Electrical fire risk assessment method for medium and low voltage distribution network

Abstract

The invention discloses an electrical fire risk assessment method for a medium and low voltage power distribution network. The method comprises the following steps: 1, constructing a layered and multivariate power distribution network electrical fire risk assessment index system; 2, based on an order relation analysis method and an entropy weight method, introducing an information bearing capacity in an information entropy optimization CRITIC method, then based on an improved CRITIC combination method, calculating an algorithm of weight of each index, and using a Kendall consistency coefficient to test subjective and objective weights to obtain a power distribution network equipment risk assessment value; 3, calculating the importance of each power distribution network device by adopting a cloud DEMATEL device importance algorithm based on a CWPHM operator; 4, solving the risk assessment value of the power distribution network according to the network topological graph and the equipment risk assessment value by using an equipment importance degree improved success flow method; and 5, dividing the state of the power distribution equipment into five grades. The risk degree of the power distribution network can be evaluated more accurately and effectively, and arrangement of equipment maintenance work is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical fire risk and early warning assessment in power systems, and in particular to an electrical fire risk assessment method for medium and low voltage distribution networks. Background Art

[0002] With the rapid development of economic construction and the acceleration of urbanization, the rapid increase in smart homes, distributed photovoltaics (PV) and electric vehicles, the operating conditions of distribution networks have become increasingly complex and changeable, increasing the risk of distribution network failures and even electrical fires.

[0003] In particular, there are blind spots in supervision in the distribution network from distribution stations in residential communities (or commercial centers) to residential units (or businesses). Problems such as aging lines and illegal electricity use lead to frequent accidents, threatening the safe and stable operation of the distribution network.

[0004] However, modern buildings and facilities are usually equipped with a large number of electrical equipment and complex circuit systems. The complexity of these equipment and lines increases the difficulty of management and maintenance, and also increases the risk of electrical fires. Therefore, it is necessary to detect and warn of distribution network-related faults in advance to reduce losses.

[0005] The distribution network is a dynamic system containing a variety of equipment. The degree of its operating risk is affected by many factors, such as distribution equipment, environmental weather, network architecture and power-consuming equipment.

[0006] The existing distribution network risk assessment is an important means to ensure the safe and stable operation of the power system. The current methods for assessing the risk level of the distribution network mostly consider the impact of distribution network lines, transformers, circuit breakers and other equipment, and the evaluation indicators do not consider the impact of electrical equipment on the distribution network. The evaluation method is single and too subjective, and does not reflect the importance of different distribution network equipment in the distribution network.

[0007] Therefore, how to evaluate the operating risk status at the end of the above-mentioned medium and low voltage distribution network in order to promptly discover and solve the parts with higher fault risks in the distribution network, formulate targeted safety management and control measures, and reduce the probability of electrical fires in the distribution network has become a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0008] In view of the above-mentioned defects of the prior art, the present invention provides a method for assessing the electrical fire risk of medium and low voltage distribution networks. The purpose of the method is to obtain equipment risk assessment values ​​and importance rankings based on the influence of new loads and distributed photovoltaic systems and other loads, so as to make a more accurate and effective assessment of the risk level of the distribution network and be more conducive to the arrangement of equipment maintenance work.

[0009] To achieve the above object, the present invention discloses a method for assessing the risk of electrical fires in medium and low voltage distribution networks, comprising the following steps:

[0010] Step 1: Construct a hierarchical and multi-dimensional distribution network electrical fire risk assessment index system including distribution equipment, lines, power loads and new loads;

[0011] Step 2: Based on the order relationship analysis (G1) method and the entropy weight method, the information entropy is introduced to optimize the information carrying capacity in the CRITIC method, and then the algorithm for obtaining the weights of each indicator is based on the improved CRITIC combination method. The Kendall consistency coefficient is used to test the subjective and objective weights to obtain the risk assessment value of the distribution network equipment.

[0012] Step 3: Use the Cloud DEMATEL device importance algorithm based on the CWPHM operator to calculate the importance of each distribution network device;

[0013] Step 4: According to the network topology and the equipment risk assessment value, the risk assessment value of the distribution network is solved by using the equipment importance improved success flow (GO) method;

[0014] Step 5: Divide the risk status of distribution equipment and distribution network into five levels: healthy, sub-healthy, caution, abnormal and severe.

[0015] Preferably, in step 1, the power load and the new load are water heaters, smart air conditioners, charging piles and distributed photovoltaic power generation systems, and the specific steps are as follows:

[0016] Step 1.1, select the main distribution equipment in the distribution network and the main power equipment and new loads in the community as secondary indicators;

[0017] Step 1.2, according to the operation characteristics of the secondary equipment corresponding to the secondary indicators, and taking into account weather environment factors and fault characteristic factors, select the third-level indicators corresponding to each of the secondary equipment;

[0018] Step 1.3: According to DL / T 2106-2020 "Guidelines for Distribution Network Equipment Status Evaluation" and DB11 / T2036-2022 "Technical Specifications for Electrical Safety of Distributed Photovoltaic Power Generation Systems", a three-level distribution network risk index scoring standard is constructed as the hierarchical and multi-dimensional distribution network electrical fire risk assessment index system.

[0019] More preferably, in step 1.1, the main power distribution equipment and distribution lines in the distribution network include low-voltage cables, transformers and circuit breakers; the main power-consuming equipment and new loads in the community include water heaters, smart air conditioners, charging piles and distributed photovoltaic power generation systems;

[0020] The first-level indicator of the three-level indicator scoring standard for distribution network risk is: distribution network A, and the second-level indicators include cables B1, transformers B2, circuit breakers B3, smart air conditioners B4, water heaters B5, distributed photovoltaic power generation systems B6 and charging piles B7;

[0021] The three-level indicators corresponding to the cable B1 include cable temperature C 11 , Cable voltage deviation C 12 、Partial discharge C 13 , Cable operation life C 14 , Cable ambient temperature C 15 and cable ambient humidity C 16 ;

[0022] The three-level indicators corresponding to transformer B2 include oil temperature C 21 、Oil level C 22 , Transformer ambient temperature C 23 , and the voltage deviation C of phase A in the three-phase 24 、B phase voltage deviation C 25 、C phase voltage deviation C 26 、A phase current deviation C 27 、B phase current deviation C 28 and C phase current deviation C 29 ;

[0023] The three-level indicators corresponding to the circuit breaker B3 include the service life of the circuit breaker C 31 , breaking times C 32 , Vacuum degree C 33 , opening synchronism C 34 And closing synchronism C 35 ;

[0024] The three-level indicators corresponding to the smart air conditioner B4 include air conditioner voltage deviation C 41 、Air conditioning current deviation C 42 and air conditioner service life C 43 ;

[0025] The three-level indicators corresponding to the water heater B5 include the water heater voltage deviation C 51 , Water heater current deviation C 52 、Service life of water heater C 53 and water heater failure times C 54 ;

[0026] The three-level indicators corresponding to the distributed photovoltaic power generation system B6 include the number of failures of the distributed photovoltaic power generation system C 61 , Distributed photovoltaic power generation system fault repair time C 62 , Common connection point voltage deviation C 63 and the penetration rate of distributed generation C 64;

[0027] The three-level indicators corresponding to the charging pile B7 include voltage stabilization accuracy C 71 、Temperature flow accuracy C 72 , current imbalance C 73 , output voltage error C 74 , output current error C 75 , Power Factor C 76 and the service life of the charging pile C 77 ;

[0028] The cable temperature C 11 The unit is: ℃, less than 70 is 4 to 5 points, between 70 and 80 is 2 to 4 points, between 80 and 90 is 0 to 2 points, and more than 90 is 0 points;

[0029] The cable voltage deviation C 12 As a percentage, less than 2 is 4 to 5 points, between 2 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0030] The partial discharge C 13 The unit is: dB, less than 8 is 4 to 5 points, between 8 and 12 is 2 to 4 points, between 12 and 15 is 0 to 2 points, and greater than 15 is 0 points;

[0031] The cable service life C 14 The unit is: year, less than 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 20 is 0 to 2 points, and more than 20 is 0 points;

[0032] The cable ambient temperature C 15 The unit is: ℃, less than 40 is 4 to 5 points, between 40 and 50 is 2 to 4 points, between 50 and 70 is 0 to 2 points, and more than 70 is 0 points;

[0033] The cable environment humidity C 16 As a percentage, less than 45 is 4 to 5 points, between 45 and 60 is 2 to 4 points, between 60 and 85 is 0 to 2 points, and more than 85 is 0 points;

