Power equipment optimal operation life decision-making method based on technical economy

Abstract

The invention discloses a power equipment optimal operation life decision-making method based on technical economy, belongs to the technical field of renewable energy sources, and aims to solve the problem that economy and technicality are not combined when the operation life of power equipment is calculated in the prior art. Constructing a static economic life calculation model of the power equipment by adopting an output result of a full-period cost calculation module in combination with a constant-speed degradation trend of the operation and maintenance cost and a fault probability of the maintenance cost; constructing a dynamic economic life measuring and calculating model of the power equipment by combining an output result of the static economic life measuring and calculating model with the time value of the fund; outputting a dynamic economic life measurement result; constructing a technical life prediction model of the power equipment based on an HGWO-LS-SVM algorithm, and outputting a technical life prediction result; and obtaining the optimal operation life of the power equipment according to the dynamic economic life measurement result and the technical life prediction result. The method is used for calculating power equipment service life.

Description

Technical Field

[0001] The invention relates to a technical-economic-based decision method for optimal operating life of electric power equipment, and belongs to the technical field of renewable energy. Background Art

[0002] The development of renewable energy such as wind power and photovoltaics plays an important role in reducing carbon emissions. However, the randomness of renewable energy such as wind power and photovoltaics and flexible loads such as electric vehicles affects the stable operation of the entire power system and brings new challenges to the investment and management of the power grid. Therefore, scientifically evaluating the operating life of power equipment and reasonably determining the timing of equipment replacement are the solid foundation and safety guarantee for building a new power system.

[0003] In the study of power equipment quality, the evolution of equipment performance has received extensive attention. Due to the wear of parts, equipment will continue to deteriorate during operation, affecting normal production activities. The main factors causing distribution network equipment failures include: material aging, external force damage, environmental impact, etc.

[0004] The existing methods for evaluating the quality of power equipment include:

[0005] In the prior art, a Gaussian cloud distribution network equipment power supply quality evaluation model is proposed, and the Technique for Order Preference by Similarity to an Ideal Solution is used for sorting. There is also a method that uses the classic bathtub curve to characterize the probability of equipment failure, and designs an electrical equipment reliability optimization method that can enhance equipment reliability. In addition, a power transformer degradation and maintenance model is established, and a bathtub curve is also used to describe the probability of equipment failure, focusing on analyzing the wear tail of the curve to study the accelerated degradation process. However, the problems with this type of method are: in the bathtub curve study, the failure rate function is discontinuous and non-smooth, and the distribution shape parameters are not fixed.

[0006] The economic life of equipment is related to the economic benefits and operating efficiency of the power grid. The economic life is the time when the average annual cost of the equipment is the lowest. When calculating the economic life of equipment, the full life cycle theory needs to be applied. The existing technology calculates the full life cycle cost and total cost of ownership of the transformer to improve the economic efficiency of the equipment. There is also a model for the economic life of equipment that finds the year with the lowest total discounted cost. However, these studies did not consider the impact of the time value of money when calculating the economic life, and ignored the impact of equipment degradation on operation and maintenance costs.

[0007] Equipment quality evaluation and technical life prediction are of great significance to the safe and reliable operation of power grids. In the existing technology, a comprehensive evaluation index system for distribution network equipment is designed from five aspects: equipment economy, performance, intelligence, coordination, and standardization. The index weights are determined by subjective and objective weighted methods, and the equipment quality level can be finally evaluated. Other technologies use fuzzy expert estimation and fuzzy set theory to evaluate the health status of power grid equipment. In addition, some methods integrate support vector regression and gray wolf algorithm to predict the future operating status of distribution transformers. In addition, based on the analysis of the causes of equipment deterioration, an improved time-varying prediction function is proposed to evaluate the future fault status of the equipment. In addition, the failure probability and expected life of the transformer are studied from both economic and technical aspects. These methods do not consider the influence of external environment such as temperature and humidity when predicting equipment quality.

[0008] Therefore, existing studies have the following problems when calculating economic life:

[0009] 1. The existing technology does not comprehensively consider the time value of money, equipment degradation rate, and equipment failure rate, which may easily lead to the problem that the calculated results are inconsistent with the actual situation.

[0010] 2. When predicting the technical life, the existing technology ignores the impact of external environment such as temperature and humidity on equipment quality.

[0011] 3. The selection of equipment replacement timing in the prior art often fails to take into account both economic and safety aspects. Summary of the invention

[0012] The purpose of the present invention is to solve the problem that the economic and technical aspects are not combined when calculating the operating life of power equipment in the prior art, and to provide a method for deciding the optimal operating life of power equipment based on technical and economic aspects.

[0013] The present invention provides a method for determining the optimal operating life of electric power equipment based on technical and economic performance, which comprises:

[0014] S1. Construct a full life cycle cost calculation module for power equipment; use the output results of the full life cycle cost calculation module combined with the uniform degradation trend of operation and maintenance costs and the failure probability of maintenance costs to construct a static economic life calculation model for power equipment; use the output results of the static economic life calculation model combined with the time value of money to construct a dynamic economic life calculation model for power equipment; output the dynamic economic life calculation results;

[0015] S2. Construct a technical life prediction model for power equipment based on the HGWO-LS-SVM algorithm and output the technical life prediction results;

[0016] S3. Obtain the optimal operating life of power equipment based on the dynamic economic life calculation results and technical life prediction results.

[0017] Preferably, the specific method of constructing the technical life prediction model of the power equipment based on the HGWO-LS-SVM algorithm and outputting the technical life prediction result in S2 includes:

[0018] S2-1. Evaluate the quality of scrapped equipment, use the kernel density estimation (KDE) method to fit the quality score with the equipment life, obtain the quality score corresponding to the critical scrap point, and obtain the kernel function;

[0019] S2-2, based on the least squares support vector machine LS-SVM, the technical life prediction model of power equipment is constructed, and the differential evolution algorithm DE is introduced into the grey wolf optimization algorithm GWO to obtain the improved grey wolf algorithm HGWO;

[0020] S2-3, using the improved grey wolf algorithm HGWO to optimize the penalty factor in LS-SVM and the kernel function obtained in S2-1 to obtain the quality scores of power equipment at different periods;

[0021] S2-4. Compare the quality scores of different periods obtained in S2-3 with the quality scores corresponding to the critical scrap points obtained in S2-1. When the quality scores of different periods are lower than the quality scores corresponding to the critical scrap points, the corresponding time is the technical life prediction result.

[0022] Preferably, the specific method of constructing the full life cycle cost estimation module of the power equipment in S1 includes:

[0023] Life cycle cost LCC for:

[0024]

[0025] Among them, C I represents the initial investment cost, C O,n represents the operation and maintenance cost in the nth year, C F,n represents the failure cost in the nth year, L R Represents the residual price of equipment, j=1,2,…,n.

