Method, device and equipment for predicting service life of standby power supply of power distribution terminal and medium

Abstract

The invention belongs to the technical field of power distribution networks, and particularly relates to a power distribution terminal standby power supply life prediction method and device, equipment and a medium. The method comprises the following steps: inputting power distribution terminal standby power supply spectral data into a pre-trained life prediction model to obtain a life prediction result of the power distribution terminal standby power supply; the training mode of the life prediction model is as follows: constructing a lightweight network model; constructing a flexible loss function; wherein the flexible loss function comprises two loss items, and the weight of each loss item is determined based on a flexible coefficient; training the lightweight network model based on the training set and the flexible loss function; in the training process, a mayfly naiad algorithm is adopted to optimize parameters and flexibility coefficients of the lightweight network model; obtaining a life prediction model after model training is completed; according to the mayfly naiad algorithm, an extreme value jumping-out method is adopted to prevent a population from falling into local optimum, and the stability and robustness of the model are enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of power distribution networks, and in particular relates to a method, device, equipment and medium for predicting the life of a backup power supply for a power distribution terminal. Background Art

[0002] There are several types of distribution network automation intelligent terminals, such as DTU (Distribution Terminal Unit) and FTU (Feeder Terminal Unit). They are characterized by large number, wide distribution range and high maintenance cost, which makes it crucial to improve the stability of the automation terminal. The instability of the distribution network automation terminal mainly comes from its power supply method. It is generally powered by a three-wire five-column PT (Potential Transformer) from the primary side. However, due to 10kV line harmonics and other reasons, PT failures frequently cause power loss in the automation terminal. New energy backup power supply has become the mainstream backup power supply method for automation terminals. Therefore, accurately predicting the life of the backup power supply of the distribution network automation terminal can greatly reduce the frequency of line inspections, reduce maintenance costs, and improve the intelligence and stability of the distribution network.

[0003] For the problem of backup power life prediction, the existing technology combines the attention mechanism and ResNet for power remaining life prediction. This method has complex modeling and requires a large amount of preliminary battery data, which increases the training time and makes the model more prone to overfitting. Summary of the invention

[0004] The purpose of the present invention is to provide a method, device, equipment and medium for predicting the life of a standby power supply for a distribution terminal, so as to solve the problem in the prior art that the attention mechanism and ResNet are combined for the prediction of the remaining life of the power supply, the modeling is complex and a large amount of preliminary battery data is required, which leads to increased training time and more prone to overfitting of the model.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for predicting the life of a backup power supply of a power distribution terminal, comprising: Obtain the backup power spectrum data of the power distribution terminal; Inputting the spectrum data of the distribution terminal backup power supply into a pre-trained life prediction model, and the life prediction model outputs a life prediction result of the distribution terminal backup power supply; The life prediction model is trained in the following manner: Constructing a lightweight network model; wherein the lightweight network model adopts the Shufflenet neural network; Get the training set for model training; Constructing a flexible loss function; wherein the flexible loss function includes two loss terms, and the weight of each loss term is determined based on the flexibility coefficient; The lightweight network model is trained based on a training set and a flexible loss function; during the training process, the parameters of the lightweight network model and the flexible coefficient are optimized using an ephemeral algorithm; after the model training is completed, the life prediction model is obtained; wherein, the ephemeral algorithm uses an extreme value jump-out method to prevent the population from falling into a local optimum.

[0006] Furthermore, the parameters of the lightweight network model and the flexibility coefficient are optimized using the mayfly algorithm, including: The parameters of the lightweight network model and the flexibility coefficient are used as parameters to be optimized, and the current position of the population is determined according to the parameters to be optimized; the population position is updated to determine whether the current population falls into a local optimum; if it falls into a local optimum, the population is moved to the next position when it falls into the local optimum. Jumped value Replace; if it does not fall into the local optimum, continue to update the position and mate the population; After the preset conditions are met, the optimization of the parameters of the lightweight network model and the flexibility coefficient is completed.

[0007] Furthermore, the jump value It is calculated as follows:

[0008] in, is the best position of the current iteration, The best position for the next iteration; λ 1、 λ 2 is made of β Parameters of decision; β Represents a binary number; is the current vector position; Represents a random number between -1 and 1; Represents a random number between -0.5 and 0.5; represents the calculation coefficient; Represents the average position of the population in the current iteration.