[0034] The oil temperature C 21 The unit is: ℃, less than 45 is 4 to 5 points, between 45 and 75 is 2 to 4 points, between 75 and 90 is 0 to 2 points, and more than 90 is 0 points;

[0035] The oil level C 22 As a percentage, more than 80 is 4 to 5 points, between 80 and 60 is 2 to 4 points, between 60 and 40 is 0 to 2 points, and less than 40 is 0 points;

[0036] The transformer ambient temperature C23 The unit is: ℃, between -5 and 40 is 4 to 5 points, between 40 and 50 is 2 to 4 points, between 50 and 60 is 0 to 2 points, and above 60 is 0 points;

[0037] The A phase voltage deviation C 24 , the B phase voltage deviation C 25 and the C phase voltage deviation C 26 All are percentages, with any value less than 2 being 4 to 5 points, between 2 and 4 being 2 to 4 points, between 4 and 5 being 0 to 2 points, and greater than 5 being 0 points;

[0038] The A phase current deviation C 27 、The B phase current deviation C 28 and the C phase current deviation C 29 All are percentages, with any value less than 4 being 4 to 5 points, between 4 and 7 being 2 to 4 points, between 7 and 10 being 0 to 2 points, and greater than 10 being 0 points;

[0039] The service life of the circuit breaker C 31 The unit is: year, less than 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 15 is 0 to 2 points, and more than 15 is 0 points;

[0040] The breaking times C 32 The unit is: times, less than 3000 is 4 to 5 points, between 3000 and 6500 is 2 to 4 points, between 6500 and 10000 is 0 to 2 points, and more than 10000 is 0 points;

[0041] The vacuum degree C 33 The unit is Pa, less than 6.6 is 4 to 5 points, between 6.6 and 13.3 is 2 to 4 points, between 13.3 and 66 is 0 to 2 points, and greater than 66 is 0 points;

[0042] The opening synchronism C 34 The unit is ms, less than 0.5 is 4 to 5 points, between 0.5 and 1.2 is 2 to 4 points, between 1.2 and 2.0 is 0 to 2 points, and greater than 2.0 is 0 points;

[0043] The closing synchronism C 35 The unit is: ms, less than 1 is 4 to 5 points, between 1 and 2 is 2 to 4 points, between 2 and 3.5 is 0 to 2 points, and greater than 3.5 is 0 points;

[0044] The air conditioning voltage deviation C 41 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0045] The air conditioning current deviation C42 As a percentage, less than 1 is 4 to 5 points, between 1 and 3.5 is 2 to 4 points, between 3.5 and 5 is 0 to 2 points, and greater than 5 is 0 points;

[0046] The service life of the air conditioner C 43 The unit is year. Less than or equal to 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 15 is 0 to 2 points, and greater than 15 is 0 points.

[0047] The water heater voltage deviation C 51 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0048] The water heater current deviation C 52 As a percentage, less than 1 is 4 to 5 points, between 1 and 3.5 is 2 to 4 points, between 3.5 and 5 is 0 to 2 points, and greater than 5 is 0 points;

[0049] The service life of the water heater C 53 The unit is year. Less than 4 is 4 to 5 points. Between 4 and 8 is 2 to 4 points. Between 8 and 12 is 0 to 2 points. More than 12 is 0 points.

[0050] The water heater failure frequency C 54 The unit is: times / year, less than or equal to 1 is 4 to 5 points, 2 is 2 to 4 points, 3 is 0 to 2 points, and greater than 3 is 0 points;

[0051] The number of failures of the distributed photovoltaic power generation system C 61 The unit is: less than or equal to 1 is 4 to 5 points, 2 is 2 to 4 points, 3 is 0 to 2 points, and greater than 3 is 0 points;

[0052] The distributed photovoltaic power generation system fault repair time C 62 The unit is: h, less than 24 is 4 to 5 points, between 24 and 48 is 2 to 4 points, between 48 and 72 is 0 to 2 points, and more than 72 is 0 points;

[0053] The common connection point voltage deviation C 63 As a percentage, less than 2 is 4 to 5 points, between 2 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0054] The distributed generation penetration rate C 64 As a percentage, between 20 and 15 is 4 to 5 points, between 15 and 10 is 2 to 4 points, between 10 and 5 is 0 to 2 points, and less than 5 is 0 points;

[0055] The voltage stabilization accuracy C 71As a percentage, less than 0.3 is 4 to 5 points, between 0.3 and 0.5 is 2 to 4 points, between 0.5 and 1 is 0 to 2 points, and greater than 1 is 0 points;

[0056] The temperature flow accuracy C 72 As a percentage, less than 0.5 is 4 to 5 points, between 0.5 and 1 is 2 to 4 points, between 1 and 2 is 0 to 2 points, and greater than 2 is 0 points;

[0057] The current imbalance degree C 73 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0058] The output voltage error C 74 As a percentage, less than 0.3 is 4 to 5 points, between 0.3 and 0.5 is 2 to 4 points, between 0.5 and 1 is 0 to 2 points, and greater than 1 is 0 points;

[0059] The output current error C 75 As a percentage, less than 0.5 is 4 to 5 points, between 0.5 and 1 is 2 to 4 points, between 1 and 2 is 0 to 2 points, and greater than 2 is 0 points;

[0060] The power factor C 76 4 to 5 points between 0.95 and 1, 2 to 4 points between 0.9 and 0.95, 0 to 2 points between 0.85 and 0.9, and 0 points for less than 0.85;

[0061] The service life of the charging pile is C 77 The unit is year; less than 3 is 4 to 5 points, between 3 and 6 is 2 to 4 points, between 6 and 10 is 0 to 2 points, and greater than 10 is 0 point.

[0062] More preferably, step 2 is as follows:

[0063] Step 2.1, performing consistency check on the subjective and objective weights obtained by each of the m third-level evaluation indicators and the n weighting methods;

[0064] If the consistency test passes, the minimum information entropy principle based on Lagrange multiplication optimization is used to optimize the subjective and objective weights, and the combined weight W of the i-th indicator is calculated. i , the specific formula is as follows:

[0065]

[0066] Among them, w i,j is the weight value of the i-th third-level indicator obtained by the j-th evaluation method;

[0067] If the consistency test fails, the improved CRITIC combination method is used to find the coefficients of each weight, and the information carrying capacity in the CRITIC method is optimized according to the information entropy in the entropy weight method. Assuming that there are m evaluation indicators and n evaluation methods, they are as follows:

[0068] Step 2.1.1: Obtain the comparative performance S of the jth method j and the contradiction between the jth method weight and the other method weights R j , the specific formula is as follows:

[0069]

[0070] Among them, G kj represents the correlation coefficient between the kth method weight and the jth method weight. Here, the Pearson correlation coefficient is used, which is the quotient of the covariance and standard deviation between two variables. The number of weighting methods is n;

[0071] G kj The calculation formula is as follows:

[0072]

[0073] Among them, w i,k is the weight value of the i-th third-level indicator obtained by the k-th evaluation method;

[0074] is the average value of the jth method weight; is the average value of the kth method weight; the calculation formula is as follows:

[0075]

[0076] Step 2.1.2: Introduce the information entropy E corresponding to the index in the entropy weight method j :

[0077]

[0078] Step 2.1.3, considering the impact of variant data on information entropy, the activation entropy is used for optimization, combined with the multiplication model used in the traditional CRITIC method:

[0079] Ij=(1-1 / (1+exp(-Ej)))SjRj;

[0080] Step 2.1.4: Get the combined weight W of the i-th indicator i for:

[0081]

[0082] Step 2.2: According to the scoring criteria, the scoring value is obtained. Combined with the indicator combination weight, the final risk assessment value H2 of the secondary indicator is:

[0083]

[0084] Among them, O i is the score of the i-th third-level indicator; W i is the combined weight of the i-th third-level indicator

[0085] More preferably, step 3 is as follows:

[0086] Step 3.1: Convert the distribution equipment importance language multi-attribute decision-making matrix and the direct impact matrix between indicators into the corresponding cloud matrix, as follows:

[0087] Step 3.1.1, select the language phrase d for the distribution network equipment l The number of is set to 7, so that the language evaluation set D = {d l |d -3 ,d -2 ,d -1 ,d0,d1,d2,d3}={extremely low, very low, low, average, high, very high, extremely high};

[0088] Using θ l represents the importance evaluation of distribution network equipment, then θ l =[0,1], the specific calculation formula is as follows:

[0089]

[0090] Among them, select the language phrase d l When the number of is 7, a≈1.37;