[0026] Preferably, the specific method of constructing the static economic life calculation model of the power equipment in S1 includes:

[0027] The average one-time cost C in the nth year I-R,n yes:

[0028]

[0029] Assuming that the equipment deteriorates at a uniform rate and the operation and maintenance cost increases year by year, the average annual operation and maintenance cost C in the nth year is O,n yes:

[0030]

[0031] Among them, C O,1 represents the initial value of the operation and maintenance cost, and α represents the incremental operation and maintenance cost caused by degradation;

[0032] The average annual maintenance cost C in the nth year F,n yes:

[0033]

[0034] in, represents the average annual periodic maintenance cost in year n, represents the average annual fault repair cost in year n;

[0035]

[0036] in, represents the initial value of the periodic maintenance cost, and β represents the incremental periodic maintenance cost caused by deterioration;

[0037]

[0038] in, represents the cost of each fault repair, and h(n) represents the failure rate in the “bathtub curve”;

[0039] The static economic life calculation model is:

[0040] C a (n) = C I-R,n +C O,n +C F,n ;

[0041] N0=[minC a (n)] -1 ;

[0042] Among them, N0 represents the static economic life, C a (n) represents the static average annual cost in year n.

[0043] Preferably, the specific method of constructing the dynamic economic life calculation model of power equipment in S1 includes:

[0044] The dynamic economic life N0′ is:

[0045] N0′=[minC a ′(n)] -1 ;

[0046] Among them, C a ′(n) represents the dynamic average annual cost in the nth year;

[0047] C a′(n)=C I ' -R,n +C′ O,n +C′ F,n ;

[0048] Among them, C I ' -R,n represents the dynamic investment cost in the nth year, C′ O,n represents the dynamic operation and maintenance cost in the nth year, C′ F,n represents the dynamic cost of failure in year n;

[0049] C I ' -R,n =C I (A / P,i,n)-L R (A / F,i,n);

[0050] Where, (A / P,i,n) represents the equal installment capital recovery factor, (A / F,i,n) represents the equal installment debt repayment fund factor, and i represents the basic internal rate of return;

[0051]

[0052] Among them, C O,1 represents the operation and maintenance cost in the first year, (P / F,i,n) represents the one-time payment present value coefficient;

[0053] in:

[0054]

[0055] Preferably, the specific method of fitting the quality score with the equipment life by using the kernel density estimation (KDE) method in S2-1 to obtain the quality score corresponding to the critical scrap point and to obtain the kernel function includes:

[0056] Kernel density estimation KDE is used to fit the quality score with the equipment life to obtain the quality score corresponding to the critical scrap point. The form of kernel density estimation is:

[0057]

[0058] in, represents the KDE fit with respect to The probability density function of represents a random variable used to represent the quality score, N represents the number of samples, h represents the bandwidth, K h represents the kernel function, represents the sample observation value, R d represents a d-dimensional real number set, k represents the kth sample;

[0059] The kernel function adopts Gaussian kernel function, probability density function It is expressed as:

[0060]

[0061] Preferably, the specific method of constructing a technical life prediction model for power equipment based on the least squares support vector machine LS-SVM described in S2-2 includes:

[0062] The LS-SVM model that extends the support vector machine SVM is proposed:

[0063] First, map the sample from the original space to the feature space:

[0064]

[0065] y represents the output vector, x represents the non-input vector, represents the mapping function, ω and b represent parameters to be determined;

[0066] The determination process of ω and b is equivalent to the minimization of the intermediate variable R:

[0067]

[0068] Among them, ||ω|| 2 represents the complexity of the control model, c represents the adjustment parameter, and controls the degree of penalty for exceeding the error. emp represents the error control function;

[0069] The minimization problem is further expressed as:

[0070]

[0071] J(ω,ξ) represents the loss function, l represents the total number of samples, j(x p ) represents the mapping function of the pth input vector, w represents the complexity parameter, x p represents the pth input vector, ξ p represents the relaxation factor;

[0072] Construct the Lagrangian function L(·):

[0073]

[0074] Among them, α p represents the Lagrange multiplier;

[0075] According to the KKT optimization conditions, we can get:

[0076]

[0077] Eliminating ω and ξ, we get:

[0078]

[0079] Where I = [1,…,1], Use kernel function instead; x m represents the mth vector, x n represents the nth vector;

[0080] The regression function f(x) is:

[0081]

[0082] Among them, K(x,x p ) represents x and x p The kernel function of

[0083] The radial basis function RBF is expressed as:

[0084]

[0085] Among them, x h represents the hth vector, x g represents the gth vector; σ is the kernel width.

[0086] Preferably, the specific method of introducing the differential evolution algorithm DE into the grey wolf optimization algorithm GWO in S2-2 to obtain the improved grey wolf algorithm HGWO includes:

[0087] First, DE is used to achieve population variation, then the optimal three individual positions are obtained based on fitness, and the distance between the remaining gray wolves and the optimal gray wolf is calculated:

[0088] D a =|C1X a (t)-X(t)|

[0089] D b =|C2X b (t)-X(t)|

[0090] D c =|C3X c (t)-X(t)|

[0091] Where t represents the number of iterations, D a , D b and D c Represent distances respectively; C1, C2 and C3 all represent random vectors, X a , X b and X c represents the three optimal positions closest to the prey, and X represents the current solution position;

[0092] The remaining gray wolves were repositioned according to the three optimal positions mentioned above:

[0093] X1(t+1)=X a (t)-A1D a

[0094] X2(t+1)=X b (t)-A2D b

[0095] X3(t+1)=X c (t)-A3D c

[0096] A1, A2 and A3 all represent convergence factors;

[0097] The final positions of the remaining individuals are:

[0098]

[0099] C = 2r1;

[0100] A = 2ar2-a;

[0101] Among them, C represents the perturbation parameter, A represents the convergence factor, r1 and r2 represent random numbers in [0,1], and a represents a variable related to the number of iterations, decreasing from 2 to 0;

[0102] Keep the better individuals and update A and C until the conditions are met.

[0103] Preferably, the specific method of obtaining the quality scores of the power equipment at different periods in S2-3 and the method of obtaining the quality scores of the power equipment at different periods in S2-4, when the quality scores at different periods are lower than the quality scores corresponding to the critical scrapping points, the corresponding time is the technical life prediction result, includes:

[0104] Map the quality-related data to the hidden space through the LS-SVM kernel;

[0105] The HGWO algorithm is applied to obtain the optimal values ​​of the adjustment parameter c and the kernel width σ;

[0106] After inputting into LS-SVM, the predicted values ​​of equipment quality scores at different periods were obtained;

[0107] When the quality scores of different periods are lower than the quality scores corresponding to the critical scrapping point, the corresponding time is the technical life prediction result:

[0108] M0=(if S(n)<S) -1 ;

[0109] Among them, M0 represents the technical life prediction result, S(n) represents the predicted value of the equipment quality score in the nth year, and S represents the quality score corresponding to the critical scrap point.