[0009] Furthermore, the flexible loss function is expressed as follows:

[0010] in, is the flexible loss function; MSE is the mean square error, MAE is the mean absolute error, is the flexibility coefficient; y i is the actual value; is the predicted value; n is the sample size.

[0011] Furthermore, in the step of obtaining the spectrum data of the backup power supply of the distribution terminal, a terahertz sensor is used to collect the spectrum data of the backup power supply of the distribution terminal.

[0012] In a second aspect of the present invention, a device for predicting the life of a backup power supply for a power distribution terminal is provided, comprising: A data acquisition module, used to acquire the spectrum data of the backup power supply of the power distribution terminal; A life prediction module, used for inputting the spectrum data of the distribution terminal backup power supply into a pre-trained life prediction model, and the life prediction model outputs the life prediction result of the distribution terminal backup power supply; The life prediction model is trained in the following manner: Constructing a lightweight network model; wherein the lightweight network model adopts the Shufflenet neural network; Get the training set for model training; Constructing a flexible loss function; wherein the flexible loss function includes two loss terms, and the weight of each loss term is determined based on the flexibility coefficient; The lightweight network model is trained based on a training set and a flexible loss function; during the training process, the parameters of the lightweight network model and the flexible coefficient are optimized using an ephemeral algorithm; after the model training is completed, the life prediction model is obtained; wherein, the ephemeral algorithm uses an extreme value jump-out method to prevent the population from falling into a local optimum.

[0013] Furthermore, the parameters of the lightweight network model and the flexibility coefficient are optimized using the mayfly algorithm, including: The parameters of the lightweight network model and the flexibility coefficient are used as parameters to be optimized, and the current position of the population is determined according to the parameters to be optimized; the population position is updated to determine whether the current population falls into a local optimum; if it falls into a local optimum, the population is moved to the next position when it falls into the local optimum. Jumped value Replace; if it does not fall into the local optimum, continue to update the position and mate the population; After the preset conditions are met, the optimization of the parameters of the lightweight network model and the flexibility coefficient is completed.

[0014] Furthermore, the jump value It is calculated as follows:

[0015] in, is the best position of the current iteration, The best position for the next iteration; λ 1、 λ 2 is made of β Parameters of decision; β Represents a binary number; is the current vector position; Represents a random number between -1 and 1; Represents a random number between -0.5 and 0.5; represents the calculation coefficient; Represents the average position of the population in the current iteration.

[0016] According to a third aspect of the present invention, there is provided an electronic device comprising a processor and a memory, wherein the processor is configured to execute a computer program stored in the memory to implement the above-mentioned method for predicting the life of a backup power supply for a distribution terminal.

[0017] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the method for predicting the life of a backup power supply for a distribution terminal as described above is implemented.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention can effectively improve the accuracy of the life prediction of the standby power supply of the distribution terminal by constructing a lightweight network model (using the Shufflenet neural network) and combining it with a flexible loss function for training. At the same time, the model parameters and flexibility coefficients are optimized using the mayfly algorithm to avoid overfitting of the model and enhance the generalization ability of the model. The convergence problem of the life prediction model of the standby power supply of the distribution network terminal is solved. Compared with other algorithms, the convergence speed is stable, the accuracy is high, and the generalization ability is strong.

[0019] Compared with the complex modeling and large amount of preliminary data requirements in the prior art, the lightweight network model adopted by the present invention reduces the amount of calculation and reduces the training cost. Through the optimization of the mayfly algorithm, the training time is further shortened and the training efficiency is improved.

[0020] The extreme value escape method used in the Mayfly algorithm effectively prevents the population from falling into local optimality during the training process, thereby enhancing the stability and robustness of the model.

[0021] By using terahertz sensors to collect spectral data of the backup power supply at the distribution terminal, more accurate and reliable input information is provided for life prediction, further improving the accuracy of the prediction.

[0022] Accurately predicting the life of the backup power supply of distribution network automation terminals can significantly reduce the frequency of line inspections and maintenance costs, and improve the intelligence and stability of the distribution network. It is of great significance to ensure the safety of the power grid and improve the reliability of power supply.

