of the measured object; and applying generalized set theory to determine the number of events in a specified set, which can be used for fuzzy set theory for decision analysis of uncertain events, etc. Each algorithm has certain limitations and advantages. For example, the prior probability required by the Bayesian algorithm is often uncertain, and the complexity increases exponentially when there are multiple variables; the neural network requires a large number of training samples to ensure the accuracy of the results, and does not have a strong mathematical explanation; the expert system requires a complete expert library for various results, etc.

[0005] Dempster-Shafer theory (DS theory), also known as evidence theory, is a mathematical framework for dealing with uncertainty and reasoning. DS theory provides a method beyond traditional probability theory for synthesizing and analyzing uncertain evidence from multiple sources. It has applications in a variety of fields such as artificial intelligence, statistics, decision analysis, and engineering. The core concepts of DS theory include: basic probability assignment, belief function, likelihood function, and Dempster combination rule. A major advantage of DS theory is that it provides a flexible way to deal with uncertainty and ambiguous information, especially when the evidence is incomplete or comes from diverse sources. However, its computational complexity and handling of conflicting evidence are sometimes challenging.

[0006] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0007] The present invention provides a multi-source power outage information fusion method, device, computer equipment and storage medium, thereby effectively solving the problems in the background technology.

[0008] In order to achieve the above object, the technical solution adopted by the present invention is: a multi-source power outage information fusion method, comprising the following steps:

[0009] Collect power outage information and extract power outage feature information as a source of evidence;

[0010] Calculating the probability distribution function value of each of the evidence sources and generating a fault table with binary attributes;

[0011] Calculate the conflict coefficient and conflict distance between the same evidence sources through the fault table to obtain the conflict distance matrix between the evidence sources;

[0012] Determine whether the non-diagonal elements in the conflict distance matrix meet the classification conditions through the set conflict distance threshold, and classify them to form evidence source classification;

[0013] The improved DS fusion rule is used to fuse the evidence sources in the evidence source classification to obtain a fusion result.

[0014] Furthermore, before extracting the power outage feature information as a source of evidence, the collected information is also cleaned, standardized and time synchronized.

[0015] Furthermore, the method of calculating the probability distribution function value of each of the evidence sources and generating a fault table with binary attributes includes the following steps:

[0016] Let the probability thresholds α1, α2, …, α of the extracted h evidence sources for judging the possibility of power outage be h ;

[0017] When the basic probability m of the i-th evidence source i ≥α i , then let the value FT of the i-th evidence in the fault table be i =1; otherwise, let FT i =0;

[0018] Obtain a fault table FT with binary attributes.

[0019] Furthermore, the conflict coefficient and conflict distance J between the same evidence sources are calculated. ij ,include:

[0020] Take two evidence sources S1 and S2 as examples, whose basic probability distribution functions are m1 and m2 respectively. In the same identification framework, there are y suspected power outage-causing components and N power outage propositions. The probability thresholds are α1 and α2, and the fault table of the two evidence sources is FT.

[0021] The value of the nth power outage proposition in table FT is recorded as A n and B n The kernels A and B composed of these values ​​are represented by A={A1,..,A n ,…,A N} and B={B1,…,B n ,…,B N};

[0022] The calculation formula of the Jouselm distance between two evidence sources is expressed as:

[0023]

[0024] D 12 =[d1…d n …d N ] T ;

[0025] Where dn is the conflict coefficient, D 12 The conflict coefficient matrix of the two evidence sources S1 and S2 is represented by the matrix, and the coefficient is modified to 0. If dn = 0, then set dn = 1 / 2 y , to avoid the conflict distance of different evidence sources to be 0, where y is the number of components suspected of causing the power outage.

[0026] Furthermore, the conflict distance matrix between the evidence sources is:

[0027] When faced with h sources of evidence, calculate the conflict coefficient matrix D between source i and source j: ij , and obtain the conflict distance matrix J between evidence sources:

[0028]

[0029] In the formula, the element J in matrix J is ij represents the conflict distance between evidence source i and evidence source j, conflict distance J ij According to the conflict coefficient matrix D ij The distance is converted from the conflict coefficient dn in the range [0,1], where 0 means the two are completely consistent and 1 means the two are in complete conflict.