[0110] Preferably, the specific method for obtaining the optimal operating life of the power equipment according to the dynamic economic life calculation results and the technical life prediction results described in S3 includes:

[0111] The optimal operating life R of power equipment is:

[0112] R = min(N0′,M0);

[0113] Among them, N0′ represents the dynamic economic life, and M0 represents the technical life prediction result.

[0114] Advantages of the present invention: This application combines the economic and technical aspects of power equipment operation and proposes an optimal operating life decision model for power equipment. It combines the uniform degradation trend of operation and maintenance costs and the failure trend of maintenance costs, and takes into account the time value of money to build a dynamic economic life calculation model. It combines internal and external factors, applies kernel density estimation to fit the equipment score probability density function, builds a technical life prediction model based on the least squares support vector machine (LS-SVM), and uses the improved gray wolf optimization algorithm to determine parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0115] Figure 1 It is a flowchart of the method for deciding the optimal operating life of electric power equipment based on technical and economic efficiency according to the present invention;

[0116] Figure 2 It is a flowchart of constructing a dynamic economic life estimation model for power equipment according to the present invention;

[0117] Figure 3 is a graph of the static economic life of the present invention;

[0118] Figure 4 is the probability density function of the device quality score;

[0119] Figure 5 It is a flowchart of constructing a technical life prediction model of power equipment based on the HGWO-LS-SVM algorithm according to the present invention;

[0120] Figure 6 This is a schematic diagram of equipment quality scores in each quarter from 2010 to 2020;

[0121] Figure 7 It is a schematic diagram of reliability curve;

[0122] Figure 8 is a schematic diagram of the failure rate curve;

[0123] Fig. 9 This is the equipment reliability curve for each quarter from 2010 to 2020;

[0124] Fig.10 This is a curve chart of the cumulative equipment failure rate in each quarter from 2010 to 2020;

[0125] Fig.11 This is a graph of the average daily maximum temperature in each quarter from 2010 to 2020;

[0126] Fig.12 This is a precipitation curve for each season from 2010 to 2020. DETAILED DESCRIPTION

[0127] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0128] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0129] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0130] Embodiment 1:

[0131] Combine the following Figure 1-Figure 5 This embodiment is described. This embodiment describes a method for determining the optimal operating life of electric power equipment based on technical and economic performance, which includes:

[0132] S1. Construct a full life cycle cost calculation module for power equipment; use the output results of the full life cycle cost calculation module combined with the uniform degradation trend of operation and maintenance costs and the failure probability of maintenance costs to construct a static economic life calculation model for power equipment; use the output results of the static economic life calculation model combined with the time value of money to construct a dynamic economic life calculation model for power equipment; output the dynamic economic life calculation results;

[0133] S2. Construct a technical life prediction model for power equipment based on the HGWO-LS-SVM algorithm and output the technical life prediction results;

[0134] S3. Obtain the optimal operating life of power equipment based on the dynamic economic life calculation results and technical life prediction results.

[0135] Furthermore, the specific method of constructing the technical life prediction model of power equipment based on the HGWO-LS-SVM algorithm and outputting the technical life prediction result in S2 includes:

[0136] S2-1. Evaluate the quality of scrapped equipment, use the kernel density estimation (KDE) method to fit the quality score with the equipment life, obtain the quality score corresponding to the critical scrap point, and obtain the kernel function;

[0137] S2-2, based on the least squares support vector machine LS-SVM, the technical life prediction model of power equipment is constructed, and the differential evolution algorithm DE is introduced into the grey wolf optimization algorithm GWO to obtain the improved grey wolf algorithm HGWO;

[0138] S2-3, using the improved grey wolf algorithm HGWO to optimize the penalty factor in LS-SVM and the kernel function obtained in S2-1 to obtain the quality scores of power equipment at different periods;

[0139] S2-4. Compare the quality scores of different periods obtained in S2-3 with the quality scores corresponding to the critical scrap points obtained in S2-1. When the quality scores of different periods are lower than the quality scores corresponding to the critical scrap points, the corresponding time is the technical life prediction result.

[0140] Furthermore, the specific method of constructing the full life cycle cost estimation module of the power equipment in S1 includes:

[0141] Life cycle cost LCC for:

[0142]

[0143] Among them, C I represents the initial investment cost, C O,n represents the operation and maintenance cost in the nth year, C F,n represents the failure cost in the nth year, L R Represents the residual price of equipment, j=1,2,…,n.

[0144] Furthermore, the specific method of constructing the static economic life calculation model of power equipment described in S1 includes:

[0145] The average one-time cost C in the nth year I-R,n yes:

[0146]

[0147] Assuming that the equipment deteriorates at a uniform rate and the operation and maintenance cost increases year by year, the average annual operation and maintenance cost C in the nth year is O,n yes:

[0148]

[0149] Among them, C O,1 represents the initial value of the operation and maintenance cost, and α represents the incremental operation and maintenance cost caused by degradation;

[0150] The average annual maintenance cost C in the nth year F,n yes:

[0151]

[0152] in, represents the average annual periodic maintenance cost in year n, represents the average annual fault repair cost in year n;

[0153]

[0154] in, represents the initial value of the periodic maintenance cost, and β represents the incremental periodic maintenance cost caused by deterioration;

[0155]

[0156] in, represents the cost of each fault repair, and h(n) represents the failure rate in the “bathtub curve”;

[0157] The static economic life calculation model is:

[0158] C a (n) = C I-R,n +C O,n +C F,n ;

[0159] N0=[minC a (n)] -1 ;

[0160] Among them, N0 represents the static economic life, C a (n) represents the static average annual cost in year n.