[0091] Step 3.1.2: Given the domain interval, that is, the evaluation score interval U = [X min ,X max ], the language phrase d l Convert to the corresponding forward cloud model A l =C l (Ex l ,En l ,He l ); The digital characteristics of the cloud model can reflect the overall characteristics of the qualitative concept, denoted as C(Ex,En,He);

[0092] Among them, Ex l is the expected distribution of cloud droplets in the domain space;

[0093] En l is entropy, representing the uncertainty and ambiguity of cloud droplet distribution;

[0094] He l is the hyperentropy, which is used to measure the uncertainty of entropy, that is, the entropy of entropy;

[0095] Calculate Expected Ex l The value of is:

[0096] Ex l =X min +θ l (X max -X min );

[0097] Step 3.1.3. Calculate entropy En l The value of is:

[0098]

[0099] Step 3.1.4. Calculate the super entropy He l The values ​​are:

[0100]

[0101] Step 3.1.5, construct the direct impact matrix P based on the cloud model;

[0102]

[0103] Among them, p gh is the degree of influence of the decision maker on the hth indicator on the gth indicator; r is the number of importance indicators;

[0104] Step 3.2: Use the C-DEMATEL combination method to determine the indicator weights, specifically:

[0105] Step 3.2.1: The cloud matrix P after the direct influence matrix between indicators is converted from step 3.1 = (p gh ) r×r , where the main diagonal element p gg =minA=(Ex -l ,En -l ,He -l ) indicates that the evaluation index has no influence on itself;

[0106] Step 3.2.2: Set the cloud score p gh Convert to score gh And normalize, get the matrix Z = (z gh ) r×r Among them, zy gh and z gh The details are as follows:

[0107]

[0108] Among them, zy gh ∈[0,1],d(p gh ,maxA) is p gh The Hamming distance between the maximum cloud maxA and d(p gh ,minA) is p gh The Hamming distance between the minimum cloud minA and the minimum cloud minA;

[0109] Step 3.2.3, calculate the Hamming distance between clouds A1 and A2, specifically:

[0110]

[0111] Step 3.2.4, get the comprehensive influence matrix V = (v gh ) r×r ;

[0112] V==Z(IZ) -1 ;

[0113] Where I is the identity matrix;

[0114] Step 3.2.5: Get the weight α of each evaluation index f , specifically:

[0115]

[0116] Among them, β f is the sum of the elements in the fth row of the comprehensive influence matrix V, δ f is the sum of the elements in the fth column of the comprehensive influence matrix V; where α f is the weight of the fth evaluation index;

[0117] Step 3.3, using the CWPHM operator to determine the comprehensive evaluation value of each distribution network equipment;

[0118] For the cloud matrix B after the multi-attribute decision-making matrix transformation, B = (B1, B2, ..., B r ), the corresponding weight vector is α=(α1,α2,…,α r ) T , let CWPHM:B r →B, then

[0119]

[0120] in, Indicates B e and The support degree; Indicates B e and The distance between

[0121] Define the normalized distance between B1 and B2 as:

[0122]

[0123] Step 3.4, quantitative value x∈U, and x~N(Ex,En′ 2 ), En′=N(En,He 2 ), the membership degree of the quantitative value x to the qualitative language concept T satisfies:

[0124]

[0125] The distribution of the membership degree y on the given domain U is called the normal cloud model, and each group (x, y) is represented as a cloud droplet. The score of a cloud droplet is defined as s = xy. Then the mathematical expectation of the score s is It is the total score of the cloud model, i.e., the importance of the distribution network equipment.

[0126] More preferably, in step 4, the steps are as follows:

[0127] Step 4.1: There is an exponential relationship between the risk assessment value of the power distribution equipment and the corresponding probability of successful operation, as follows:

[0128]

[0129] Among them, H2 is the risk assessment value of the power distribution equipment; DB is the probability of successful operation of the power distribution equipment;

[0130] Step 4.2: Starting from the bottom layer of the distribution network hierarchy diagram, all devices in the same layer can be equivalent to one device. For all devices connected in series on the branch line, the probability of successful operation is:

[0131]

[0132] in, is the successful operation probability of the kth device in the j1th branch line of the i1th layer; is the probability of successful operation of all devices in the j1-th branch line of the i1-th layer connected in series; is the importance of the kth device in the j1th branch line of the i1th layer; m1 is the total number of all devices in the j1th branch line of the i1th layer;

[0133] After the devices on the same branch line are connected in series and equivalent, the probability of successful operation of the series equivalent devices connected in parallel on the branch lines on the same layer is:

[0134]

[0135] In the formula, is the probability of successful operation of all devices in layer i1; is the ratio of the importance of all devices in the j1-th branch line of the i1-th layer to the importance of all devices in this layer;

[0136] Step 4.3: Repeat steps 4.1 and 4.2 to obtain the probability of successful operation of the distribution network. By converting the formula, the risk assessment value H1 of the first-level distribution network can be deduced. The specific formula is as follows:

[0137]

[0138] In this step, by introducing the equivalence theory in the GO method for optimizing equipment importance, and based on the relationship between the equipment risk assessment value and the probability of successful operation of the equipment, the risk assessment value of the distribution equipment is derived into the risk assessment value of the distribution network.

[0139] More preferably, step 5 is to add a sub-health level on the basis of the four levels of healthy, caution, abnormal and severe in DL / T 2106-2020 "Guidelines for Distribution Network Equipment Status Evaluation", as follows:

[0140] The risk level is the risk assessment value corresponding to the health status [4,5], the representative color is green, and the performance description is that the distribution network or equipment is in excellent condition, there is no abnormality in operation, and the incidence of electrical fire is very low;

[0141] The risk level is the risk assessment value corresponding to the sub-health mentioned above is [3,4], the representative color is blue, and the performance description is that the distribution network or equipment is in good condition, there may be minor problems, and the incidence of electrical fires is low;

[0142] The risk level is the risk assessment value corresponding to the above-mentioned attention is [2,3], the representative color is yellow, the performance description is that the distribution network or equipment status is general, there may be some abnormalities, and the incidence of electrical fires is general;

[0143] The risk level is the risk assessment value corresponding to the anomaly, which is [1,2], and the representative color is orange. The performance description is that the distribution network or equipment is in poor condition, there are many anomalies, and the incidence of electrical fires is high;

[0144] The risk level is severe and the corresponding risk assessment value is [0,1], the representative color is red, and the performance description is that the distribution network or equipment is in a very poor state, there are many anomalies, and the incidence of electrical fires is very high.

[0145] Beneficial effects of the present invention:

[0146] According to the impact of new loads and distributed photovoltaic systems on the risk level of the distribution network, the present invention constructs an electrical fire risk assessment index system and scoring standard for the distribution network involving distribution lines, distribution equipment and power-consuming equipment, which reflects the risk level of the distribution network in a more scientific and comprehensive manner.

[0147] The present invention adopts the improved CRITIC method to combine the subjective and objective weights of G1 and entropy weight method, which can not only retain expert opinions but also make the weighting results more reasonable; the cloud DEMATEL linguistic multi-attribute decision-making method is used to calculate the equipment importance, which effectively handles the fuzzy and uncertain relationship between indicators; the GO method is improved by using the equipment importance to more accurately reflect the overall risk level of the distribution network.

[0148] The application of the present invention can obtain the equipment risk assessment value and importance ranking, which is beneficial to the arrangement of equipment maintenance work; the improved CRITIC method and the improved GO method can accurately and effectively evaluate the risk level of the distribution network, and accurately reflect the improvement of the risk level of the distribution network caused by equipment maintenance and replacement.

[0149] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0150] Figure 1 The following is a flowchart showing an implementation of an embodiment of the present invention.

[0151] Figure 2 A network topology diagram in an embodiment of the present invention is shown.

[0152] Figure 3 A simplified network topology diagram of a 10 kV distribution network in a certain location according to an embodiment of the present invention is shown.

[0153] Figure 4 The risk assessment value of the distribution equipment layer in one embodiment of the present invention is shown. DETAILED DESCRIPTION

[0154] Example

[0155] like Figure 1 and Figure 2 As shown, the method for assessing the electrical fire risk of medium and low voltage distribution networks includes the following steps:

[0156] Step 1: Construct a hierarchical and multi-dimensional distribution network electrical fire risk assessment index system including distribution equipment, lines, power loads and new loads;

[0157] Step 2: Based on the order relationship analysis (G1) method and the entropy weight method, the information entropy is introduced to optimize the information carrying capacity in the CRITIC method, and then the algorithm for obtaining the weights of each indicator is based on the improved CRITIC combination method. The Kendall consistency coefficient is used to test the subjective and objective weights to obtain the risk assessment value of the distribution network equipment.