[0023] The device for predicting the life of a backup power supply for a power distribution terminal, an electronic device and a computer-readable storage medium provided by the present invention also solve the problems raised in the background technology section. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 A schematic diagram of a flow chart of a method for predicting the life of a backup power supply for a power distribution terminal according to an embodiment of the present invention; Figure 2 A training principle diagram of a life prediction model in an embodiment of the present invention; Figure 3 This is a comparison diagram of the model convergence curves in the embodiments of the present invention; Figure 4 This is a flow chart of channel shuffling in an embodiment of the present invention; Figure 5 : This is a comparison diagram of the frequency domain spectra before and after noise reduction in an embodiment of the present invention; wherein (a) the spectrum before noise reduction and (b) the spectrum after noise reduction; Figure 6 A comparison diagram of the convergence curves of the algorithm of the embodiment of the present invention and other algorithms; Figure 7 It is a confusion matrix diagram of the model on the test set in an embodiment of the present invention; Figure 8 This is a structural block diagram of a device for predicting the life of a backup power supply for a power distribution terminal according to an embodiment of the present invention; Fig. 9 The present invention is a structural block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0026] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention.

[0027] Terahertz (THz) refers to electromagnetic waves with a frequency range of 0.1~10 THz (wavelength of 0.03~3mm). It has some spectral characteristics of far infrared and X-rays, and has good penetration and chemical selectivity. Terahertz spectroscopy can be used to analyze the composition, structure and dynamic properties of materials, and is widely used in material classification, material defect detection and other aspects.

[0028] Shufflenet is a convolutional neural network architecture designed specifically for mobile devices with very limited computing power.

[0029] Mayfly Algorithm (MA) is an emerging swarm intelligence optimization algorithm that simulates the flight and mating behavior of mayflies, increases the diversity of the algorithm, combines the advantages of particle swarm, genetic and firefly algorithms, has strong global search capabilities, and performs well in multi-objective optimization, nonlinear optimization, discrete optimization and other problems.

[0030] Example 1 The present invention provides a method for predicting the life of a distribution terminal backup power supply. The spectrum data of the distribution terminal backup power supply is obtained and input into a pre-trained life prediction model for life prediction. In order to improve the prediction accuracy, a flexible loss function is constructed, and the mayfly algorithm is used to optimize the model parameters and flexibility coefficients, thereby avoiding overfitting of the model and enhancing the generalization ability of the model.

[0031] like Figure 1 As shown, a method for predicting the life of a backup power supply for a power distribution terminal includes the following specific steps: S1. Obtain the spectrum data of the backup power supply of the power distribution terminal.

[0032] Optionally, this solution uses a terahertz sensor to collect spectrum data of the backup power supply of the distribution terminal.

[0033] S2. Inputting the spectrum data of the distribution terminal backup power supply into a pre-trained life prediction model, and the life prediction model outputs a life prediction result of the distribution terminal backup power supply.

[0034] In step S2, the life prediction model is trained in the following manner: S21, constructing a lightweight network model; wherein the lightweight network model adopts the Shufflenet neural network; In step S21, the lightweight network model aims to achieve the best model accuracy with limited computing resources. The core of the model architecture adopts two new operations: pointwise group convolution (PGCONV) and channel shuffle. These two operations solve the drawbacks of traditional convolution and greatly reduce the computational complexity of the model while maintaining accuracy. PGCONV is a channel sparse connection method that groups different feature maps of the input layer, and then uses different convolution kernels to convolve each group. In this way, each convolution kernel no longer processes all input channels, but only processes a part of the input channels, which greatly reduces the computational complexity of convolution. The disadvantage of this operation is that after grouping, the feature maps of each group cannot communicate across groups, which indirectly reduces the accuracy of the model. Another core operation of the algorithm, channel shuffle, solves this problem. The operation example is as follows. Figure 4 As shown in the figure. For the network model with 12 channels of the backup power spectrum of the power distribution terminal, the input layer is divided into 3 groups, first reshaped into two dimensions (3, 4), then transposed to (4, 3), and finally reshaped into one dimension, that is, only simple dimension operations and transpositions are needed to achieve uniform channel shuffling. Each group contains the characteristics of other groups, solving the problem of inter-group communication failure.