[0030] Furthermore, after obtaining the conflict distance matrix between the evidence sources, the method further includes:

[0031] Calculate the similarity S between any two evidence sources i and j ij =1-J ij , generate the similarity distance matrix S, the sum of the similarity distances between evidence source i and other evidence sources minus one represents the support level Sup(i) of the multi-evidence source system for evidence source i:

[0032]

[0033] The support level of each evidence source is normalized to the probability space, and the credibility Crd(i) of evidence source i in the multi-evidence source system is obtained:

[0034]

[0035] Furthermore, judging whether the non-diagonal elements in the conflict distance matrix meet the classification conditions by using the set conflict distance threshold and performing classification comprises the following steps:

[0036] S1: If max(Jij)≤Jα, it means that there is no strong conflict between the evidences and no classification is required; if max(Jij)>Jα, further classification is required, where J ij represents the conflict distance between evidence source i and evidence source j, and Jα represents the set conflict distance threshold;

[0037] S2: Find the minimum value Jij of the non-diagonal elements in the distance matrix J, i≠j, form the evidence classification G by the evidence source i and the evidence source j, and replace i and j;

[0038] S3: Find the maximum value of the distance between evidence source i and evidence source j and other evidence, which constitutes the distance between evidence classification G and other evidence, that is, JGf = max(Jif, Jjf), and update the distance matrix J;

[0039] S4: If the minimum value of the non-diagonal elements of the updated distance matrix J is not greater than the threshold Jα, steps S1 and S2 are repeated until min(Jij)>Jα, and the evidence source classification is completed.

[0040] Furthermore, the step of fusing the evidence sources in the evidence source classification using the improved DS fusion rule comprises the following steps:

[0041] The weight of the evidence classification depends on the credibility of each evidence source within the classification Crd(i), so the weight Wk of the classification Gk is:

[0042] W k =Crd(1)+Crd(2)+…+Crd(e);

[0043] The improved weighted fusion rule based on classification DS evidence theory is obtained as follows:

[0044]

[0045] Wherein, formula (1) indicates that the e evidence sources in the formed classification Gk are fused respectively to obtain the classification fusion result; formula (2) indicates that the g evidence classifications are weighted fused to obtain the final fusion result.

[0046] The present invention also includes a multi-source power outage information fusion device, using the above method, including:

[0047] A collection unit, used to collect power outage information and extract power outage feature information as a source of evidence;

[0048] A probability distribution unit, used to calculate the probability distribution function value of each of the evidence sources and generate a fault table with binary attributes;

[0049] A conflict distance matrix unit, used to calculate the conflict coefficient and conflict distance between the same evidence source through the fault table to obtain the conflict distance matrix between the evidence sources;

[0050] A classification unit, used to determine whether the non-diagonal elements in the conflict distance matrix meet the classification conditions through a set conflict distance threshold, and classify them to form an evidence source classification;

[0051] The fusion unit is used to fuse the evidence sources in the evidence source classification by using the improved DS fusion rule to obtain a fusion result.

[0052] The present invention also includes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described above is implemented.

[0053] The present invention also includes a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0054] The beneficial effects of the present invention are as follows: when the DS evidence theory is used for fusion, a multi-evidence classification process is added to the present invention, which greatly reduces the conflicts between different evidence sources and makes the fusion result more accurate. The DS evidence theory fusion algorithm is suitable for solving the unknown and uncertainty of the problem, and does not require the acquisition of training samples under each power outage mode in advance. It has unique advantages in solving multi-attribute diagnosis problems of subjective uncertainty judgment. Considering that the conflict coefficient of the traditional combination rule represents the overall conflict situation between all evidence sources, it cannot reflect the degree of conflict between each evidence source, which will cause the fusion result to be contrary to common sense. It is proposed to measure the conflict coefficient and conflict distance between evidence to ensure that the conflict between evidence does not affect the accuracy of the fusion result. The fusion method effectively improves the reliability of subsequent decision-making and helps to accurately identify the cause of the power outage. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 is a flow chart of the method in Example 1;