[0158] Step 3: Use the Cloud DEMATEL device importance algorithm based on the CWPHM operator to calculate the importance of each distribution network device;

[0159] Step 4: Figure 2 The network topology diagram and equipment risk assessment value shown in the figure are used, and the risk assessment value of the distribution network is solved by using the equipment importance improved success flow (GO) method;

[0160] Step 5: Divide the status of distribution equipment and distribution network risk into five levels: healthy, sub-healthy, caution, abnormal and severe.

[0161] In some embodiments, in step 1, the power load and the new load are water heaters, smart air conditioners, charging piles and distributed photovoltaic power generation systems. The specific steps are as follows:

[0162] Step 1.1, select the main distribution equipment in the distribution network and the main power equipment and new loads in the community as secondary indicators;

[0163] Step 1.2, according to the operation characteristics of the secondary equipment corresponding to the secondary indicators, and taking into account the weather environment factors and the fault characteristic factors, select the third-level indicators corresponding to each secondary equipment;

[0164] Step 1.3: According to DL / T 2106-2020 "Guidelines for Distribution Network Equipment Status Evaluation" and DB11 / T2036-2022 "Technical Specifications for Electrical Safety of Distributed Photovoltaic Power Generation Systems", a three-level distribution network risk index scoring standard is constructed as a layered and diversified distribution network electrical fire risk assessment indicator system.

[0165] In practical applications, from the two dimensions of distribution equipment and distribution system, based on comprehensive consideration of the characteristics of distribution equipment, distribution lines and power loads, a hierarchical and diversified distribution network electrical fire risk assessment index system for distribution equipment, lines, power loads and new loads is constructed.

[0166] In some embodiments, in step 1.1, the main power distribution equipment and distribution lines in the distribution network include low-voltage cables, transformers, and circuit breakers; the main power-consuming equipment and new loads in the community include water heaters, smart air conditioners, charging piles, and distributed photovoltaic power generation systems;

[0167] The first-level indicator of the three-level indicator scoring standard for distribution network risk is: distribution network A, and the second-level indicators include cables B1, transformers B2, circuit breakers B3, smart air conditioners B4, water heaters B5, distributed photovoltaic power generation systems B6 and charging piles B7;

[0168] The three-level indicators corresponding to cable B1 include cable temperature C 11 , Cable voltage deviation C 12 、Partial discharge C 13 , Cable operation life C 14 , Cable ambient temperature C 15 and cable ambient humidity C 16 ;

[0169] The three-level indicators corresponding to transformer B2 include oil temperature C 21 、Oil level C 22 , Transformer ambient temperature C 23 , and the voltage deviation C of phase A in the three-phase 24 、B phase voltage deviation C 25 、C phase voltage deviation C 26 、A phase current deviation C 27 、B phase current deviation C 28 and C phase current deviation C 29 ;

[0170] The three-level indicators corresponding to circuit breaker B3 include circuit breaker service life C 31 , breaking times C 32 , Vacuum degree C 33 , opening synchronism C 34 And closing synchronism C 35 ;

[0171] The three-level indicators corresponding to the smart air conditioner B4 include air conditioner voltage deviation C 41 、Air conditioning current deviation C 42 and air conditioner service life C 43 ;

[0172] The three-level indicators corresponding to the water heater B5 include the water heater voltage deviation C 51 , Water heater current deviation C 52 、Service life of water heater C 53 and water heater failure times C 54 ;

[0173] The three-level indicators corresponding to the distributed photovoltaic power generation system B6 include the number of failures of the distributed photovoltaic power generation system C 61 , Distributed photovoltaic power generation system fault repair time C 62 , Common connection point voltage deviation C 63 and the penetration rate of distributed generation C 64 ;

[0174] The three-level indicators corresponding to the charging pile B7 include voltage stabilization accuracy C 71 、Temperature flow accuracy C 72 , current imbalance C 73 , output voltage error C 74 , output current error C 75 , Power Factor C 76 and the service life of the charging pile C 77 ;

[0175] Cable temperature C 11 The unit is: ℃, less than 70 is 4 to 5 points, between 70 and 80 is 2 to 4 points, between 80 and 90 is 0 to 2 points, and more than 90 is 0 points;

[0176] Cable voltage deviation C 12 As a percentage, less than 2 is 4 to 5 points, between 2 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0177] Partial discharge C 13 The unit is: dB, less than 8 is 4 to 5 points, between 8 and 12 is 2 to 4 points, between 12 and 15 is 0 to 2 points, and greater than 15 is 0 points;

[0178] Cable operation life C 14 The unit is: year, less than 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 20 is 0 to 2 points, and more than 20 is 0 points;

[0179] Cable ambient temperature C 15 The unit is: ℃, less than 40 is 4 to 5 points, between 40 and 50 is 2 to 4 points, between 50 and 70 is 0 to 2 points, and more than 70 is 0 points;

[0180] Cable ambient humidity C 16 As a percentage, less than 45 is 4 to 5 points, between 45 and 60 is 2 to 4 points, between 60 and 85 is 0 to 2 points, and more than 85 is 0 points;

[0181] Oil temperature C 21 The unit is: ℃, less than 45 is 4 to 5 points, between 45 and 75 is 2 to 4 points, between 75 and 90 is 0 to 2 points, and more than 90 is 0 points;

[0182] Oil level C 22 As a percentage, more than 80 is 4 to 5 points, between 80 and 60 is 2 to 4 points, between 60 and 40 is 0 to 2 points, and less than 40 is 0 points;

[0183] Transformer ambient temperature C 23The unit is: ℃, between -5 and 40 is 4 to 5 points, between 40 and 50 is 2 to 4 points, between 50 and 60 is 0 to 2 points, and above 60 is 0 points;

[0184] Phase A voltage deviation C 24 、B phase voltage deviation C 25 and C phase voltage deviation C 26 All are percentages, with any value less than 2 being 4 to 5 points, between 2 and 4 being 2 to 4 points, between 4 and 5 being 0 to 2 points, and greater than 5 being 0 points;

[0185] A phase current deviation C 27 、B phase current deviation C 28 and C phase current deviation C 29 All are percentages, with any value less than 4 being 4 to 5 points, between 4 and 7 being 2 to 4 points, between 7 and 10 being 0 to 2 points, and greater than 10 being 0 points;

[0186] Circuit breaker service life C 31 The unit is: year, less than 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 15 is 0 to 2 points, and more than 15 is 0 points;

[0187] Breaking times C 32 The unit is: times, less than 3000 is 4 to 5 points, between 3000 and 6500 is 2 to 4 points, between 6500 and 10000 is 0 to 2 points, and more than 10000 is 0 points;

[0188] Vacuum degree C 33 The unit is Pa, less than 6.6 is 4 to 5 points, between 6.6 and 13.3 is 2 to 4 points, between 13.3 and 66 is 0 to 2 points, and greater than 66 is 0 points;

[0189] Opening synchronism C 34 The unit is ms, less than 0.5 is 4 to 5 points, between 0.5 and 1.2 is 2 to 4 points, between 1.2 and 2.0 is 0 to 2 points, and greater than 2.0 is 0 points;

[0190] Closing synchronism C 35 The unit is: ms, less than 1 is 4 to 5 points, between 1 and 2 is 2 to 4 points, between 2 and 3.5 is 0 to 2 points, and greater than 3.5 is 0 points;

[0191] Air conditioning voltage deviation C 41 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0192] Air conditioning current deviation C 42As a percentage, less than 1 is 4 to 5 points, between 1 and 3.5 is 2 to 4 points, between 3.5 and 5 is 0 to 2 points, and greater than 5 is 0 points;

[0193] Air conditioner service life C 43 The unit is year. Less than or equal to 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 15 is 0 to 2 points, and greater than 15 is 0 points.

[0194] Water heater voltage deviation C 51 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0195] Water heater current deviation C 52 As a percentage, less than 1 is 4 to 5 points, between 1 and 3.5 is 2 to 4 points, between 3.5 and 5 is 0 to 2 points, and greater than 5 is 0 points;

[0196] Water heater service life C 53 The unit is year. Less than 4 is 4 to 5 points. Between 4 and 8 is 2 to 4 points. Between 8 and 12 is 0 to 2 points. More than 12 is 0 points.