[0035] Specifically, this solution can optimize the lightweight network from the following four aspects: 1) Minimize memory access cost (MAC) by making the number of input and output channels equal. Assume that the number of input and output channels of the network is and , the feature map size is ,but The number of floating point operations per second (FLOPs) of convolution is ; According to inequality (1), , MAC takes the minimum value.

[0036] (1) 2) Avoid excessive use of group convolutions, which can lead to increased memory access. The FLOPs of PGCONV is , where g is the number of groups. From formula (2), we can get: fixed input When C remains unchanged, an increase in g will increase MAC.

[0037] (2) 3) Reduce network fragmentation operations to improve training speed. Network fragmentation refers to the parallel and mixed branches of the algorithm. Although too many fragmentation operations can slightly improve network accuracy, they will reduce the efficiency of the model, which is not worth the loss on mobile devices with weak computing power.

[0038] 4) Reduce element-level operations to improve network training speed. Experimental results show Although operations like these have small FLOPs, they involve large memory accesses, so try to avoid them.

[0039] S22, obtaining a training set for model training; In step S22, the spectral data of the distribution terminal backup power supplies of different specifications and different aging degrees and the corresponding labels are constructed into a data set, and the data set is divided into a training set and a test set.

[0040] S23. Construct a flexible loss function; wherein the flexible loss function includes two loss terms, and the weight of each loss term is determined based on the flexibility coefficient.

[0041] In step S23, the flexible loss function is expressed as follows: (3) in, is the flexible loss function; MSE is the mean square error, MAE is the mean absolute error, is the flexibility coefficient; y i is the actual value; is the predicted value; n is the sample size.

[0042] The applicant needs to explain that the reason for choosing MSE to participate in the new loss function is that its calculation method makes the algorithm more penalized for large errors, which is conducive to the rapid convergence of the model and is more suitable for mobile devices with low computing power studied in this scheme. The reason for choosing MAE to participate in the new loss function is that it is more robust to outliers, which is conducive to reducing the impact of outliers on the model and reducing the overfitting of the model. As one of the dimensions of multi-objective optimization of the Mayfly algorithm, it participates in the iteration together.

[0043] S24. The lightweight network model is trained based on a training set and a flexible loss function. During the training process, the parameters of the lightweight network model and the flexible coefficient are optimized using an ephemeral algorithm. After the model training is completed, the life prediction model is obtained. The ephemeral algorithm uses an extreme value jump-out method to prevent the population from falling into a local optimum.

[0044] In step S24, the parameters of the lightweight network model and the flexibility coefficient are optimized using the ephemeral algorithm, including: taking the parameters of the lightweight network model and the flexibility coefficient as the parameters to be optimized, determining the current position of the population according to the parameters to be optimized; updating the population position, judging whether the current population falls into a local optimum; if it falls into a local optimum, making the population at the next position when it falls into the local optimum Jumped value Replace; if it does not fall into the local optimum, continue to update the position and mate the population; after reaching the preset conditions, complete the optimization of the parameters of the lightweight network model and the flexibility coefficient.

[0045] Specifically, before training, initialize the model and algorithm parameters, including initializing the lightweight network model: setting the structure of the convolution layer (including convolution kernel size, number, step size, padding method, etc.), pooling layer (type, size, step size, etc.), and fully connected layer (number of input and output channels, activation function, etc.), and initializing the Mayfly algorithm parameters: including population size, mating probability, mutation probability, upper limit of iterations, learning rate (for network training), and flexibility coefficient. (initialized to 0.5).

[0046] Specifically, the mayfly algorithm randomly generates two groups of mayflies, representing males and females respectively, and each mayfly is randomly placed in the problem space as a candidate solution. Indicates that female mayflies use express, For male mayflies exist The position at For female mayflies exist The positions of male and female mayflies are expressed by formulas (4) and (5) respectively.

[0047] (4) (5) in, is the velocity vector of the i-th mayfly at time t+1.

[0048] In the fitness calculation, the performance of each mayfly is evaluated according to a predefined objective function.

[0049] Specifically, the algorithm optimizes the position update rule as follows: male mayflies adjust their positions based on their own experience and the experience of their neighbors, while female mayflies fly to the best male for mating.