[0161] Furthermore, the specific method of constructing the dynamic economic life estimation model of power equipment described in S1 includes:

[0162] The dynamic economic life N0′ is:

[0163] N0′=[minC a ′(n)] -1 ;

[0164] Among them, C a ′(n) represents the dynamic average annual cost in the nth year;

[0165] C a ′(n)=C I ' -R,n +C′ O,n +C′ F,n ;

[0166] Among them, C I ' -R,n represents the dynamic investment cost in the nth year, C′ O,n represents the dynamic operation and maintenance cost in the nth year, C′ F,n represents the dynamic cost of failure in year n;

[0167] C I ' -R,n =C I (A / P,i,n)-L R (A / F,i,n);

[0168] Where, (A / P,i,n) represents the equal installment capital recovery factor, (A / F,i,n) represents the equal installment debt repayment fund factor, and i represents the basic internal rate of return;

[0169]

[0170] Among them, C O,1 represents the operation and maintenance cost in the first year, (P / F,i,n) represents the one-time payment present value coefficient;

[0171] in:

[0172]

[0173] Furthermore, the kernel density estimation (KDE) method described in S2-1 is used to fit the quality score with the equipment life to obtain the quality score corresponding to the critical scrap point, and the specific method of obtaining the kernel function includes:

[0174] Kernel density estimation KDE is used to fit the quality score with the equipment life to obtain the quality score corresponding to the critical scrap point. The form of kernel density estimation is:

[0175]

[0176] in, represents the KDE fit with respect to The probability density function of represents a random variable used to represent the quality score, N represents the number of samples, h represents the bandwidth, K h represents the kernel function, represents the sample observation value, R d represents a d-dimensional real number set, k represents the kth sample;

[0177] The kernel function adopts Gaussian kernel function, probability density function It is expressed as:

[0178]

[0179] Furthermore, the specific method of constructing the technical life prediction model of power equipment based on the least squares support vector machine LS-SVM described in S2-2 includes:

[0180] The LS-SVM model that extends the support vector machine SVM is proposed:

[0181] First, map the sample from the original space to the feature space:

[0182]

[0183] y represents the output vector, x represents the non-input vector, represents the mapping function, ω and b represent parameters to be determined;

[0184] The determination process of ω and b is equivalent to the minimization of the intermediate variable R:

[0185]

[0186] Among them, ||ω|| 2 represents the complexity of the control model, c represents the adjustment parameter, and controls the degree of penalty for exceeding the error. emp represents the error control function;

[0187] The minimization problem is further expressed as:

[0188]

[0189] J(ω,ξ) represents the loss function, l represents the total number of samples, j(x p ) represents the mapping function of the pth input vector, w represents the complexity parameter, x p represents the pth input vector, ξ p represents the relaxation factor;

[0190] Construct the Lagrangian function L(·):

[0191]

[0192] Among them, α p represents the Lagrange multiplier;

[0193] According to the KKT optimization conditions, we can get:

[0194]

[0195] Eliminating ω and ξ, we get:

[0196]

[0197] Where I = [1,…,1], Use kernel function instead; xm represents the mth vector, x n represents the nth vector;

[0198] The regression function f(x) is:

[0199]

[0200] Among them, K(x,x p ) represents x and x p The kernel function of

[0201] The radial basis function RBF is expressed as:

[0202]

[0203] Among them, x h represents the hth vector, x g represents the gth vector; σ is the kernel width.

[0204] Furthermore, the specific method of introducing the differential evolution algorithm DE into the grey wolf optimization algorithm GWO in S2-2 to obtain the improved grey wolf algorithm HGWO includes:

[0205] First, DE is used to achieve population variation, then the optimal three individual positions are obtained based on fitness, and the distance between the remaining gray wolves and the optimal gray wolf is calculated:

[0206] D a =|C1X a (t)-X(t)|

[0207] D b =|C2X b (t)-X(t)|

[0208] D c =|C3X c (t)-X(t)|

[0209] Where t represents the number of iterations, D a , D b and D c Represent distances respectively; C1, C2 and C3 all represent random vectors, X a , X b and X c represents the three optimal positions closest to the prey, and X represents the current solution position;

[0210] The remaining gray wolves were repositioned according to the three optimal positions mentioned above:

[0211] X1(t+1)=X a (t)-A1D a

[0212] X2(t+1)=X b (t)-A2D b

[0213] X3(t+1)=X c (t)-A3D c

[0214] A1, A2 and A3 all represent convergence factors;

[0215] The final positions of the remaining individuals are:

[0216]

[0217] C = 2r1;

[0218] A = 2ar2-a;

[0219] Among them, C represents the perturbation parameter, A represents the convergence factor, r1 and r2 represent random numbers in [0,1], and a represents a variable related to the number of iterations, decreasing from 2 to 0;

[0220] Keep the better individuals and update A and C until the conditions are met.

[0221] Furthermore, the specific method of obtaining the quality scores of the power equipment at different periods in S2-3 and the corresponding time as the technical life prediction result when the quality scores at different periods are lower than the quality scores corresponding to the critical scrapping point in S2-4 includes:

[0222] Map the quality-related data to the hidden space through the LS-SVM kernel;

[0223] The HGWO algorithm is applied to obtain the optimal values ​​of the adjustment parameter c and the kernel width σ;

[0224] After inputting into LS-SVM, the predicted values ​​of equipment quality scores at different periods were obtained;

[0225] When the quality scores of different periods are lower than the quality scores corresponding to the critical scrapping point, the corresponding time is the technical life prediction result:

[0226] M0=(if S(n)<S) -1 ;

[0227] Among them, M0 represents the technical life prediction result, S(n) represents the predicted value of the equipment quality score in the nth year, and S represents the quality score corresponding to the critical scrap point.

[0228] Furthermore, the specific method for obtaining the optimal operating life of the power equipment according to the dynamic economic life calculation results and the technical life prediction results described in S3 includes:

[0229] The optimal operating life R of power equipment is:

[0230] R = min(N0′,M0);

[0231] Among them, N0′ represents the dynamic economic life, and M0 represents the technical life prediction result.

[0232] In the present invention, power equipment has the characteristics of being multi-faceted, multi-source and heterogeneous. This application mainly studies distribution network equipment. When calculating the economic life and predicting the technical life, three key issues need to be addressed: (1) Acquisition of data on equipment cost, status, performance, etc. each year; (2) Determination of the evolution of equipment performance over time; (3) Evaluation of equipment quality each year. The specific solutions to the above three key issues are as follows:

[0233] The data fields included in the multi-business system are shown in Table 1:

[0234] Table 1

[0235] Business System Data Fields Production Management System Fields such as ledger, retirement, defect, failure, maintenance, and operation Enterprise Resource Planning System Fields such as ledger, assets, costs, materials, and supplies Safety supervision system Fields such as reliable operation events and statistics, power outage events and statistics, etc. Lightning Location System Lightning information, lightning tripping, etc. Environmental Weather System Temperature, humidity, rainfall, altitude and other fields

[0236] According to the data fields contained in the multi-business system, relevant information on economic life estimation and technical life prediction is extracted.

[0237] There are three patterns of evolution of power equipment degradation over time, namely acceleration, uniform speed and deceleration. When internal or external causes change, the equipment quality status will recover. Internal causes mainly refer to equipment maintenance, such as replacing severely worn parts. External causes mainly refer to the impact of the environment on equipment quality, such as coastal areas experiencing a high temperature and high humidity season.