[0197] Water heater failure times C 54 The unit is: times / year, less than or equal to 1 is 4 to 5 points, 2 is 2 to 4 points, 3 is 0 to 2 points, and greater than 3 is 0 points;

[0198] Distributed photovoltaic power generation system failure times C 61 The unit is: less than or equal to 1 is 4 to 5 points, 2 is 2 to 4 points, 3 is 0 to 2 points, and greater than 3 is 0 points;

[0199] Distributed photovoltaic power generation system fault repair time C 62 The unit is: h, less than 24 is 4 to 5 points, between 24 and 48 is 2 to 4 points, between 48 and 72 is 0 to 2 points, and more than 72 is 0 points;

[0200] Common connection point voltage deviation C 63 As a percentage, less than 2 is 4 to 5 points, between 2 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0201] Distributed power generation penetration rate C 64 As a percentage, between 20 and 15 is 4 to 5 points, between 15 and 10 is 2 to 4 points, between 10 and 5 is 0 to 2 points, and less than 5 is 0 points;

[0202] Voltage regulation accuracy C 71As a percentage, less than 0.3 is 4 to 5 points, between 0.3 and 0.5 is 2 to 4 points, between 0.5 and 1 is 0 to 2 points, and greater than 1 is 0 points;

[0203] Temperature flow accuracy C 72 As a percentage, less than 0.5 is 4 to 5 points, between 0.5 and 1 is 2 to 4 points, between 1 and 2 is 0 to 2 points, and greater than 2 is 0 points;

[0204] Current imbalance C 73 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points;

[0205] Output voltage error C 74 As a percentage, less than 0.3 is 4 to 5 points, between 0.3 and 0.5 is 2 to 4 points, between 0.5 and 1 is 0 to 2 points, and greater than 1 is 0 points;

[0206] Output current error C 75 As a percentage, less than 0.5 is 4 to 5 points, between 0.5 and 1 is 2 to 4 points, between 1 and 2 is 0 to 2 points, and greater than 2 is 0 points;

[0207] Power Factor C 76 4 to 5 points between 0.95 and 1, 2 to 4 points between 0.9 and 0.95, 0 to 2 points between 0.85 and 0.9, and 0 points for less than 0.85;

[0208] Charging pile service life C 77 The unit is year; less than 3 is 4 to 5 points, between 3 and 6 is 2 to 4 points, between 6 and 10 is 0 to 2 points, and greater than 10 is 0 point.

[0209] The details are shown in the following table:

[0210]

[0211] In some embodiments, step 2 is as follows:

[0212] Step 2.1, perform consistency test on the subjective and objective weights obtained by each of the m third-level evaluation indicators and the n weighting methods;

[0213] If the consistency test passes, the minimum information entropy principle based on Lagrange multiplication optimization is used to optimize the subjective and objective weights, and the combined weight W of the i-th indicator is calculated. i , the specific formula is as follows:

[0214]

[0215] Among them, w i,jis the weight value of the i-th third-level indicator obtained by the j-th evaluation method;

[0216] If the consistency check fails, the improved CRITIC combination method is used to find the coefficients of each weight, and the information carrying capacity in the CRITIC method is optimized according to the information entropy in the entropy weight method, specifically:

[0217] Step 2.1.1: Obtain the comparative performance S of the jth method j and the contradiction between the jth method weight and the other method weights R j , the specific formula is as follows:

[0218]

[0219] Among them, G kj represents the correlation coefficient between the kth method weight and the jth method weight. Here, the Pearson correlation coefficient is used, which is the quotient of the covariance and standard deviation between two variables. The number of weighting methods is n;

[0220] G kj The calculation formula is as follows:

[0221]

[0222] Among them, w i,k is the weight value of the i-th third-level indicator obtained by the k-th evaluation method;

[0223] is the average value of the jth method weight; is the average value of the kth method weight; the calculation formula is as follows:

[0224]

[0225] Step 2.1.2: Introduce the information entropy E corresponding to the index in the entropy weight method j :

[0226]

[0227] Step 2.1.3, considering the impact of variant data on information entropy, the activation entropy is used for optimization, combined with the multiplication model used in the traditional CRITIC method:

[0228] Ij=(1-1 / (1+exp(-Ej)))SjRj;

[0229] Step 2.1.4: Get the combined weight W of the i-th indicator i for:

[0230]

[0231] Step 2.2: According to the scoring criteria, the scoring value is obtained. Combined with the indicator combination weight, the final risk assessment value H2 of the secondary indicator is:

[0232]

[0233] Among them, O i is the score of the i-th third-level indicator; W i is the combined weight of the i-th third-level indicator.

[0234] In some embodiments, step 3 is as follows:

[0235] Step 3.1: Convert the distribution equipment importance language multi-attribute decision-making matrix and the direct impact matrix between indicators into the corresponding cloud matrix, as follows:

[0236] Step 3.1.1, select the language phrase d for the distribution network equipment l The number of is set to 7, so that the language evaluation set D = {d l |d -3 ,d -2 ,d -1 ,d0,d1,d2,d3}={extremely low, very low, low, average, high, very high, extremely high};

[0237] Using θ l represents the importance evaluation of distribution network equipment, then θ l =[0,1], the specific calculation formula is as follows:

[0238]

[0239] Among them, select the language phrase d l When the number of is 7, a≈1.37;

[0240] Step 3.1.2: Given the domain interval, that is, the evaluation score interval U = [X min ,X max ], the language phrase d l Convert to the corresponding forward cloud model A l =C l (Ex l ,En l ,He l ); The digital characteristics of the cloud model can reflect the overall characteristics of the qualitative concept, denoted as C(Ex,En,He);

[0241] Among them, Ex l is the expected distribution of cloud droplets in the domain space;

[0242] En l is entropy, representing the uncertainty and ambiguity of cloud droplet distribution;

[0243] He l is the hyperentropy, which is used to measure the uncertainty of entropy, that is, the entropy of entropy;

[0244] Calculate Expected Ex l The value of is:

[0245] Ex l =X min +θ l (X max -X min );

[0246] Step 3.1.3. Calculate entropy En l The value of is:

[0247]

[0248]

[0249] Step 3.1.4. Calculate the super entropy He l The values ​​are:

[0250]

[0251] In practical applications, if we assume that the given domain is U = [0, 100], then according to steps 3.1.1 to 3.1.4, the seven clouds that can be calculated are:

[0252] C -3 (0,29.65,1.228),C -2 (22.10,26.63,2.234),C -1 (38.23, 21.08, 4.086), C0(50, 19.28, 4.683), C1(61.77, 21.08, 4.086), C2(77.90, 26.63, 2.234) and C3(100, 29.65, 1.228).

[0253] Step 3.1.5, construct the direct impact matrix P based on the cloud model;

[0254]

[0255] Among them, p gh is the degree of influence of the decision maker on the hth indicator on the gth indicator; r is the number of importance indicators;

[0256] Step 3.2: Use the C-DEMATEL method to determine the indicator weights, specifically:

[0257] Step 3.2.1: The cloud matrix P after the direct influence matrix between indicators is converted from step 3.1 = (p gh ) r×r , where the main diagonal element p gg =minA=(Ex -l ,En -l ,He -l ) indicates that the evaluation index has no influence on itself;

[0258] Step 3.2.2: Set the cloud score p gh Convert to score gh And normalize, get the matrix Z = (z gh ) r×r Among them, zy h and z h The details are as follows:

[0259]

[0260] Among them, zy gh ∈[0,1],d(p gh ,maxA) is p gh The Hamming distance between the maximum cloud maxA and d(p gh ,minA) is p gh The Hamming distance between the minimum cloud minA and the minimum cloud minA;

[0261] Step 3.2.3, calculate the Hamming distance between clouds A1 and A2, specifically:

[0262]

[0263] Step 3.2.4, get the comprehensive influence matrix V = (v gh ) r×r ;

[0264] V==Z(IZ) -1 ;

[0265] Where I is the identity matrix;

[0266] Step 3.2.5: Get the weight α of each evaluation index f , specifically:

[0267]

[0268] Among them, β f is the sum of the elements in the fth row of the comprehensive influence matrix V, δ f is the sum of the elements in the fth column of the comprehensive influence matrix V; where α f is the weight of the fth evaluation index;

[0269] Step 3.3, using the CWPHM operator to determine the comprehensive evaluation value of each distribution network equipment;

[0270] For the cloud matrix B after the multi-attribute decision-making matrix transformation, B = (B1, B2, ..., B r ), the corresponding weight vector is α=(α1,α2,…,α r ) T , let CWPHM:B r →B, then

[0271]

[0272] in, Indicates B e and The support degree; Indicates B e and The distance between

[0273] Define the normalized distance between B1 and B2 as:

[0274]

[0275] Step 3.4, quantitative value x∈U, and x~N(Ex,En′ 2 ), En′=N(En,He 2 ), the membership degree of the quantitative value x to the qualitative language concept T satisfies:

[0276]

[0277] The distribution of the membership degree y on the given domain U is called the normal cloud model, and each group (x, y) is represented as a cloud droplet. The score of a cloud droplet is defined as s = xy. Then the mathematical expectation of the score s is It is the total score of the cloud model, i.e., the importance of the distribution network equipment.