[0050] The speeds of male and female mayflies are determined by equations (6) and (7), respectively.

[0051] (6) (7) (8) in, Representative Mayfly exist Dimensions The speed of time, , is the attraction coefficient, is the individual visibility distance; e is the base of natural logarithms; r p Represents the current individual position and The distance r g Represents the current individual position and The distance is calculated by formula (8). In formula (8), x i is the current location, X i is the reference position, is the Euclidean distance between the current position and the reference position; x ij and X ij In the dimensions j The current position and reference position on n is the dimension of the search space. For Mayflies The individual historical optimal position of is the current optimal individual position; is the position of the male mayfly at moment t; is the position of the female mayfly at time t; r mf is the distance of the female mayfly from the male mayfly; is the random walk coefficient, for A random number between .

[0052] Algorithm mating process: Produce offspring mayflies through mating. The production method is shown in formula (9), where the best male and the best female produce the first offspring , the second best male and the second best female produce a second offspring .

[0053] (9) Among them, male is the father, female is the mother; L is a random number.

[0054] Iteration and termination of the algorithm: Repeat the above steps until the termination condition is met (the maximum number of iterations is reached or the optimal solution that meets the accuracy requirements is found).

[0055] Specifically, Figure 2 As shown in the figure, this scheme proposes a new extreme value escape algorithm JEV to solve the problem that the model is prone to fall into the extreme value trap. and And the corresponding judgment conditions, so that the next position of the population when it falls into the local optimum Jumped value Replace, thus jumping out of the extreme value trap, The judgment condition is shown in formula (10). is the flexibility coefficient, and its value range is .

[0056] (10) Bounce value The calculation method of is shown in formula (11).

[0057] (11) in, is the best position of the current iteration, The best position for the next iteration; λ 1、 λ 2 is made of β Decision parameters. is the current vector position, , represents a random number between -1 and 1; , represents a random number between -0.5 and 0.5. ,but 1, the rest is 0. The current vector position and the current best position in the simultaneous participation and parameters The randomness of the selection makes the population more diverse, and the search process for the optimal value more flexible and efficient. represents the calculation coefficient, where T The current iteration number , MaxT is the preset maximum number of iterations; Indicates the average position of the population in the current iteration; β Represents a binary number, when , β for 0 ,otherwise β is 1, where , d It is the algorithm dimension.

[0058] The method of the present invention can perform model training and life prediction on mobile devices with low computing power. In practical applications, maintenance personnel can timely detect the life of the distribution network terminal backup power supply during line inspection, thereby significantly reducing the line inspection frequency, reducing maintenance costs, and improving the intelligence and stability of the distribution network. By analyzing the model convergence curve, prediction confusion matrix and other simulations of the distribution terminal backup power supply with different parameters of the algorithm of this scheme, the results show that the method of this scheme can accurately predict the remaining life of the backup power supply under different loss levels, with an average accuracy of up to 92.8%.

[0059] Figure 3 The model convergence curves in regression analysis were compared between this method, the Levy flight method and the control group. Figure 3 As shown in the figure, although Levy flight is more helpful for the algorithm to jump out of the local optimal solution than the control group, its random strategy causes the algorithm to generate more invalid attempts during the search process, thereby increasing the calculation time, and ultimately leading to slow model convergence. In addition, due to its weak stability, it still falls into the local optimal situation in some experiments. Although the convergence speed of this method is not as fast as that of the Levy flight method in the early iterations, the overall convergence is stable, there is no fall into the local optimal situation, and the final accuracy is high.

[0060] Below, this solution is further explained and illustrated in conjunction with a simulation example.

[0061] 1. Collection of initial data The experiment uses a BT-PTS-FC portable terahertz spectrometer, which has the advantages of small size and large radio frequency range. The specific parameters are shown in Table 1.

