Flexible DC power distribution network fault prediction method, device, equipment and medium

Abstract

The invention discloses a flexible DC power distribution network fault prediction method and device, equipment and a medium. The method comprises the steps of collecting power grid operation data of a flexible DC power distribution network; performing feature extraction on the power grid operation data to obtain feature variables; a preset optimization target model is adopted to optimize the preset initial weight vector, and feature weight distribution is adjusted; and calculating a fault probability according to the feature weight distribution, the feature vector and a pre-constructed fault probability prediction model. According to the scheme, the adaptability and precision of fault prediction can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method, device, equipment and medium for predicting faults in a flexible direct current distribution network. Background Art

[0002] With the transformation of the global energy structure and the proposal of the "dual carbon" goal, flexible DC distribution networks have gradually become an important direction for the development of future power systems. Flexible DC distribution networks have the advantages of high efficiency, flexibility and controllability, and have shown significant application value in large-scale renewable energy access, long-distance transmission and power systems with large load fluctuations.

[0003] However, an important challenge faced by flexible DC distribution networks in actual operation is the rapidity and complexity of their fault processes. Especially when grid data fluctuates frequently, traditional fault outage prediction methods are often difficult to respond in a timely manner and are prone to prediction deviations when the grid operating status changes rapidly, resulting in low fault prediction accuracy. Summary of the invention

[0004] In view of the above-mentioned defects, the present invention provides a flexible DC distribution network fault prediction method, device, equipment and medium, which can improve the adaptability and accuracy of fault prediction.

[0005] An embodiment of the present invention provides a method for predicting faults in a flexible DC distribution network, the method comprising: Collect grid operation data of flexible DC distribution network; Extracting features from the power grid operation data to obtain feature variables; The preset initial weight vector is optimized using the preset optimization target model to adjust the feature weight distribution; The failure probability is calculated according to the feature weight distribution, the feature vector and a pre-built failure probability prediction model.

[0006] Preferably, feature extraction is performed on the power grid operation data to obtain feature variables, including: Performing missing value processing and outlier processing on the power grid operation data; According to the processed power grid operation data, key features that affect the fault detection of the flexible DC distribution network are extracted to obtain the characteristic variables.

[0007] Furthermore, the key features include voltage fluctuation frequency, power mutation rate and load change rate; Among them, the voltage fluctuation frequency , the power mutation rate , the load change rate , is the voltage at the i-th time point, is the average value of voltage, N is the number of time points, and are the power values ​​at the current time point and the previous time point respectively, and are the load values ​​at the current time point and the previous time point respectively.

[0008] As a preferred solution, a preset initial weight vector is optimized using a preset optimization target model to adjust the feature weight distribution, including: Determine the initial weight vector of the characteristic variable based on the acquired historical power grid operation data; The gradient descent algorithm is used to minimize the preset optimization target model, and the initial weight vector is optimized in real time to adjust the feature weight distribution.

[0009] Furthermore, the optimization target model includes: ; in, is the loss function, i.e. the optimization objective, represents the feature weight vector, represents the input variable after feature weighting, w represents the weight vector of the optimization target model, represents the input feature vector of the i-th sample, represents the true label value of the i-th sample, Indicates the total number of samples of input data.

[0010] Preferably, the initial weight vector of the characteristic variable is determined based on the acquired historical power grid operation data, including: When the load change rate calculated according to the historical power grid operation data exceeds a preset first threshold, it is identified that the load is in a high load condition, and a preset first vector is set as the initial weight vector; When the proportion of the voltage fluctuation frequency in the preset abnormal index range monitored according to the historical power grid operation data exceeds the preset second threshold, it is identified that it is in a low load condition and the preset second vector is set as the initial weight vector.