[0278] In certain embodiments, in step 4, the steps are as follows:

[0279] Step 4.1: There is an exponential relationship between the risk assessment value of the power distribution equipment and the corresponding probability of successful operation, as follows:

[0280]

[0281] Among them, H2 is the risk assessment value of the power distribution equipment; DB is the probability of successful operation of the power distribution equipment;

[0282] Step 4.2: Starting from the bottom layer of the distribution network hierarchy diagram, all devices in the same layer can be equivalent to one device. For all devices connected in series on the branch line, the probability of successful operation is:

[0283]

[0284] in, is the successful operation probability of the kth device in the j1th branch line of the i1th layer; is the probability of successful operation of all devices in the j1-th branch line of the i1-th layer connected in series; is the importance of the kth device in the j1th branch line of the i1th layer; m1 is the total number of all devices in the j1th branch line of the i1th layer;

[0285] After the devices on the same branch line are connected in series and equivalent, the probability of successful operation of the series equivalent devices connected in parallel on the branch lines on the same layer is:

[0286]

[0287] In the formula, is the probability of successful operation of all devices in layer i1; is the ratio of the importance of all devices in the j1-th branch line of the i1-th layer to the importance of all devices in this layer;

[0288] Step 4.3: Repeat steps 4.1 and 4.2 to obtain the probability of successful operation of the distribution network. By converting the formula, the risk assessment value H1 of the first-level distribution network can be deduced. The specific formula is as follows:

[0289]

[0290] In this step, by introducing the equivalence theory in the GO method for optimizing equipment importance, and based on the relationship between the equipment risk assessment value and the probability of successful operation of the equipment, the risk assessment value of the distribution equipment is derived into the risk assessment value of the distribution network.

[0291] In some embodiments, step 5 is to add a sub-health level on the basis of the four levels of healthy, caution, abnormal and severe in DL / T 2106-2020 "Guidelines for Distribution Network Equipment Status Evaluation", as follows:

[0292] The risk level is healthy, and the corresponding risk assessment value is [4,5]. The representative color is green, and the performance description is that the distribution network or equipment is in excellent condition, there is no abnormality in operation, and the incidence of electrical fire is very low;

[0293] The risk level of sub-health corresponds to a risk assessment value of [3,4], represented by blue, and the performance description is that the distribution network or equipment is in good condition, there may be minor problems, and the incidence of electrical fires is low;

[0294] The risk level is attention, and the corresponding risk assessment value is [2,3], the representative color is yellow, and the performance description is that the distribution network or equipment status is general, there may be some abnormalities, and the incidence of electrical fires is general;

[0295] The risk level is abnormal, and the corresponding risk assessment value is [1,2]. The representative color is orange, and the performance description is that the distribution network or equipment is in poor condition, there are many abnormalities, and the incidence of electrical fires is high;

[0296] The risk level is severe and the corresponding risk assessment value is [0,1], the representative color is red, and the performance description is that the distribution network or equipment is in a very poor state, there are many anomalies, and the incidence of electrical fires is very high.

[0297] The details are as follows:

[0298] Risk Level Risk Assessment Value Representative Color Performance Description healthy [4,5] green The distribution network or equipment is in excellent condition, operating normally, and the incidence of electrical fires is very low. Sub-health [3，4] blue The distribution network or equipment is in good condition, with possible minor problems and low incidence of electrical fires Notice [2,3] yellow The distribution network or equipment is in normal condition, there may be some abnormalities, and the incidence of electrical fires is normal abnormal [1，2] orange color The distribution network or equipment is in poor condition, with many abnormalities and a high incidence of electrical fires serious [0,1] red The distribution network or equipment is in poor condition, with many anomalies and a high incidence of electrical fires

[0299] Example 2

[0300] Taking the 10kV distribution network data of a certain community as an example, the simplified network topology diagram is as follows: Figure 3 As shown, it contains 4 low-voltage lines, 4 transformers, 2 circuit breakers, 1 distributed photovoltaic power generation system (PV), 2 charging piles (M9, M10), 5 water heaters and 3 smart air conditioners (M2, M5, M8). Select the distribution network related data of 4 years ago (Example 1), 2 years ago (Example 2) and now (Example 3) to obtain the risk assessment value of distribution network equipment as follows: Figure 4 The distribution network risk assessment values ​​of the three examples before and after equipment importance optimization are shown in the following table:

[0301] category Example 1 Example 2 Example 3 Risk assessment value before optimization 3.5339 3.0322 2.3094 Optimized risk assessment value 3.6502 3.3084 2.8356 Risk Sub-health Sub-health Notice

[0302] From the three examples in the table above, we can see that the risk levels of Example 1, Example 2, and Example 3 before and after the optimization of the GO method using the equipment importance are "sub-healthy", "sub-healthy", and "caution", respectively, and as the operating time increases, the overall risk assessment value of the distribution network decreases. The risk assessment value of the distribution network before optimization is low, especially in Example 3, the difference between the risk assessment values ​​of the distribution network before and after optimization is large, with a difference of 0.5262. This is because when the equipment on the same branch line is equivalent, the probability of successful operation of all equipment follows a multiplication function relationship, which makes the equipment with a lower risk assessment value have a greater impact on the overall system. Since the risk assessment value of cable L1 is only 1.5626, the overall risk assessment value of the distribution network before optimization is seriously lowered. The method in this paper optimizes this shortcoming and is more in line with the actual risk level of the distribution network.

[0303] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for assessing electrical fire risks in medium and low voltage distribution networks; characterized in that: The steps include: Step 1: Construct a hierarchical and multi-dimensional distribution network electrical fire risk assessment index system including distribution equipment, lines, power loads and new loads; Step 2: Based on the order relationship analysis (G1) method and the entropy weight method, the information entropy is introduced to optimize the information carrying capacity in the CRITIC method, and then the algorithm for obtaining the weights of each indicator is based on the improved CRITIC combination method. The Kendall consistency coefficient is used to test the subjective and objective weights to obtain the risk assessment value of the distribution network equipment. Step 3: Use the Cloud DEMATEL device importance algorithm based on the CWPHM operator to calculate the importance of each distribution network device; Step 4: According to the network topology and the equipment risk assessment value, the risk assessment value of the distribution network is solved by using the equipment importance improved success flow (GO) method; Step 5: Divide the status of distribution equipment and distribution network risk into five levels: healthy, sub-healthy, caution, abnormal and severe.

2. The method for assessing electrical fire risk in medium and low voltage distribution networks according to claim 1, characterized in that: In step 1, the power load and new load are water heaters, smart air conditioners, charging piles and distributed photovoltaic power generation systems. The specific steps are as follows: Step 1.1, select the main distribution equipment in the distribution network and the main power equipment and new loads in the community as secondary indicators; Step 1.2, according to the operation characteristics of the secondary equipment corresponding to the secondary indicators, and taking into account weather environment factors and fault characteristic factors, select the third-level indicators corresponding to each of the secondary equipment; Step 1.3: According to DL / T 2106-2020 "Guidelines for Distribution Network Equipment Status Evaluation" and DB11 / T2036-2022 "Technical Specifications for Electrical Safety of Distributed Photovoltaic Power Generation Systems", a three-level distribution network risk index scoring standard is constructed as the hierarchical and multi-dimensional distribution network electrical fire risk assessment index system.