[0062] Table 1 Instrument parameters

[0063] The experiment selected two different specifications of distribution terminal backup power supplies, and their specifications are shown in Table 2. The ultraviolet accelerated experiment was used to cause different degrees of use loss, and its remaining life was used as the sample label. In order to make the algorithm model have a better performance, the remaining life is evenly distributed in 0-15 years when making the training set samples. In order to make the test set samples closer to reality, when making the test set, the samples with a remaining life of 0-3 years accounted for 10%, the samples with a remaining life of 3-6 years accounted for 15%, the samples with a remaining life of 6-9 years accounted for 20%, the samples with a remaining life of 9-12 years accounted for 25%, and the samples with a remaining life of 12-15 years accounted for 30%, and they were evenly distributed in each stage. The specific distribution of the samples is shown in Table 3.

[0064] Table 2 Specifications of backup power supply

[0065] Table 3 Experimental sample distribution

[0066] The prepared samples are sequentially used with a terahertz spectrometer to obtain their ATR spectra. The instrument probe transmits terahertz waves to the samples and collector in turn, and the ATR spectrum of the current sample is displayed on the host.

[0067] 2. Data Preprocessing and Noise Reduction Due to human operation factors and equipment accuracy, the collected data will have some abnormal points, missing points and outliers. This solution uses the following preprocessing method to remove them.

[0068] 1) Abnormal points like , then the point is identified as an outlier point and the The outliers are removed by this method. y t represents the data value at time point t, is the average value of the data.

[0069] 2) Missing Points For missing points in the data, spline interpolation is used to fill in the gaps. Spline interpolation can produce similar effects to high-order polynomial interpolation, but the calculation is simpler and the fitting effect is better. In addition, changing the shape of the fitting curve only requires changing the control amount at the node, which makes spline interpolation more flexible.

[0070] 3) Outliers When the situation of formula (12) or (13) occurs, based on the adoption of The method of transferring data y t Remove.

[0071] (12) (13) Among them, Δ y t is the change in data values ​​at adjacent time points; is the sample standard deviation.

[0072] In order to facilitate the regression analysis of the algorithm and to observe the changing characteristics of the spectrum at each frequency within the band, the ATR spectrum is zero-filled and then fast Fourier transformed to obtain its average frequency domain spectrum, as shown in Figure 5 In order to retain the frequency bands with rich feature information, the SVD denoising method is used to denoise the spectrum. The denoising effect is shown in Figure 5 As shown in (b).

[0073] 3. Use the developed new algorithm to establish and optimize the life prediction model The hybrid algorithm of this scheme is used to train 400 training set samples of DC48V, and the model convergence curve is compared with the Shufflenet-HDC method and the MOA-VMD method. The results are as follows: Figure 6 As shown in the figure, it can be seen that the Shufflenet-HDC method converges faster in the first half, but the final accuracy of the model is not enough. The accuracy of the MOA-VMD method is slightly better than that of MOA-VMD. The convergence speed of the algorithm of this scheme is generally stable, and the final prediction accuracy is the highest.

[0074] The trained model was applied to the 200 test sets of DC48V for verification, and its final accuracy was 96.9%. The model confusion matrix is ​​as follows: Figure 7 As shown in the figure, it can be seen that the overall performance of the prediction model on the test set is stable, and there is no fluctuation in the prediction accuracy at a certain stage.

[0075] In order to test the generalization ability of the model, the algorithm of this scheme is trained using DC48V and DC24V respectively, and the trained model is applied to the 200 test sets of DC24V, and the Shufflenet-HDC and MOA-VMD methods are compared. The average accuracy is shown in Table 4. It can be seen that the performance of the model trained with the DC48V training set and the model trained with DC24V on the DC24V test set is not much different, with an average difference of 4.3%, while the gap between the other two methods is larger, proving that their generalization ability is not as good as that of the algorithm of this scheme.

[0076] Table 4 Prediction accuracy of different algorithms on the DC24V test set

[0077] Example 2 like Figure 8 As shown, based on the same inventive concept as the above embodiment, the present invention also provides a device for predicting the life of a backup power supply for a power distribution terminal, comprising: A data acquisition module, used to acquire the spectrum data of the backup power supply of the power distribution terminal; A life prediction module, used for inputting the spectrum data of the distribution terminal backup power supply into a pre-trained life prediction model, and the life prediction model outputs the life prediction result of the distribution terminal backup power supply; The life prediction model is trained in the following manner: Constructing a lightweight network model; wherein the lightweight network model adopts the Shufflenet neural network; Get the training set for model training; Constructing a flexible loss function; wherein the flexible loss function includes two loss terms, and the weight of each loss term is determined based on the flexibility coefficient; The lightweight network model is trained based on a training set and a flexible loss function; during the training process, the parameters of the lightweight network model and the flexible coefficient are optimized using an ephemeral algorithm; after the model training is completed, the life prediction model is obtained; wherein, the ephemeral algorithm uses an extreme value jump-out method to prevent the population from falling into a local optimum.