[0238] The evolution of the failure rate of power equipment over time has three periods, namely, the running-in period, the stable period, and the loss period. Figure 3 The "bathtub curve" is shown in the figure. After the equipment is put into operation, it needs to be debugged many times during the running-in period, and small faults often occur; after the debugging is completed, the failure rate is very low and enters a stable period; as the equipment ages as a whole, the failure rate is high and enters a loss period. In addition to the "bathtub curve", the reliability function can also reflect the equipment failure rate.

[0239] When fitting the reliability function and the "bathtub curve", it is necessary to ensure that both are greater than zero, the curve is continuous, and it decreases first and then increases. Studies have shown that it is usually necessary to introduce at least three variables. There are currently more than dozens of fitting results. After comparing the accuracy of various fitting models, the reliability function and "bathtub curve" used in this article are as follows:

[0240] Reliability function:

[0241]

[0242] Failure rate function:

[0243]

[0244] Among them, R(t) is the function of reliability with respect to life; h(t) is the function of failure rate with respect to life; α and β are shape parameters, α∈(0,1), β>0; λ is the adjustment factor of failure rate function; T is the design life of the equipment; t is the service year of the equipment, t∈(0,T);

[0245] The quality evaluation process of power equipment is as follows:

[0246] Following the principles of comprehensiveness, systematicness, independence, and quantification, a quality evaluation index system for power equipment is constructed, including the equipment's operating status, availability factor, number of failures, downtime, defect level, and fault elimination time.

[0247] The weights of various indicators are determined by using the hierarchical analysis method and entropy weight method. In order to improve objectivity and scientificity, the combined weighting method can be used;

[0248] To construct a comprehensive evaluation model, commonly used ones include fuzzy comprehensive evaluation, TOPSIS, matter-element extension, etc.

[0249] By extracting data of various indicators of equipment at different periods from multiple business systems and conducting quality assessment according to the above process, the quality assessment results of the equipment at different periods can be obtained and converted into quality scores.

[0250] Dynamic economic life calculation model based on the time value of money:

[0251] A cost estimation model for the entire life cycle of power equipment was constructed. Taking into account the uniform degradation trend of operation and maintenance costs and the failure trend of maintenance costs, a static economic life estimation model for equipment was proposed. Finally, considering the time value of money, a dynamic economic life estimation model for equipment was proposed. The calculation process is as follows: Figure 2 shown.

[0252] Equipment life cycle cost model:

[0253] The life cycle cost of power equipment includes: initial investment cost, operation and maintenance cost, overhaul cost, and residual value, which can be expressed as:

[0254]

[0255] In the formula, C LCC Represents the life cycle cost of power equipment; C I represents the initial investment cost; C O,t represents the operation and maintenance cost in year t; C F,t represents the failure cost in year t; L R Indicates the residual value of the equipment.

[0256] Static economic life calculation model considering degradation failure:

[0257] In the life cycle cost, the operation and maintenance cost increases year by year due to equipment degradation, and the maintenance cost also increases year by year due to the increase in equipment failure rate. The calculation method of each annual average cost is as follows:

[0258] Average annual one-time cost: Average annual one-time cost refers to the annual amortization of the investment cost after deducting the residual value.

[0259]

[0260] In the formula, C I-R,n represents the average annual one-time cost in year n.

[0261] Average annual operation and maintenance cost: Assuming that the equipment deteriorates at a uniform rate, the operation and maintenance cost will increase by a certain amount each year.

[0262]

[0263] In the formula, C O,n is the average annual operation and maintenance cost in the nth year; C O,1 is the initial value of the operation and maintenance cost; α is the incremental operation and maintenance cost caused by degradation.

[0264] Average annual maintenance cost: Maintenance costs include regular maintenance costs and fault maintenance costs. Regular maintenance is carried out according to the maintenance guidelines for various types of equipment. Considering the uniform deterioration of equipment, the regular maintenance cost increases by a certain amount year by year. Fault maintenance is carried out according to the fault situation, and the annual cost is related to the probability of failure.

[0265]

[0266] In the formula, C F,n is the average annual maintenance cost in the nth year; is the average annual periodic maintenance cost in year n; is the average annual fault repair cost in year n.

[0267]

[0268] In the formula, is the initial value of the periodic maintenance cost; β is the incremental periodic maintenance cost caused by deterioration.

[0269]

[0270] In the formula, is the repair cost for each failure; h is the failure rate in the “bathtub curve”.

[0271] Static economic life: During the life cycle of power equipment, the average annual cost changes with the use time, showing a "U" curve that first decreases and then increases. The time when the average annual cost is the lowest (N0) is the economic life. Figure 3 shown.

[0272] The static economic life calculation method is:

[0273] C a (n) = C I-R,n +C O,n +C F,n

[0274] N0=[minC a (n)] -1

[0275] In the formula, C a (n) is the static average annual cost in the nth year; N0 is the static economic life.

[0276] Dynamic economic life calculation model considering the time value of money:

[0277] The operating life of power equipment is generally long, and the time value of money has a greater impact on the economic life. The dynamic economic life calculation model is:

[0278] C′ a (n) = C′ I-R,n +C′ O,n +C′ F,n

[0279] N′0=[min C′ a (n)] -1

[0280] In the formula, C′ a (n) is the dynamic average annual cost in the nth year; N′0 is the dynamic economic life; the other variables with a prime in the upper right corner all represent dynamic costs.

[0281] After considering the time value of money, the cost calculation methods are revised as follows:

[0282] C′ I-R,n =C I (A / P,i,n)-L R (A / F,i,n)

[0283]

[0284] The corresponding capital equivalent formula is:

[0285]

[0286] In the formula, i represents the basic internal rate of return; (A / P,i,n) represents the equal installment capital recovery coefficient; (A / F,i,n) represents the equal installment debt repayment fund coefficient; and (P / F,i,n) represents the one-time payment present value coefficient.

[0287] Technical life prediction model based on HGWO-LS-SVM: Considering internal factors (reliability, failure rate) and external factors (temperature, humidity), a technical life prediction model for power equipment is constructed based on the least squares support vector machine (LS-SVM). The differential evolution algorithm (DE) is introduced into the gray wolf optimization algorithm (GWO), and the improved gray wolf algorithm (HGWO) is used to optimize the penalty factor and kernel function of LS-SVM to predict the quality scores of the equipment at different periods. When the quality score is lower than the time corresponding to the given standard, it is the technical life of the equipment.

[0288] Kernel density estimation: In order to study the relationship between the quality and life of power equipment, the quality of scrapped equipment is evaluated, and the quality score is fitted with the equipment life using the kernel density estimation (KDE) method to find the quality score corresponding to the critical scrap point. Kernel density estimation is used to estimate the unknown density function and is one of the non-parametric test methods. Compared with participation estimation, the result of kernel density estimation is closer to the true distribution, and the general form is:

[0289]

[0290] In the formula, is a random variable, indicating the quality score; is the KDE fitting The probability density function of h It is the kernel function, which has several types such as Gaussian, Triangle, Epanechnikov, etc.