[0057] Figure 2 It is a structural schematic diagram of the device in Example 1;

[0058] Figure 3 is a flow chart of the method in Example 2;

[0059] Figure 4 This is a flow chart of classification of multiple evidence sources in Example 2;

[0060] Figure 5 A schematic diagram of the structure of a computer device. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0062] Embodiment 1:

[0063] like Figure 1 A multi-source power outage information fusion method is shown, comprising the following steps:

[0064] Collect power outage information and extract power outage feature information as a source of evidence;

[0065] Calculate the probability distribution function value of each evidence source and generate a fault table with binary attributes;

[0066] The conflict coefficient and conflict distance between the same evidence sources are calculated through the fault table to obtain the conflict distance matrix between the evidence sources;

[0067] The conflict distance threshold is set to determine whether the non-diagonal elements in the conflict distance matrix meet the classification conditions, and then the elements are classified to form the evidence source classification.

[0068] The improved DS fusion rule is used to fuse the evidence sources in the evidence source classification to obtain the fusion result.

[0069] When a power outage or other fault occurs in the power grid, the devices and systems therein will generate a large amount of alarm information and data. Since the information from these different sources may differ or conflict, information fusion is required to provide a more accurate and comprehensive power outage or fault diagnosis. The DS evidence theory fusion algorithm is suitable for solving the unknown and uncertainty of the problem. It does not require the acquisition of training samples under each power outage mode in advance, and has unique advantages in solving multi-attribute diagnosis problems with subjective uncertainty judgments.

[0070] This embodiment considers that the conflict coefficient of the traditional combination rule represents the overall conflict situation among all evidence sources, but cannot reflect the degree of conflict among various evidence sources, which may cause the fusion result to be contrary to common sense. It is proposed to measure the conflict coefficient and conflict distance among evidences to ensure that the accuracy of the fusion result is not affected when there is a conflict among evidences.

[0071] When using DS evidence theory for fusion, a multi-evidence classification process is added, which greatly reduces the conflicts between different evidence sources and makes the fusion result more accurate. The fusion method proposed in this embodiment effectively improves the reliability of subsequent decision-making and helps to accurately identify the cause of the power outage.

[0072] In this embodiment, before extracting the power outage feature information as a source of evidence, the collected information is also cleaned, standardized and time synchronized.

[0073] The probability distribution function value of each evidence source is calculated, and a fault table with binary attributes is generated, including the following steps:

[0074] Let the probability thresholds α1, α2, …, α of the extracted h evidence sources for judging the possibility of power outage be h ;

[0075] When the basic probability m of the i-th evidence source i ≥α i , then let the value FT of the i-th evidence in the fault table be i =1; otherwise, let FT i =0;

[0076] Obtain a fault table FT with binary attributes.

[0077] As a preferred embodiment of the above, the conflict coefficient and conflict distance J between the same evidence sources are calculated. ij ,include:

[0078] Take two evidence sources S1 and S2 as examples, whose basic probability distribution functions are m1 and m2 respectively. In the same identification framework, there are y suspected power outage-causing components and N power outage propositions. The probability thresholds are α1 and α2, and the fault table of the two evidence sources is FT.

[0079] The value of the nth power outage proposition in table FT is recorded as A n and B n The kernels A and B composed of these values ​​are represented by A={A1,..,A n ,…,A N} and B={B1,…,B n ,…,B N};

[0080] The calculation formula of the Jouselm distance between two evidence sources is expressed as:

[0081]

[0082] D 12 =[d1…d n …d N ] T ;

[0083] Where dn is the conflict coefficient, D 12 The conflict coefficient matrix of the two evidence sources S1 and S2 is represented by the matrix, and the coefficient is modified to 0. If dn = 0, then set dn = 1 / 2 y , to avoid the conflict distance of different evidence sources to be 0, where y is the number of components suspected of causing the power outage.