[0078] Specifically, the parameters of the lightweight network model and the flexibility coefficient are optimized using the mayfly algorithm, including: The parameters of the lightweight network model and the flexibility coefficient are used as parameters to be optimized, and the current position of the population is determined according to the parameters to be optimized; the population position is updated to determine whether the current population falls into a local optimum; if it falls into a local optimum, the population is moved to the next position when it falls into the local optimum. Jumped value Replace; if it does not fall into the local optimum, continue to update the position and mate the population; After the preset conditions are met, the optimization of the parameters of the lightweight network model and the flexibility coefficient is completed.

[0079] Specifically, the jump value It is calculated as follows:

[0080] in, is the best position of the current iteration, The best position for the next iteration; λ 1、 λ 2 is made of β Decision parameters. is the current vector position, , represents a random number between -1 and 1; , represents a random number between -0.5 and 0.5. ,but 1, the rest is 0. The current vector position and the current best position in the simultaneous participation and parameters The randomness of the selection makes the population more diverse, and the search process for the optimal value more flexible and efficient. represents the calculation coefficient, where T The current iteration number , MaxT is the preset maximum number of iterations; Indicates the average position of the population in the current iteration; β Represents a binary number, when ,β is 0 ,otherwise β is 1, where , d It is the algorithm dimension.

[0081] Example 3 like Fig. 9 As shown, the present invention also provides an electronic device 100 for implementing a method for predicting the life of a backup power supply for a power distribution terminal; The electronic device 100 includes a memory 101 , at least one processor 102 , a computer program 103 stored in the memory 101 and executable on the at least one processor 102 , and at least one communication bus 104 .

[0082] The memory 101 can be used to store a computer program 103. The processor 102 implements the steps of a method for predicting the life of a distribution terminal backup power supply in Example 1 by running or executing the computer program stored in the memory 101 and calling data stored in the memory 101.

[0083] The memory 101 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 may include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0084] At least one processor 102 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and uses various interfaces and lines to connect various parts of the entire electronic device 100.

[0085] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a method for predicting the life of a backup power supply for a power distribution terminal, and the processor 102 can execute the plurality of instructions to implement: Obtain the backup power spectrum data of the power distribution terminal; Inputting the spectrum data of the distribution terminal backup power supply into a pre-trained life prediction model, and the life prediction model outputs a life prediction result of the distribution terminal backup power supply; The life prediction model is trained in the following manner: Constructing a lightweight network model; wherein the lightweight network model adopts the Shufflenet neural network; Get the training set for model training; Constructing a flexible loss function; wherein the flexible loss function includes two loss terms, and the weight of each loss term is determined based on the flexibility coefficient; The lightweight network model is trained based on a training set and a flexible loss function; during the training process, the parameters of the lightweight network model and the flexible coefficient are optimized using an ephemeral algorithm; after the model training is completed, the life prediction model is obtained; wherein, the ephemeral algorithm uses an extreme value jump-out method to prevent the population from falling into a local optimum.

[0086] Example 4 If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0087] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0089] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0090] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0091] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for predicting the life of a backup power supply for a power distribution terminal, characterized in that: include: Obtain the backup power spectrum data of the power distribution terminal; Inputting the spectrum data of the distribution terminal backup power supply into a pre-trained life prediction model, and the life prediction model outputs a life prediction result of the distribution terminal backup power supply; The life prediction model is trained in the following manner: Constructing a lightweight network model; wherein the lightweight network model adopts the Shufflenet neural network; Get the training set for model training; Constructing a flexible loss function; wherein the flexible loss function includes two loss terms, and the weight of each loss term is determined based on the flexibility coefficient; The lightweight network model is trained based on a training set and a flexible loss function; during the training process, the parameters of the lightweight network model and the flexible coefficient are optimized using an ephemeral algorithm; after the model training is completed, the life prediction model is obtained; wherein, the ephemeral algorithm uses an extreme value jump-out method to prevent the population from falling into a local optimum.