3. The method for assessing electrical fire risk in medium and low voltage distribution networks according to claim 2, characterized in that: In step 1.1, the main power distribution equipment and distribution lines in the distribution network include low-voltage cables, transformers and circuit breakers; the main power-consuming equipment and new loads in the community include water heaters, smart air conditioners, charging piles and distributed photovoltaic power generation systems; The first-level indicator of the three-level indicator scoring standard for distribution network risk is: distribution network A, and the second-level indicators include cables B1, transformers B2, circuit breakers B3, smart air conditioners B4, water heaters B5, distributed photovoltaic power generation systems B6 and charging piles B7; The three-level indicators corresponding to the cable B1 include cable temperature C 11 , Cable voltage deviation C 12 、Partial discharge C 13 , Cable operation life C 14 , Cable ambient temperature C 15 and cable ambient humidity C 16 ; The three-level indicators corresponding to transformer B2 include oil temperature C 21 、Oil level C 22 , Transformer ambient temperature C 23 , and the voltage deviation C of phase A in the three-phase 24 、B phase voltage deviation C 25 、C phase voltage deviation C 26 、A phase current deviation C 27 、B phase current deviation C 28 and C phase current deviation C 29 ; The three-level indicators corresponding to the circuit breaker B3 include the service life of the circuit breaker C 31 , breaking times C 32 , Vacuum degree C 33 , opening synchronism C 34 And closing synchronism C 35 ; The three-level indicators corresponding to the smart air conditioner B4 include air conditioner voltage deviation C 41 , Air conditioning current deviation C 42 and air conditioner service life C 43 ; The three-level indicators corresponding to the water heater B5 include the water heater voltage deviation C 51 , Water heater current deviation C 52 、Service life of water heater C 53 and water heater failure times C 54 ; The three-level indicators corresponding to the distributed photovoltaic power generation system B6 include the number of failures of the distributed photovoltaic power generation system C 61 , Distributed photovoltaic power generation system fault repair time C 62 , Common connection point voltage deviation C 63 and the penetration rate of distributed generation C 64 ; The three-level indicators corresponding to the charging pile B7 include voltage stabilization accuracy C 71 、Temperature flow accuracy C 72 , current imbalance C 73 , output voltage error C 74 , output current error C 75 , Power Factor C 76 and the service life of the charging pile C 77 ; The cable temperature C 11 The unit is: ℃, less than 70 is 4 to 5 points, between 70 and 80 is 2 to 4 points, between 80 and 90 is 0 to 2 points, and more than 90 is 0 points; The cable voltage deviation C 12 As a percentage, less than 2 is 4 to 5 points, between 2 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points; The partial discharge C 13 The unit is: dB, less than 8 is 4 to 5 points, between 8 and 12 is 2 to 4 points, between 12 and 15 is 0 to 2 points, and greater than 15 is 0 points; The cable service life C 14 The unit is: year, less than 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 20 is 0 to 2 points, and more than 20 is 0 points; The cable ambient temperature C 15 The unit is: ℃, less than 40 is 4 to 5 points, between 40 and 50 is 2 to 4 points, between 50 and 70 is 0 to 2 points, and more than 70 is 0 points; The cable environment humidity C 16 As a percentage, less than 45 is 4 to 5 points, between 45 and 60 is 2 to 4 points, between 60 and 85 is 0 to 2 points, and more than 85 is 0 points; The oil temperature C 21 The unit is: ℃, less than 45 is 4 to 5 points, between 45 and 75 is 2 to 4 points, between 75 and 90 is 0 to 2 points, and more than 90 is 0 points; The oil level C 22 As a percentage, more than 80 is 4 to 5 points, between 80 and 60 is 2 to 4 points, between 60 and 40 is 0 to 2 points, and less than 40 is 0 points; The transformer ambient temperature C 23 The unit is: ℃, between -5 and 40 is 4 to 5 points, between 40 and 50 is 2 to 4 points, between 50 and 60 is 0 to 2 points, and above 60 is 0 points; The A phase voltage deviation C 24 , the B phase voltage deviation C 25 and the C phase voltage deviation C 26 All are percentages, with any value less than 2 being 4 to 5 points, between 2 and 4 being 2 to 4 points, between 4 and 5 being 0 to 2 points, and greater than 5 being 0 points; The A phase current deviation C 27 、The B phase current deviation C 28 and the C phase current deviation C 29 All are percentages, with any value less than 4 being 4 to 5 points, between 4 and 7 being 2 to 4 points, between 7 and 10 being 0 to 2 points, and greater than 10 being 0 points; The service life of the circuit breaker C 31 The unit is: year, less than 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 15 is 0 to 2 points, and more than 15 is 0 points; The breaking times C 32 The unit is: times, less than 3000 is 4 to 5 points, between 3000 and 6500 is 2 to 4 points, between 6500 and 10000 is 0 to 2 points, and more than 10000 is 0 points; The vacuum degree C 33 The unit is Pa, less than 6.6 is 4 to 5 points, between 6.6 and 13.3 is 2 to 4 points, between 13.3 and 66 is 0 to 2 points, and greater than 66 is 0 points; The opening synchronism C 34 The unit is ms, less than 0.5 is 4 to 5 points, between 0.5 and 1.2 is 2 to 4 points, between 1.2 and 2.0 is 0 to 2 points, and greater than 2.0 is 0 points; The closing synchronism C 35 The unit is: ms, less than 1 is 4 to 5 points, between 1 and 2 is 2 to 4 points, between 2 and 3.5 is 0 to 2 points, and greater than 3.5 is 0 points; The air conditioning voltage deviation C 41 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points; The air conditioning current deviation C 42 As a percentage, less than 1 is 4 to 5 points, between 1 and 3.5 is 2 to 4 points, between 3.5 and 5 is 0 to 2 points, and greater than 5 is 0 points; The service life of the air conditioner C 43 The unit is year. Less than or equal to 5 is 4 to 5 points, between 5 and 10 is 2 to 4 points, between 10 and 15 is 0 to 2 points, and greater than 15 is 0 points. The water heater voltage deviation C 51 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points; The water heater current deviation C 52 As a percentage, less than 1 is 4 to 5 points, between 1 and 3.5 is 2 to 4 points, between 3.5 and 5 is 0 to 2 points, and greater than 5 is 0 points; The service life of the water heater C 53 The unit is year. Less than 4 is 4 to 5 points. Between 4 and 8 is 2 to 4 points. Between 8 and 12 is 0 to 2 points. More than 12 is 0 points. The water heater failure frequency C 54 The unit is: times / year, less than or equal to 1 is 4 to 5 points, 2 is 2 to 4 points, 3 is 0 to 2 points, and greater than 3 is 0 points; The number of failures of the distributed photovoltaic power generation system C 61 The unit is: less than or equal to 1 is 4 to 5 points, 2 is 2 to 4 points, 3 is 0 to 2 points, and greater than 3 is 0 points; The distributed photovoltaic power generation system fault repair time C 62 The unit is: h, less than 24 is 4 to 5 points, between 24 and 48 is 2 to 4 points, between 48 and 72 is 0 to 2 points, and more than 72 is 0 points; The common connection point voltage deviation C 63 As a percentage, less than 2 is 4 to 5 points, between 2 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points; The distributed generation penetration rate C 64 As a percentage, between 20 and 15 is 4 to 5 points, between 15 and 10 is 2 to 4 points, between 10 and 5 is 0 to 2 points, and less than 5 is 0 points; The voltage stabilization accuracy C 71 As a percentage, less than 0.3 is 4 to 5 points, between 0.3 and 0.5 is 2 to 4 points, between 0.5 and 1 is 0 to 2 points, and greater than 1 is 0 points; The temperature flow accuracy C 72 As a percentage, less than 0.5 is 4 to 5 points, between 0.5 and 1 is 2 to 4 points, between 1 to 2 is 0 to 2 points, and greater than 2 is 0 points; The current imbalance degree C 73 As a percentage, less than 3 is 4 to 5 points, between 3 to 5 is 2 to 4 points, between 5 to 7 is 0 to 2 points, and greater than 7 is 0 points; The output voltage error C 74 As a percentage, less than 0.3 is 4 to 5 points, between 0.3 and 0.5 is 2 to 4 points, between 0.5 and 1 is 0 to 2 points, and greater than 1 is 0 points; The output current error C 75 As a percentage, less than 0.5 is 4 to 5 points, between 0.5 and 1 is 2 to 4 points, between 1 to 2 is 0 to 2 points, and greater than 2 is 0 points; The power factor C 76 4 to 5 points between 0.95 and 1, 2 to 4 points between 0.9 and 0.95, 0 to 2 points between 0.85 and 0.9, and 0 points for less than 0.85; The service life of the charging pile is C 77 The unit is year; less than 3 is 4 to 5 points, between 3 and 6 is 2 to 4 points, between 6 and 10 is 0 to 2 points, and greater than 10 is 0 point.