[0084] The conflict distance matrix between evidence sources is:

[0085] When faced with h sources of evidence, calculate the conflict coefficient matrix D between source i and source j: ij , and obtain the conflict distance matrix J between evidence sources:

[0086]

[0087] In the formula, the element J in matrix J is ij represents the conflict distance between evidence source i and evidence source j, conflict distance J ij According to the conflict coefficient matrix D ij The distance is converted from the conflict coefficient dn in the range [0,1], where 0 means the two are completely consistent and 1 means the two are in complete conflict.

[0088] In this embodiment, after obtaining the conflict distance matrix between the evidence sources, the following steps are also included:

[0089] Calculate the similarity S between any two evidence sources i and j ij =1-J ij , generate the similarity distance matrix S, the sum of the similarity distances between evidence source i and other evidence sources minus one represents the support level Sup(i) of the multi-evidence source system for evidence source i:

[0090]

[0091] The support level of each evidence source is normalized to the probability space, and the credibility Crd(i) of evidence source i in the multi-evidence source system is obtained:

[0092]

[0093] The conflict distance threshold is set to determine whether the non-diagonal elements in the conflict distance matrix meet the classification conditions, and then classify them, including the following steps:

[0094] S1: If max(Jij)≤Jα, it means that there is no strong conflict between the evidences and no classification is required; if max(Jij)>Jα, further classification is required, where J ij represents the conflict distance between evidence source i and evidence source j, and Jα represents the set conflict distance threshold;

[0095] S2: Find the minimum value Jij of the non-diagonal elements in the distance matrix J, i≠j, form the evidence classification G by the evidence source i and the evidence source j, and replace i and j;

[0096] S3: Find the maximum value of the distance between evidence source i and evidence source j and other evidence, which constitutes the distance between evidence classification G and other evidence, that is, JGf = max(Jif, Jjf), and update the distance matrix J;

[0097] S4: If the minimum value of the non-diagonal elements of the updated distance matrix J is not greater than the threshold Jα, steps S1 and S2 are repeated until min(Jij)>Jα, and the evidence source classification is completed.

[0098] The improved DS fusion rule is used to fuse the evidence sources in the evidence source classification, including the following steps:

[0099] The weight of the evidence classification depends on the credibility of each evidence source within the classification Crd(i), so the weight Wk of the classification Gk is:

[0100] W k =Crd(1)+Crd(2)+…+Crd(e);

[0101] The improved weighted fusion rule based on classification DS evidence theory is obtained as follows:

[0102]

[0103] Wherein, formula (1) indicates that the e evidence sources in the formed classification Gk are fused respectively to obtain the classification fusion result; formula (2) indicates that the g evidence classifications are weighted fused to obtain the final fusion result.

[0104] like Figure 2 As shown, this embodiment also includes a multi-source power outage information fusion device, using the above method, including:

[0105] A collection unit, used to collect power outage information and extract power outage feature information as a source of evidence;

[0106] A probability distribution unit, used to calculate the probability distribution function value of each evidence source and generate a fault table with binary attributes;

[0107] The conflict distance matrix unit is used to calculate the conflict coefficient and conflict distance between the same evidence sources through the fault table to obtain the conflict distance matrix between the evidence sources;

[0108] A classification unit is used to determine whether the non-diagonal elements in the conflict distance matrix meet the classification conditions through a set conflict distance threshold, and to classify them to form a classification of evidence sources;

[0109] The fusion unit is used to fuse the evidence sources in the evidence source classification by using the improved DS fusion rule to obtain a fusion result.

[0110] Embodiment 2:

[0111] like Figure 3 As shown, this embodiment includes a multi-source power outage information fusion method based on DS evidence theory, including the following steps:

[0112] Step 1: Set up information collection points specifically for collecting power outage information in the active distribution network, perform data cleaning, data standardization and time synchronization on the collected information, and extract power outage feature information from the pre-processed power outage information as a source of evidence.