[0291] This application uses the Gaussian kernel function, It can be expressed as:

[0292]

[0293] The quality scores of transformer equipment that have reached their technical life are mainly distributed on both sides of 60 points, such as Figure 4 shown.

[0294] LS-SVM model: LS-SVM is an extension of support vector machine (SVM), which can effectively improve the solution speed and convergence accuracy. First, the sample is mapped from the original space to the feature space

[0295]

[0296] In the formula, ω and b are parameters to be determined; is the mapping function; x is the input vector; y is the output vector.

[0297] The determination process of ω and b can be equivalently transformed into the minimization of the following problem, namely:

[0298]

[0299] In the formula, ||ω|| 2 Controls the complexity of the model; c is a tuning parameter that controls the degree of penalty for exceeding the error; R emp is the error control function.

[0300] The minimization optimization problem can be further expressed as:

[0301]

[0302] In the formula, ξ i is the relaxation factor.

[0303] Construct the Lagrangian function:

[0304]

[0305] In the formula, α i is the Lagrange multiplier.

[0306] According to the KKT optimization conditions, we can get:

[0307]

[0308] Eliminating ω and ξ, we get:

[0309]

[0310] Where, I = [1,…,1]; We can use the kernel function K(x,x i )replace.

[0311] Finally, the regression function is:

[0312]

[0313] The key issue is to determine the appropriate kernel function. The radial basis function (RBF) kernel function has strong adaptability and good performance. The RBF expression is as follows:

[0314]

[0315] Where σ is the kernel width.

[0316] Parameter optimization based on HGWO:

[0317] GWO has the advantages of fewer parameters, fast speed, and high accuracy, but it is easy to fall into local optimality. DE has the advantages of strong robustness and simple structure, and its introduction can improve the shortcomings of GWO. The HGWO process is: first use DE to realize population variation, then get the optimal three individual positions according to fitness, and calculate the distance between the remaining gray wolves and the optimal gray wolf.

[0318] D a =|C1X a (t)-X(t)|

[0319] D b =|C2X b (t)-X(t)|

[0320] D c =|C3X c (t)-X(t)|

[0321] In the formula, t represents the number of iterations; D a , D b and D c represents distance; C1, C2 and C3 represent random vectors; X a , X b , X c are the three optimal positions closest to the prey; X represents the current solution position;

[0322] The remaining gray wolves were repositioned according to the three optimal positions mentioned above:

[0323] X1(t+1)=X a (t)-A1D a

[0324] X2(t+1)=X b (t)-A2D b

[0325] X3(t+1)=X c (t)-A3D c

[0326] Where A is the convergence factor. The final positions of the remaining individuals are:

[0327]

[0328] C=2r1

[0329] A=2ar2-a

[0330] Where: C is the disturbance parameter; r1 and r2 are random numbers in [0,1]; a is related to the number of iterations, decreasing from 2 to 0.

[0331] Then keep the better individuals, update A and C, and determine whether the conditions are met. If so, output the result. If not, recalculate.

[0332] Prediction model accuracy test:

[0333] Select the residual sum of squares e SSE , mean absolute percentage error e MAPE , mean square error e MSE To measure the prediction accuracy, the expressions of each indicator are as follows:

[0334]

[0335] Where Y i and represent the i-th actual value and predicted value respectively.

[0336] HGWO-LS-SVM prediction process:

[0337] A technical life prediction model for power equipment is proposed based on the HGWO-LS-SVM algorithm. First, the quality-related data are mapped to the hidden space through the LS-SVM kernel. Then the HGWO algorithm is applied to obtain the optimal values ​​of c and σ. After inputting into LS-SVM, the predicted values ​​of the equipment quality scores at different periods are obtained. The time when the quality score is lower than the given standard for the first time is found, which is the predicted life of the equipment.

[0338] M0=(if S(n)<S) -1

[0339] Where M0 is the predicted technical life of the equipment; S(n) is the quality score predicted in the nth year; S is the given equipment operation quality standard. Figure 5 shown.

[0340] The optimal operating life decision process under the combination of two lifespans:

[0341] According to the dynamic economic life calculation results and technical life prediction results, the optimal operating life of power equipment is determined. The specific process is as follows: Figure 1 .

[0342] R=min(N0′,M0)

[0343] Where R is the optimal operating life.

[0344] In the present invention, a 10kV oil-immersed distribution transformer in operation of a power company is taken as an example to extract cost and quality related data information from the ERP and PMS systems. The distribution transformer asset information is shown in Table 2.

[0345] Table 2

[0346] Device Name 94840 distribution transformer Device Code 15M00000013266481 Unit Details State Grid Power Supply Company in a City PMS equipment type code 302 Asset Classification Code 160200001 Rated capacity 500kVA Voltage level AC 10kV Running status In operation Health status Normal state Device Model SCB9-D-500 / 10 Date of manufacture 2010 / 4 / 1 Commissioning date 2010 / 6 / 1

[0347] The cost data of 94840 distribution transformer is shown in Table 3.

[0348] Table 3

[0349] Device Name 94840 distribution transformer Lifespan limit 18 years Use period until 2020 10 years Residual value of assets 5％ Original value of assets 57593.7 yuan Net residual value of assets 2879.69 yuan Initial operation and maintenance costs 1439.84 yuan Operation and maintenance cost growth 400 NTD Initial periodic maintenance 863.91 yuan Increased cost of periodic maintenance 300 NTD Average cost per troubleshooting 30,000 NTD

[0350] Since the temperature and precipitation in different seasons in the province vary greatly, which will have a certain impact on the equipment quality, the time granularity of all quality-related data is quarterly, with a total of 5 groups and 44 dimensions. In order to simplify the analysis, each quarter of each year is represented in a digital sequence format, as shown in Table 4.

[0351] Table 4

[0352] date 2010 / 3 / 1 2010 / 6 / 1 2010 / 9 / 1 2010 / 12 / 1 2014 / 3 / 1 … 2020 / 9 / 1 2020 / 12 / 1 sequence 1 2 3 4 5 … 43 44

[0353] First, the comprehensive quality evaluation method can be used to obtain the quality scores of 94840 distribution transformers in each quarter, such as Figure 6 .

[0354] Then, using the fitting method, according to the historical failure law of distribution network equipment, the values ​​of each parameter are: α = 0.4, β = 0.1, λ = 0.01, η = 1. The reliability curve and failure rate curve are drawn as follows Figure 7 and Figure 8 .

[0355] Interception Figure 7 and Figure 8 The reliability and cumulative failure rate of distribution transformers in each quarter from 2010 to 2020 are obtained by using the first 44-dimensional data. Fig. 9 and Fig.10 shown.