[0011] Preferably, the construction process of the fault probability prediction model specifically includes: Divide the pre-collected feature data into a training set and a test set; Initialize the population of the genetic algorithm and randomly generate a number of initial solutions as initial individuals; Performance parameters for fault prediction through pre-built fitness functions; According to the performance index calculated by the fitness function, a preset selection method is used to select individuals with better fitness as the next generation; The preset crossover probability adjustment model is used to dynamically adjust the crossover probability of the population; Use the preset mutation probability adjustment model to dynamically adjust the population's mutation probability; Calculate the parameters of the current generation of individuals. In the optimization process of each generation, calculate the standard deviation change rate and fitness gain rate of the power grid data in real time as the input of the fitness function. Adjust the crossover probability and mutation probability in real time to iterate until the fitness function calculation result reaches a preset threshold or the iteration reaches a preset maximum number of iterations. Output the latest individuals as the parameters of the Logistic regression model to obtain the fault probability prediction model.

[0012] Furthermore, the crossover probability adjustment model is ; The mutation probability adjustment model is: ; The performance parameters include precision and recall; The accuracy ; The recall rate is ; in, is the crossover probability, and the rate of change of the standard deviation of the power grid operation data , and are the standard deviations of the current generation population data and the previous generation population data, is the preset value, fitness gain rate , and is the average fitness of the current generation population data and the previous generation population data; is the mutation probability, represents the nonlinear dynamics of enhanced mutation probability, TP, TN, FP, and FN represent the number of true positive examples, true negative examples, false positive examples, and false negative examples, respectively. and are the minimum and maximum crossover probability, respectively. and are the minimum and maximum values ​​of the mutation probability, respectively.

[0013] Preferably, the failure probability prediction model is ; in, represents the feature weight vector, represents the input variable after feature weighting, represents the feature vector, represents the optimized feature weight vector, is the output result of the flexible DC distribution network fault prediction model, which means that under a given feature weight vector In case of failure Probability of occurrence.

[0014] An embodiment of the present invention further provides a flexible DC distribution network fault prediction device, the device comprising: A collection module, used to collect grid operation data of the flexible DC distribution network; An extraction module, used for extracting features from the power grid operation data to obtain feature variables; An optimization module, used to optimize a preset initial weight vector using a preset optimization target model and adjust feature weight distribution; A calculation module is used to calculate the failure probability based on the feature weight distribution, the feature vector and a pre-built failure probability prediction model.

[0015] An embodiment of the present invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the flexible DC distribution network fault prediction method as described in any one of the above embodiments is implemented.

[0016] An embodiment of the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the flexible DC distribution network fault prediction method as described in any one of the above embodiments.

[0017] The flexible DC distribution network fault prediction method, device, equipment and medium provided by the present invention collects the grid operation data of the flexible DC distribution network; extracts the features of the grid operation data to obtain feature variables; optimizes the preset initial weight vector using a preset optimization target model to adjust the feature weight distribution; calculates the fault probability according to the feature weight distribution, the feature vector and the pre-constructed fault probability prediction model. This solution can improve the adaptability and accuracy of fault prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of a flow chart of a method for predicting faults in a flexible DC distribution network provided by an embodiment of the present invention; Figure 2 It is a structural schematic diagram of a flexible DC distribution network fault prediction device provided by an embodiment of the present invention; Figure 3 It is a structural diagram of a terminal device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] 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.

[0020] See also Figure 1 , is a flow chart of a method for predicting faults in a flexible DC distribution network provided by an embodiment of the present invention, the method comprising steps S1 to S3: Step S1, collecting grid operation data of the flexible DC distribution network; Step S2, extracting features from the power grid operation data to obtain feature variables; Step S3, optimizing the preset initial weight vector using the preset optimization target model to adjust the feature weight distribution; Step S4, calculating the failure probability according to the feature weight distribution, the feature vector and a pre-built failure probability prediction model.

[0021] During the specific implementation of this embodiment, the power grid operation data including voltage, current, power, load change rate, etc. are collected in real time through the monitoring equipment of the distribution network.

[0022] By analyzing the grid operation data, the key characteristic variables that affect the fault detection of the flexible DC distribution network are taken as input features, with the goal of enabling the model to identify faults more effectively.