[0113] Step 2: Record the probability thresholds α1, α2, …, α of the h evidence sources extracted in step 1 to determine the possibility of power outage h , when the basic probability m of the i-th evidence source i ≥α i , then let the value FT of the i-th evidence in the fault table be i =1; otherwise, let FT i =0, we can get a fault table FT with binary attributes. The fault table can intuitively determine whether there is a possibility of conflict between different evidence sources for the same proposition, and then calculate the conflict coefficient and conflict distance through the Jouselm function.

[0114] Preferably, for multiple evidence sources, the conflict coefficient represents the overall conflict situation among all evidence sources, and cannot reflect the degree of conflict among individual evidence sources. The essence of conflict among evidence is the conflict among evidence sources when the degree of support for a proposition by an evidence source differs greatly from the degree of support for the proposition by other evidence sources. Existing methods mainly redistribute the original evidence probability by calculating similarity metrics, confidence function entropy, and global conflict coefficients, and modify the focal element and weight of the conflicting evidence source to reduce the degree of conflict. However, this method also reduces the ability of conflicting evidence to explain the results, and cannot be fully applied to the field of multi-source power outage information fusion. Therefore, this paper proposes an improved weighted fusion algorithm based on the DS evidence theory of classification, which measures the conflict coefficient and conflict distance between evidence sources through the Jouselme function, avoiding the problem of using classical combination rules for information fusion, which will produce results contrary to common sense.

[0115] Taking two evidence sources S1 and S2 as examples, their basic probability distribution functions are m1 and m2 respectively. In the same identification framework, there are y suspected power outage components and N power outage propositions. The probability thresholds are α1 and α2. The fault table of the two evidence sources is FT. The value of the nth power outage proposition in table FT is recorded as A n and B n The kernels A and B composed of these values ​​are represented by A={A1,..,A n ,…,A N} and B={B1,…,B n ,…,B N}, the calculation formula of the Jouselm distance between two evidence sources can be expressed as:

[0116]

[0117] D 12 =[d1…d n …d N ] T (3)

[0118] Among them, dn is the conflict coefficient, D 12 The conflict coefficient matrix of the two evidence sources S1 and S2 is represented by the matrix, and the coefficient is modified to 0. If dn = 0, then set dn = 1 / 2 y , to avoid the problem of conflict distance of different evidence sources being 0, where y is the number of components suspected of causing the power outage.

[0119] When there are h evidence sources, the conflict coefficient matrix D between evidence source i and evidence source j is calculated by formula (2): ij , and then the conflict distance matrix J between evidence sources can be obtained by formula (1):

[0120]

[0121] The element J in matrix J ij represents the conflict distance between evidence source i and evidence source j, conflict distance J ij According to the conflict coefficient matrix D ij The distance is converted from the conflict coefficient dn in the range [0,1], where 0 means the two are completely consistent and 1 means the two are in complete conflict.

[0122] In contrast to the conflict distance matrix, the similarity S between any two evidence sources i and j is ij =1-J ij , a similarity distance matrix S can be generated. The sum of the similarity distances between evidence source i and other evidence sources minus one represents the support level Sup(i) of the multi-evidence source system for evidence source i:

[0123]

[0124] By normalizing the support level of each evidence source to the probability space, we can obtain the credibility Crd(i) of evidence source i in the multi-evidence source system:

[0125]

[0126] Step 3: By setting the conflict distance threshold Jα, determine whether the non-diagonal elements (i.e., i≠j) in the conflict distance matrix J obtained in step 2 meet the classification conditions, and then classify the multiple evidence sources to form evidence classification. The process of classifying multiple evidence sources to form evidence classification is as follows: Figure 4 As shown, the specific steps can be described as:

[0127] 1. If max(Jij)≤Jα, it means that there is no strong conflict between the evidences and no classification is required; if max(Jij)>Jα, further classification is required.