2. The method for predicting the life of a backup power supply for a power distribution terminal according to claim 1, characterized in that: The parameters of the lightweight network model and the flexibility coefficient are optimized using the mayfly algorithm, including: The parameters of the lightweight network model and the flexibility coefficient are used as parameters to be optimized, and the current position of the population is determined according to the parameters to be optimized; Update the population position to determine whether the current population is trapped in the local optimum; if it is trapped in the local optimum, make the population in the next position when it is trapped in the local optimum Jumped value Replace; if it does not fall into the local optimum, continue to update the position and mate the population; After the preset conditions are met, the optimization of the parameters of the lightweight network model and the flexibility coefficient is completed.

3. The method for predicting the life of a backup power supply for a power distribution terminal according to claim 2, characterized in that: The bounce value It is calculated as follows: in, is the best position of the current iteration, The best position for the next iteration; λ 1、 λ 2 is made of β Parameters of decision; β Represents a binary number; is the current vector position; Represents a random number between -1 and 1; Represents a random number between -0.5 and 0.5; represents the calculation coefficient; Represents the average position of the population in the current iteration.

4. The method for predicting the life of a backup power supply for a power distribution terminal according to claim 1, characterized in that: The flexible loss function is expressed as follows: in, is the flexible loss function; MSE is the mean square error, MAE is the mean absolute error, is the flexibility coefficient; y i is the actual value; is the predicted value; n is the sample size.

5. The method for predicting the life of a backup power supply for a power distribution terminal according to claim 1, characterized in that: In the step of obtaining the spectrum data of the backup power supply of the distribution terminal, a terahertz sensor is used to collect the spectrum data of the backup power supply of the distribution terminal.

6. A device for predicting the life of a backup power supply for a power distribution terminal, characterized in that: include: A data acquisition module, used to acquire the spectrum data of the backup power supply of the power distribution terminal; A life prediction module, used for inputting the spectrum data of the distribution terminal backup power supply into a pre-trained life prediction model, and the life prediction model outputs the life prediction result of the distribution terminal backup power supply; The life prediction model is trained in the following manner: Constructing a lightweight network model; wherein the lightweight network model adopts the Shufflenet neural network; Get the training set for model training; Constructing a flexible loss function; wherein the flexible loss function includes two loss terms, and the weight of each loss term is determined based on the flexibility coefficient; The lightweight network model is trained based on a training set and a flexible loss function; during the training process, the parameters of the lightweight network model and the flexible coefficient are optimized using an ephemeral algorithm; after the model training is completed, the life prediction model is obtained; wherein, the ephemeral algorithm uses an extreme value jump-out method to prevent the population from falling into a local optimum.

7. The device for predicting the life of a backup power supply for a power distribution terminal according to claim 6, characterized in that: The parameters of the lightweight network model and the flexibility coefficient are optimized using the mayfly algorithm, including: The parameters of the lightweight network model and the flexibility coefficient are used as parameters to be optimized, and the current position of the population is determined according to the parameters to be optimized; Update the population position to determine whether the current population is trapped in the local optimum; if it is trapped in the local optimum, make the population in the next position when it is trapped in the local optimum Jumped value Replace; if it does not fall into the local optimum, continue to update the position and mate the population; After the preset conditions are met, the optimization of the parameters of the lightweight network model and the flexibility coefficient is completed.

8. The device for predicting the life of a backup power supply for a power distribution terminal according to claim 7, characterized in that: The bounce value It is calculated as follows: in, is the best position of the current iteration, The best position for the next iteration; λ 1、 λ 2 is made of β Parameters of decision; β Represents a binary number; is the current vector position; Represents a random number between -1 and 1; Represents a random number between -0.5 and 0.5; represents the calculation coefficient; Represents the average position of the population in the current iteration.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the method for predicting the life of a backup power supply for a distribution terminal as described in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by the processor, the method for predicting the life of the backup power supply for the distribution terminal as described in any one of claims 1 to 5 is implemented.

Images