4. The method for assessing electrical fire risk in medium and low voltage distribution networks according to claim 4, characterized in that: Step 2 is as follows: Step 2.1, performing consistency check on the subjective and objective weights obtained by each of the m third-level evaluation indicators and the n weighting methods; If the consistency test passes, the minimum information entropy principle based on Lagrange multiplication optimization is used to optimize the subjective and objective weights, and the combined weight W of the i-th indicator is calculated. i , the specific formula is as follows: Among them, w i,j is the weight value of the i-th third-level indicator obtained by the j-th evaluation method; If the consistency test fails, the improved CRITIC combination method is used to find the coefficients of each weight, and the information carrying capacity in the CRITIC method is optimized according to the information entropy in the entropy weight method. Assuming that there are m evaluation indicators and n evaluation methods, they are as follows: Step 2.1.1: Obtain the comparative performance S of the jth method j and the contradiction between the jth method weight and the other method weights R j , the specific formula is as follows: Among them, G kj represents the correlation coefficient between the kth method weight and the jth method weight. Here, the Pearson correlation coefficient is used, which is the quotient of the covariance and standard deviation between two variables. The number of weighting methods is n; G kj The calculation formula is as follows: Among them, w i,k is the weight value of the i-th third-level indicator obtained by the k-th evaluation method; is the average value of the jth method weight; is the average value of the kth method weight; the calculation formula is as follows: Step 2.1.2: Introduce the information entropy E corresponding to the index in the entropy weight method j : Step 2.1.3, considering the impact of variant data on information entropy, the activation entropy is used for optimization, combined with the multiplication model used in the traditional CRITIC method: I j =(1-1 / (1+exp(-E j )))S j R j ; Step 2.1.4: Get the combined weight W of the i-th indicator i for: Step 2.2: According to the scoring criteria, the scoring value is obtained. Combined with the indicator combination weight, the final risk assessment value H2 of the secondary indicator is: Among them, O i is the score of the i-th third-level indicator; W i is the combined weight of the i-th third-level indicator.

5. The method for assessing electrical fire risk in medium and low voltage distribution networks according to claim 1, characterized in that: Step 3 is as follows: Step 3.1: Convert the distribution equipment importance language multi-attribute decision-making matrix and the direct impact matrix between indicators into the corresponding cloud matrix, as follows: Step 3.1.1, select the language phrase d for the distribution network equipment l The number of is set to 7, so that the language evaluation set D = {d l |d -3 ,d -2 ,d -1 ,d0,d1,d2,d3}={extremely low, very low, low, average, high, very high, extremely high}; Using θ l represents the importance evaluation of distribution network equipment, then θ l =[0,1], the specific calculation formula is as follows: Among them, select the language phrase d l When the number of is 7, a≈1.37; Step 3.1.2: Given the domain interval, that is, the evaluation score interval U = [X min ,X max ], the language phrase d l Convert to the corresponding forward cloud model A l =C l (Ex l ,En l ,He l ); The digital characteristics of the cloud model can reflect the overall characteristics of the qualitative concept, denoted as C(Ex,En,He); Among them, Ex l is the expected distribution of cloud droplets in the domain space; En l is entropy, representing the uncertainty and ambiguity of cloud droplet distribution; He l is the hyperentropy, which is used to measure the uncertainty of entropy, that is, the entropy of entropy; Calculate Expected Ex l The value of is: Ex l =X min +θ l (X max -X min ); Step 3.1.

3. Calculate entropy En l The value of is: Step 3.1.

4. Calculate the excess entropy He l The values ​​are: In practical applications, if we assume that the given domain is U = [0, 100], then according to steps 3.1.1 to 3.1.4, the seven clouds that can be calculated are: Step 3.1.5, construct the direct impact matrix P based on the cloud model; Among them, p gh is the degree of influence of the decision maker on the hth indicator on the gth indicator; r is the number of importance indicators; Step 3.2: Use the C-DEMATEL method to determine the indicator weights, specifically: Step 3.2.1: The cloud matrix P after the direct influence matrix between indicators is converted from step 3.1 = (p gh ) r×r , where the main diagonal element p gg =minA=(Ex -l ,En -l ,He -l ) indicates that the evaluation index has no influence on itself; Step 3.2.2: Set the cloud score p gh Convert to score gh And normalize, get the matrix Z = (z gh ) r×r Among them, zy gh and z gh The details are as follows: Among them, zy gh ∈[0,1],d(p gh ,maxA) is p gh The Hamming distance between the maximum cloud maxA and d(p gh ,minA) is p gh The Hamming distance between the minimum cloud minA and the minimum cloud minA; Step 3.2.3, calculate the Hamming distance between clouds A1 and A2, specifically: Step 3.2.4, get the comprehensive influence matrix V = (v gh ) r×r ; V==Z(I-Z) -1 ; Where I is the identity matrix; Step 3.2.5: Get the weight α of each evaluation index f , specifically: Among them, β f is the sum of the elements in the fth row of the comprehensive influence matrix V, δ f is the sum of the elements in the fth column of the comprehensive influence matrix V; where α f is the weight of the fth evaluation index; Step 3.3, using the CWPHM operator to determine the comprehensive evaluation value of each distribution network equipment; For the cloud matrix B after the multi-attribute decision-making matrix transformation, B = (B1, B2, ..., B r ), the corresponding weight vector is α=(α1,α2,…,α r ) T , let CWPHM:B r →B, then in, Indicates B e and The support degree; Indicates B e and The distance between Define the normalized distance between B1 and B2 as: Step 3.4, quantitative value x∈U, and x~N(Ex,En′ 2 ), En′=N(En,He 2 ), the membership degree of the quantitative value x to the qualitative language concept T satisfies: The distribution of the membership degree y on the given domain U is called the normal cloud model, and each group (x, y) is represented as a cloud droplet. The score of a cloud droplet is defined as s = xy. Then the mathematical expectation of the score s is It is the total score of the cloud model, i.e., the importance of the distribution network equipment.

6. The method for assessing electrical fire risk in medium and low voltage distribution networks according to claim 5, characterized in that: In step 4, the steps are as follows: Step 4.1: There is an exponential relationship between the risk assessment value of the power distribution equipment and the corresponding probability of successful operation, as follows: Among them, H2 is the risk assessment value of the power distribution equipment; DB is the probability of successful operation of the power distribution equipment; Step 4.2: Starting from the bottom layer of the distribution network hierarchy diagram, all devices in the same layer can be equivalent to one device. For all devices connected in series on the branch line, the probability of successful operation is: in, is the successful operation probability of the kth device in the j1th branch line of the i1th layer; DB i1,j1 is the probability of successful operation of all devices in the j1-th branch line of the i1-th layer connected in series; is the importance of the kth device in the j1th branch line of the i1th layer; m1 is the total number of all devices in the j1th branch line of the i1th layer; After the devices on the same branch line are connected in series and equivalent, the probability of successful operation of the series equivalent devices connected in parallel on the branch lines on the same layer is: In the formula, DB i1 is the probability of successful operation of all devices in layer i1; b i1,j1 is the ratio of the importance of all devices in the j1-th branch line of the i1-th layer to the importance of all devices in this layer; Step 4.3: Repeat steps 4.1 and 4.2 to obtain the distribution network's successful operation probability DB. 11 , by converting the formula, the risk assessment value H1 of the first-level distribution network can be deduced. The specific formula is as follows:

7. The method for assessing electrical fire risk in medium and low voltage distribution networks according to claim 6, characterized in that: Step 5 is to add a sub-health level to the four levels of healthy, caution, abnormal and severe in DL / T 2106-2020 "Guidelines for Distribution Network Equipment Status Evaluation", as follows: The risk level is the risk assessment value corresponding to the health status [4,5], the representative color is green, and the performance description is that the distribution network or equipment is in excellent condition, there is no abnormality in operation, and the incidence of electrical fire is very low; The risk level is the risk assessment value corresponding to the sub-health mentioned above is [3,4], the representative color is blue, and the performance description is that the distribution network or equipment is in good condition, there may be minor problems, and the incidence of electrical fires is low; The risk level is the risk assessment value corresponding to the above-mentioned attention is [2,3], the representative color is yellow, the performance description is that the distribution network or equipment status is general, there may be some abnormalities, and the incidence of electrical fires is general; The risk level is the risk assessment value corresponding to the anomaly, which is [1,2], and the representative color is orange. The performance description is that the distribution network or equipment is in poor condition, there are many anomalies, and the incidence of electrical fires is high; The risk level is severe and the corresponding risk assessment value is [0,1], the representative color is red, and the performance description is that the distribution network or equipment is in a very poor state, there are many anomalies, and the incidence of electrical fires is very high.

Images