[0128] 2. Find the minimum value Jij, (i≠j) of the non-diagonal elements in the distance matrix J, form the evidence classification G by the evidence source i and the evidence source j, and replace i and j.

[0129] 3. Find the maximum value of the distance between evidence source i and evidence source j and other evidence, which constitutes the distance between evidence classification G and other evidence, that is, JGf = max(Jif, Jjf), and update the distance matrix J;

[0130] 4. If the minimum value of the non-diagonal elements of the updated distance matrix J is not greater than the threshold Jα, repeat steps 1 and 2 until min(Jij)>Jα, which means that the distances between the categories meet the classification requirements and the classification of the evidence sources is completed. Thus, the h evidence sources that meet the conditions are divided into g evidence categories G1, G2, …, Gg, where any category Gk consists of e evidence sources, that is:

[0131] G k ={m1,m2,…,m e} (7)

[0132] Step 4: Fusion the evidence sources classified in step 3, and obtain the credibility of evidence source i Crd(i) in step 2, which represents the weight of evidence source i. The weight of evidence classification depends on the credibility Crd(i) of each evidence source within the classification. It can be seen that the weight Wk of classification Gk is:

[0133] W k =Crd(1)+Crd(2)+…+Crd(e) (8)

[0134] Then the improved weighted fusion rule based on classification DS evidence theory can be obtained as follows:

[0135]

[0136] In formula (9), formula (1) indicates that the e evidence sources in the formed classification Gk are fused separately to obtain the classification fusion result; formula (2) indicates that the g evidence classifications are weightedly fused to obtain the final fusion result.

[0137] See also Figure 5 A computer device 400 provided in an embodiment of the present application includes: a processor 410 and a memory 420, wherein the memory 420 stores a computer program executable by the processor 410, and when the computer program is executed by the processor 410, the above method is executed.

[0138] The embodiment of the present application further provides a storage medium 430 on which a computer program is stored. When the computer program is run by the processor 410, the above method is executed.

[0139] Among them, the storage medium 430 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0140] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. "Multiple" means two or more, unless otherwise clearly and specifically defined.

[0141] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0142] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0143] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0144] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0145] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0146] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0147] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A multi-source power outage information fusion method, characterized in that: The steps include: Collect power outage information and extract power outage feature information as a source of evidence; Calculating the probability distribution function value of each of the evidence sources and generating a fault table with binary attributes; Calculate the conflict coefficient and conflict distance between the same evidence sources through the fault table to obtain the conflict distance matrix between the evidence sources; Determine whether the non-diagonal elements in the conflict distance matrix meet the classification conditions through the set conflict distance threshold, and classify them to form evidence source classification; The improved DS fusion rule is used to fuse the evidence sources in the evidence source classification to obtain a fusion result.

2. The multi-source power outage information fusion method according to claim 1, characterized in that: Before extracting the power outage feature information as a source of evidence, the collected information is also cleaned, standardized and time synchronized.

3. The multi-source power outage information fusion method according to claim 1, characterized in that: The method of calculating the probability distribution function value of each evidence source and generating a fault table with binary attributes includes the following steps: Let the probability thresholds α1, α2, …, α of the extracted h evidence sources for judging the possibility of power outage be h ; When the basic probability m of the i-th evidence source i ≥α i , then let the value FT of the i-th evidence in the fault table be i =1; On the contrary, let FT i =0; Obtain a fault table FT with binary attributes.

4. The multi-source power outage information fusion method according to claim 3 is characterized in that: The conflict coefficient and conflict distance J between the same evidence sources are calculated. ij ,include: For two evidence sources S1 and S2, their basic probability distribution functions are m1 and m2 respectively. In the same identification framework, there are y suspected power outage-causing components and N power outage propositions. The probability thresholds are α1 and α2, and the fault table of the two evidence sources is FT. The value of the nth power outage proposition in table FT is recorded as A n and B n The kernels A and B composed of these values ​​are represented by A={A1,..,A n ,…,A N } and B={B1,…,B n ,…,B N }; The calculation formula of the Jouselm distance between two evidence sources is expressed as: D 12 =[d1…d n …d N ] T ; Where dn is the conflict coefficient, D 12 The conflict coefficient matrix of the two evidence sources S1 and S2 is represented by the matrix, and the coefficient is 0. If dn = 0, then dn = 1 / 2 y , to avoid the conflict distance of different evidence sources to be 0, where y is the number of components suspected of causing the power outage.