[0023] Incorporating these features into the input variables of the Logistic regression model enables the model to capture the possibility of power grid failure more accurately.

[0024] In order to further improve the accuracy of power grid fault prediction, the present invention, based on dynamic feature weighting, strengthens the robustness and accuracy of the model by illustrating different power grid operating conditions, clarifying the setting method of the optimal parameter vector and its precise relationship with the predicted probability. That is, dynamic weights are set and optimized. First, the weight vector is initialized, and then the weight vector is optimized in real time using the gradient descent algorithm, so that the model can adjust the feature weight distribution according to the current power grid operating conditions.

[0025] Based on the feature weighted optimization, a fault probability prediction model based on the equipment status is generated through an improved weighted optimization Logistic regression model. The fault probability is calculated according to the feature weight distribution, the feature vector and the pre-built fault probability prediction model; Whether a fault occurs in the flexible DC distribution network is determined according to the fault probability.

[0026] That is, when the predicted fault probability exceeds the preset fault threshold, it is determined that the flexible DC distribution network has a fault.

[0027] The present application provides a fault prediction algorithm with adaptive adjustment capability, which has high adaptability and accuracy in fault power outage prediction.

[0028] In another embodiment provided by the present invention, the step S1 specifically includes the following steps: After obtaining the power grid operation data, data cleaning is also required.

[0029] Data cleaning includes missing value processing and outlier processing. Since missing data accounts for a small part, the deletion method is used to directly discard missing data; for abnormal data, the average value of the same nature is used to interpolate and replace the abnormal data. Numerical data is standardized to unify the data scale.

[0030] After data cleaning, feature selection and definition will be performed.

[0031] By analyzing the grid operation data, the key characteristic variables that affect the fault detection of the flexible DC distribution network are used as input features, so that the model can identify faults more effectively.

[0032] In another embodiment provided by the present invention, the extracted key features include voltage fluctuation frequency, power mutation rate and load change rate; Voltage fluctuation frequency : Defined as the standard deviation rate of change of voltage per unit time, it characterizes the fluctuation of voltage under grid fault conditions, especially in scenarios such as short circuit and overload.

[0033] Voltage fluctuation frequency ; in, represents the voltage at the i-th time point, is the average value of voltage, and N is the number of time points.

[0034] Power mutation rate It is used to describe the degree of power change per unit time. It can capture the characteristics of drastic power changes caused by abnormal power grid equipment or drastic changes in load. This is a reflection of the high sensitivity to changes in equipment status.

[0035] The power mutation rate .

[0036] in, and are the power values ​​at the current and previous time points respectively.

[0037] Load change rate It describes the sudden fluctuations of load and reveals the load transfer and abnormal load disturbance between different nodes of the power grid. Compared with other variables such as voltage and current, it can more accurately reflect the balance and dynamic changes of power grid load.

[0038] The load change rate .

[0039] in, and are the load values ​​at the current and previous time points respectively.

[0040] The three variables describe the characteristics of the power grid from different perspectives, but their fluctuations show a synergistic effect in time. For example, when the voltage fluctuation frequency is abnormal, the power mutation rate will usually fluctuate violently at the same time, and the load change rate will also have a corresponding surge or drop in the load distribution. This dynamic consistency determines their core role as "key characteristic variables". The superficial characteristics of power grid operation failures (such as voltage drops and current anomalies) can only serve as the basis for the first-level identification and cannot fully reflect the nature of the failure. The voltage fluctuation frequency, power mutation rate and load change rate, as the second-level key features, are a further in-depth identification of the fault mode, which can significantly improve the identification accuracy.

[0041] In another embodiment provided by the present invention, step S3 in the above embodiment specifically includes the following steps: Firstly, the weight vector is initialized. Based on the historical grid operation data and combined with the feature selection algorithm, the initial weight vectors of voltage fluctuation frequency, power mutation rate and load change rate are determined.