[0356] Finally, according to the environmental meteorological system, the average daily maximum temperature and precipitation of the area are obtained, such as Fig.11 and Fig.12 .

[0357] Among the above-mentioned quarterly data from 2010 to 2020, dimensions 1:36 are sample sets used for model training; dimensions 37:44 are test sets used to verify the accuracy of model predictions; dimensions 44:79 are sets to be predicted used to judge the technical life of equipment.

[0358] Static economic life calculation:

[0359] Based on the asset information and cost data of distribution transformers, the static economic life is calculated. The calculation results are as follows:

[0360] Table 5

[0361]

[0362] It can be seen that the static economic life of the distribution transformer is 12 years, which is lower than the expected service life of the equipment. This is mainly due to the failure to consider the impact of the time value of money.

[0363] Dynamic economic life calculation: After considering the time value of money, the dynamic economic life calculation results are as follows:

[0364] Table 6

[0365]

[0366] It can be seen that the lowest average annual cost of distribution transformers is 13,899 yuan, and the dynamic economic life is 14 years, which is close to the actual service life of the equipment. Since the distribution network equipment has a long service life, the time value of money cannot be ignored. After considering the time of money, the calculated economic life is basically consistent with the actual forecast of the power supply bureau.

[0367] As the service life of the equipment increases, the average annual investment cost and average annual net residual value of the distribution transformer decrease year by year, and the rate of decrease gradually slows down. Considering that the equipment deteriorates at a uniform rate, the average annual operation and maintenance cost and the average annual regular maintenance cost increase at a uniform rate year by year. The average annual fault maintenance cost decreases first and then increases, which is consistent with the "bathtub curve" law representing faults.

[0368] The economic life calculation method proposed in this application can reduce the difficulty of obtaining the annual expenses of various equipment. It only needs to judge the equipment degradation trend and failure probability. For the degradation trend and failure probability, this project has carried out sufficient qualitative analysis and scientific quantitative modeling to find out the law of equipment operation and characterize it with mathematical models. The calculation method optimizes the timing of equipment scrapping and replacement, and provides a decision-making reference for technical transformation and overhaul.

[0369] In the past, the theoretical economic life calculated often did not match the actual economic life. The theoretical economic life was usually ahead of schedule. This problem mainly occurred because the time value of money was not considered. According to the above transformer economic life calculation results, the dynamic economic life lags behind the static economic life. Provincial power grid companies can apply this method to improve the accuracy of economic life calculation in the physical asset analysis and evaluation table.

[0370] The technical model of power equipment has achieved the transformation from "passive" replacement after equipment problems occur to active "prevention" before problems occur. The reliable operation of the power system depends to a large extent on the reliable operation of the equipment. Once a serious failure occurs in electrical equipment, it will cause an accident if it is not maintained and repaired in time, and the power operation and maintenance management is in the "passive side". The application of the technical life prediction method of this project can accurately grasp the future operation and maintenance activities of the equipment, and the power operation and maintenance management is in the "active side", which improves the level of equipment operation and maintenance management.

[0371] Each provincial company can formulate maintenance, inspection and replacement plans according to the trend of power equipment quality score changes. The current power equipment quality management only focuses on the shortcomings of the current state, without considering the impact of the past state on the future. The HGWO-LS-SVM technical life prediction model proposed in this paper can be used to analyze the development trend, so as to truly understand the operating status of the equipment and formulate targeted operation and maintenance plans.

[0372] In summary, the optimal operating life decision result is 14 years when the economic life and technical life of power equipment are combined. The economic life of equipment reflects the operating efficiency of the distribution network. The reasonable application of the economic life calculation method can achieve accurate investment and lean operation and maintenance; the technical life of equipment reflects the frequency of failures in the distribution network. The reasonable application of the technical life prediction method can achieve safe operation and reliable service. Only by combining the economic life and technical life can we take into account both reliable operation and economic efficiency.

[0373] Improving the operation and maintenance management level of power equipment is the safety guarantee and effective support for building a new power system with new energy as the main body. In order to avoid premature replacement of power equipment that has not reached the life limit and can still operate safely, and to prevent delayed replacement of power equipment that has reached the life limit and cannot operate safely, this paper proposes an optimal operation life decision method for power equipment under the combination of economic life and technical life. The empirical results show that: 1) The economic life calculation method not only considers the equipment degradation trend and failure probability, but also takes into account the time value of money. The economic life is finally calculated to be 14 years, and the average annual minimum cost is 13,899 yuan, which solves the problem that the static economic life is inconsistent with the actual life. 2) The technical life prediction method considers multiple internal and external factors, and HGWO improves the convergence speed and solution accuracy of the LS-SVM optimization parameter acquisition process. The technical life is finally predicted to be 16 years, which can provide a decision reference for technical transformation and overhaul. 3) Finally, the two life decision results are combined to determine the optimal operation life of power equipment as 14 years, which can not only ensure safe operation but also save costs.

[0374] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in a manner different from that described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be used in other described embodiments.

Claims

1. A method for determining the optimal operating life of power equipment based on technical and economic factors, characterized in that: It includes: S1. Build a full life cycle cost estimation module for power equipment; The output results of the full-cycle cost calculation module are combined with the uniform degradation trend of the operation and maintenance cost and the failure probability of the maintenance cost to construct a static economic life calculation model for power equipment; the output results of the static economic life calculation model are combined with the time value of money to construct a dynamic economic life calculation model for power equipment; and the dynamic economic life calculation results are output; S2. Construct a technical life prediction model for power equipment based on the HGWO-LS-SVM algorithm and output the technical life prediction results; S3. Obtain the optimal operating life of power equipment based on the dynamic economic life calculation results and technical life prediction results.

2. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 1 is characterized in that: S2 The specific method of constructing the technical life prediction model of power equipment based on the HGWO-LS-SVM algorithm and outputting the technical life prediction result includes: S2-1. Evaluate the quality of scrapped equipment, use the kernel density estimation (KDE) method to fit the quality score with the equipment life, obtain the quality score corresponding to the critical scrap point, and obtain the kernel function; S2-2, based on the least squares support vector machine LS-SVM, the technical life prediction model of power equipment is constructed, and the differential evolution algorithm DE is introduced into the grey wolf optimization algorithm GWO to obtain the improved grey wolf algorithm HGWO; S2-3, using the improved grey wolf algorithm HGWO to optimize the penalty factor in LS-SVM and the kernel function obtained in S2-1 to obtain the quality scores of power equipment at different periods; S2-4. Compare the quality scores of different periods obtained in S2-3 with the quality scores corresponding to the critical scrap points obtained in S2-1. When the quality scores of different periods are lower than the quality scores corresponding to the critical scrap points, the corresponding time is the technical life prediction result.

3. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 1 is characterized in that: The specific method of constructing the full life cycle cost estimation module of power equipment in S1 includes: Life cycle cost LCC for: Among them, C I represents the initial investment cost, C O,n represents the operation and maintenance cost in the nth year, C F,n represents the failure cost in the nth year, L R Represents the residual price of equipment, j=1,2,…,n.

4. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 3 is characterized in that: The specific method of constructing the static economic life calculation model of power equipment described in S1 includes: The average one-time cost C in the nth year I-R,n yes: Assuming that the equipment deteriorates at a uniform rate and the operation and maintenance cost increases year by year, the average annual operation and maintenance cost C in the nth year is O,n yes: Among them, C O,1 represents the initial value of the operation and maintenance cost, and α represents the incremental operation and maintenance cost caused by degradation; The average annual maintenance cost C in the nth year F,n yes: in, represents the average annual periodic maintenance cost in year n, represents the average annual fault repair cost in year n; in, represents the initial value of the periodic maintenance cost, and β represents the incremental periodic maintenance cost caused by deterioration; in, represents the cost of each fault repair, and h(n) represents the failure rate in the "bathtub curve"; The static economic life calculation model is: C a (n)=C I-R,n +C O,n +C F,n ; N0=[minC a (n)] -1 ; Among them, N0 represents the static economic life, C a (n) represents the static average annual cost in year n.

5. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 4 is characterized in that: The specific method of constructing the dynamic economic life calculation model of power equipment described in S1 includes: The dynamic economic life N0′ is: N0′=[minC a ′(n)] -1 ; Among them, C a ′(n) represents the dynamic average annual cost in the nth year; C a ′(n)=C I ′ -R,n +C′ O,n +C′ F,n ; Among them, C I ' -R,n represents the dynamic investment cost in the nth year, C′ O,n represents the dynamic operation and maintenance cost in the nth year, C′ F,n represents the dynamic cost of failure in year n; C I ′ -R,n =C I ·(A / P,i,n)-L R ·(A / F,i,n); Where, (A / P,i,n) represents the equal installment capital recovery factor, (A / F,i,n) represents the equal installment debt repayment fund factor, and i represents the basic internal rate of return; Among them, C O,1 represents the operation and maintenance cost in the first year, (P / F,i,n) represents the one-time payment present value coefficient; in:

6. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 2 is characterized in that: S2-1 describes a method for fitting the quality score with the equipment life by using the kernel density estimation (KDE) method to obtain the quality score corresponding to the critical scrap point, and the specific method for obtaining the kernel function includes: Kernel density estimation KDE is used to fit the quality score with the equipment life to obtain the quality score corresponding to the critical scrap point. The form of kernel density estimation is: in, represents the KDE fit with respect to The probability density function of represents a random variable used to represent the quality score, N represents the number of samples, h represents the bandwidth, K h represents the kernel function, represents the sample observation value, R d represents a d-dimensional real number set, k represents the kth sample; The kernel function adopts Gaussian kernel function, probability density function It is expressed as:

7. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 6 is characterized in that: The specific method of constructing the technical life prediction model of power equipment based on the least squares support vector machine LS-SVM described in S2-2 includes: The LS-SVM model that extends the support vector machine SVM is proposed: First, map the sample from the original space to the feature space: y represents the output vector, x represents the non-input vector, represents the mapping function, ω and b represent parameters to be determined; The determination process of ω and b is equivalent to the minimization of the intermediate variable R: Among them, ||ω|| 2 represents the complexity of the control model, c represents the adjustment parameter, and controls the degree of penalty for exceeding the error. emp represents the error control function; The minimization problem is further expressed as: s.t.y p =j(x p )w+b+x p (p=1,2,…,l); J(ω,ξ) represents the loss function, l represents the total number of samples, j(x p ) represents the mapping function of the pth input vector, w represents the complexity parameter, x p represents the pth input vector, ξ p represents the relaxation factor; Construct the Lagrangian function L(·): Among them, α p represents the Lagrange multiplier; According to the KKT optimization conditions, we can get: Eliminating ω and ξ, we get: Where I = [1,…,1], Use kernel function instead; x m represents the mth vector, x n represents the nth vector; The regression function f(x) is: Among them, K(x,x p ) represents x and x p The kernel function of The radial basis function RBF is expressed as: Among them, x h represents the hth vector, x g represents the gth vector; σ is the kernel width.

8. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 7 is characterized in that: The specific method of introducing the differential evolution algorithm DE into the grey wolf optimization algorithm GWO to obtain the improved grey wolf algorithm HGWO as described in S2-2 includes: First, DE is used to achieve population variation, then the optimal three individual positions are obtained based on fitness, and the distance between the remaining gray wolves and the optimal gray wolf is calculated: D a =|C1X a (t)-X(t)| D b =|C2X b (t)-X(t)| D c =|C3X c (t)-X(t)| Where t represents the number of iterations, D a , D b and D c Represent distances respectively; C1, C2 and C3 all represent random vectors, X a , X b and X c represents the three optimal positions closest to the prey, and X represents the current solution position; The remaining gray wolves were repositioned according to the three optimal positions mentioned above: X1(t+1)=X a (t)-A1D a X2(t+1)=X b (t)-A2D b X3(t+1)=X c (t)-A3D c A1, A2 and A3 all represent convergence factors; The final positions of the remaining individuals are: C＝2r1； A = 2ar2-a; Among them, C represents the perturbation parameter, A represents the convergence factor, r1 and r2 represent random numbers in [0,1], and a represents a variable related to the number of iterations, decreasing from 2 to 0; Keep the better individuals and update A and C until the conditions are met.

9. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 8, characterized in that: The specific method for obtaining the quality scores of power equipment at different periods as described in S2-3 and the corresponding time as the technical life prediction result when the quality scores at different periods are lower than the quality scores corresponding to the critical scrapping point as described in S2-4 includes: Map the quality-related data to the hidden space through the LS-SVM kernel; The HGWO algorithm is applied to obtain the optimal values ​​of the adjustment parameter c and the kernel width σ; After inputting into LS-SVM, the predicted values ​​of equipment quality scores at different periods were obtained; When the quality scores of different periods are lower than the quality scores corresponding to the critical scrapping point, the corresponding time is the technical life prediction result: M0=(if S(n)<S) -1 ; Among them, M0 represents the technical life prediction result, S(n) represents the predicted value of the equipment quality score in the nth year, and S represents the quality score corresponding to the critical scrap point.

10. The method for determining the optimal operating life of electric power equipment based on technical and economic performance according to claim 1, characterized in that: The specific method for obtaining the optimal operating life of power equipment according to the dynamic economic life calculation results and technical life prediction results as described in S3 includes: The optimal operating life R of power equipment is: R = min(N0′,M0); Among them, N0′ represents the dynamic economic life, and M0 represents the technical life prediction result.

Images