5. The multi-source power outage information fusion method according to claim 1, characterized in that: The conflict distance matrix between the evidence sources is: When faced with h sources of evidence, calculate the conflict coefficient matrix D between source i and source j: ij , we get the conflict distance matrix J between evidence sources: In the formula, the element J in matrix J is ij represents the conflict distance between evidence source i and evidence source j, conflict distance J ij According to the conflict coefficient matrix D ij The distance is converted from the conflict coefficient dn in , and the distance range is [0,1]. A distance of 0 means that the two are completely consistent, and a distance of 1 means that the two are in complete conflict.

6. The multi-source power outage information fusion method according to claim 5, characterized in that: After obtaining the conflict distance matrix between the evidence sources, the method further includes: Calculate the similarity S between any two evidence sources i and j ij =1-J ij , generate the similarity distance matrix S, the sum of the similarity distances between evidence source i and other evidence sources minus one represents the support level Sup(i) of the multi-evidence source system for evidence source i: The support degree of each evidence source is normalized to the probability space, and the credibility Crd(i) of evidence source i in the multi-evidence source system is obtained:

7. The multi-source power outage information fusion method according to claim 1, characterized in that: The step of judging whether the non-diagonal elements in the conflict distance matrix meet the classification conditions by using the set conflict distance threshold and performing classification comprises the following steps: S1: If max(J ij )≤Jα, it means that there is no strong conflict between the evidences and no classification is required; if max(Jij)>Jα, further classification is required, where J ij represents the conflict distance between evidence source i and evidence source j, and Jα represents the set conflict distance threshold; S2: Find the minimum value J of the non-diagonal elements in the distance matrix J ij , i≠j, evidence source i and evidence source j form evidence classification G, and replace i and j; S3: Find the maximum value of the distance between evidence source i and evidence source j and other evidence, which constitutes the distance between evidence category G and other evidence, that is, JGf = max(J if ,J jf ), update the distance matrix J; S4: If the minimum value of the non-diagonal elements of the updated distance matrix J is not greater than the threshold Jα, repeat steps S1 and S2 until min(J ij )>Jα, the classification of evidence sources is completed.

8. The multi-source power outage information fusion method according to claim 6, characterized in that: The step of fusing the evidence sources in the evidence source classification using the improved DS fusion rule comprises the following steps: The weight of evidence classification depends on the credibility of each evidence source within the classification, Crd(i), so it can be seen that the classification G k The weight W k for: W k =Crd(1)+Crd(2)+…+Crd(e); The improved weighted fusion rule based on classification DS evidence theory is obtained as follows: In the formula, formula (1) represents the classification G formed respectively k The e evidence sources in are fused to obtain the classified fusion result; Formula (2) represents the weighted fusion of g evidence classifications to obtain the final fusion result.

9. A multi-source power outage information fusion device, characterized in that: Use of the method according to any one of claims 1 to 8, comprising: A collection unit, used to collect power outage information and extract power outage feature information as a source of evidence; A probability distribution unit, used to calculate the probability distribution function value of each of the evidence sources and generate a fault table with binary attributes; A conflict distance matrix unit, used to calculate the conflict coefficient and conflict distance between the same evidence source through the fault table to obtain the conflict distance matrix between the evidence sources; A classification unit, used to determine whether the non-diagonal elements in the conflict distance matrix meet the classification conditions through a set conflict distance threshold, and classify them to form an evidence source classification; The fusion unit is used to fuse the evidence sources in the evidence source classification by using the improved DS fusion rule to obtain a fusion result.

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

11. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images