[0042] The weight vector is then optimized in real time using the gradient descent algorithm, with the preset optimization target model minimized as the goal, so that the model can adjust the feature weight distribution according to the current grid operating conditions.

[0043] It should be noted that the optimization goal of the weight is to minimize the loss function. In order to avoid the imbalance of feature weight distribution, the regularization term is introduced to limit the range of weight change. The formula is: .

[0044] Among them, λ is the regularization coefficient, which is used to prevent the model from overfitting.

[0045] In another embodiment provided by the present invention, the optimization target model includes: ; in, is the loss function, i.e. the optimization objective, represents the feature weight vector, represents the input variable after feature weighting, w represents the weight vector of the optimization target model, represents the input feature vector of the i-th sample, represents the true label value of the i-th sample, Indicates the total number of samples of input data.

[0046] w represents the weight vector of the optimization model, which is a key parameter in the logistic regression model and is used to quantify the influence of each input feature on the prediction target. In the present invention, the value of w is optimized by genetic algorithm and dynamically adjusted to adapt to different grid operating conditions. The weight w value of different features (such as voltage fluctuation frequency, power mutation rate, load change rate) represents its importance in predicting fault events.

[0047] Represents the input feature vector of the i-th sample, which contains multiple key feature variables, such as voltage fluctuation frequency, power mutation rate, load change rate, etc. Corresponding to a moment or state point in actual grid operation, the characteristic value is derived from real-time collected grid monitoring data (such as voltage, current, and load changes). is the core input of the model, which is weighted by features (with w and combined), is used to predict the failure probability of the i-th sample.

[0048] is the true label value of the i-th sample, which is used to indicate the grid state corresponding to the sample: =1: Indicates that a fault has occurred. =0 indicates that the power grid operates normally. It comes from the historical operation data of the power grid and is generated by annotating the fault records. It represents the supervisory signal for model training.

[0049] It represents the total number of samples of input data, that is, the number of samples contained in the historical operation data of the power grid. The size of determines the coverage of training data and the robustness of the model. The power grid operation samples cover a variety of scenarios such as high load, low load, and multiple point faults. Determine the loss function The summation range is the basic indicator for model training. It provides abundant training samples and improves the adaptability of the model to complex scenarios. This embodiment provides an optimization mechanism with strong adaptability, which significantly improves the ability of power grid fault identification.

[0050] In another embodiment provided by the present invention, when determining the initial weight vector, the following steps are specifically included: The voltage fluctuation frequency, power mutation rate and load change rate are determined based on the historical power grid operation data to detect the load condition.

[0051] When the load change rate calculated according to the historical grid operation data exceeds the preset first threshold, that is, the load change rate fluctuates significantly, it is recognized that it is in a high load condition, and the initial weight vector of the high load condition is .

[0052] When the proportion of the voltage fluctuation frequency in the preset abnormal index range monitored according to the historical power grid operation data exceeds the preset second threshold, that is, the voltage fluctuation frequency is the main abnormal index, and it is identified that it is in a low load condition, the initial weight vector of the low load condition is .

[0053] It should be noted that, in this embodiment, the initial weight vector provided is only a preferred embodiment, and in other embodiments, other initial weight vectors may be set under different working conditions.

[0054] In another embodiment provided by the present invention, when constructing a fault probability prediction model, the following steps are specifically included: Conduct a comparative test of grid state characteristics. In order to verify the performance of the model in grid fault detection, the training set and test set are divided into 8:2.

[0055] It has been verified that the prediction accuracy of the model under different operating conditions has been improved from 85% of the traditional method to 94%. In complex working conditions with severe load fluctuations, the recall rate has been improved from 82% to 91%.

[0056] Establish a genetic algorithm-based optimization model. First, initialize the population. Randomly generate a certain number of initial solutions (chromosomes), each of which represents a set of parameters (such as weights and biases) of the Logistics regression model. The population is .in represents the i-th chromosome, that is, the parameter solution.

[0057] Construct a fitness function. This is used to evaluate the quality of each solution. The fitness function is based on the predictive performance of the model.

[0058] Perform a selection operation and use the roulette wheel selection method to select individuals with higher fitness to enter the next generation based on the results of the fitness function.

[0059] Perform mutation operation. Based on the traditional genetic algorithm, the present invention proposes a dynamic crossover probability adjustment method that is deeply coupled with the power grid operation characteristics for distribution network fault detection. The crossover probability of the population is dynamically adjusted using a preset crossover probability adjustment model; the mutation probability of the population is dynamically adjusted using a preset mutation probability adjustment model; Execution of the adaptive adjustment mechanism. First, calculate the parameters of the current generation. During the optimization process of each generation, calculate the standard deviation change rate of the power grid data in real time. and fitness gain rate , as the input of the adaptive adjustment formula, the crossover rate and mutation rate are adjusted in real time, thereby dynamically adjusting the optimization strategy of the genetic algorithm.

[0060] Iterative optimization process. Selection, crossover and mutation operations are performed according to the adjusted parameters to generate a new population and evaluate the fitness of the population.

[0061] The iteration process is repeated until the fitness function reaches the preset threshold or the maximum number of iterations is reached. The performance indicators of the optimized model are verified and evaluated on the test set to ensure the adaptability of the model under different power grid operating conditions.

[0062] Until the fitness function calculation result reaches a preset threshold or the iteration reaches a preset maximum number of iterations, the latest individual is output as a parameter of the Logistic regression model to obtain the fault probability prediction model.

[0063] In order to adapt to different grid conditions, the present invention designs a multi-condition adaptive convergence mechanism to ensure the efficiency and accuracy of the optimization process.

[0064] High load conditions: Increase the crossover probability to greater than 0.7, the mutation probability to greater than 0.5, and enhance the global search capability.

[0065] Low fluctuation conditions: reduce the crossover probability to less than 0.5, the mutation probability to less than 0.3, and enhance local optimization capabilities.

[0066] After testing, it was found that under different power grid conditions, the fault location time was shortened by 25% and the detection accuracy reached 95%.

[0067] The adaptive parameter adjustment mechanism proposed in this application scheme breaks through the application limitations of existing technologies in the field of distribution network fault detection, innovatively combines grid feature feedback and dynamic nonlinear adjustment methods, and significantly improves the efficiency and accuracy of fault detection. Through experimental verification, this scheme performs better than traditional methods in multi-operating condition optimization scenarios, providing a strong guarantee for the safe operation of complex power grids.

[0068] In another embodiment provided by the present invention, the dynamic crossover probability adjustment adopted by the crossover probability adjustment model is: ; The mutation probability adjustment model is: ; The performance parameters include precision and recall; The accuracy ; The recall rate is ;in, is the crossover probability, and the rate of change of the standard deviation of the power grid operation data , and are the standard deviations of the current generation population data and the previous generation population data, is the preset value, fitness gain rate , and is the average fitness of the current generation population data and the previous generation population data; is the mutation probability. When the power grid data fluctuates greatly (i.e. When it is larger), the mutation rate Increase, enhance the global search capability, and prevent falling into the local optimal solution. represents the nonlinear dynamics of enhanced mutation probability, TP, TN, FP, and FN represent the number of true positive examples, true negative examples, false positive examples, and false negative examples, respectively. and are the minimum and maximum crossover probability, respectively. and are the minimum and maximum values ​​of the mutation probability, respectively.

[0069] C r Represents the crossover probability used in the current genetic algorithm, which determines the probability of gene crossover operation occurring in the genetic algorithm. and are the minimum and maximum values ​​of the crossover probability, respectively, to ensure the adjustment range of the crossover probability.

[0070] M r Represents the probability of gene mutation operation occurring in the current genetic algorithm. and are the minimum and maximum values ​​of the mutation probability, respectively, which limit the range of the mutation probability. In order to use the sine function to adjust the mutation probability and make its fluctuation more flexible.

[0071] This application scheme overcomes the limitation of traditional methods that rely solely on fitness changes by introducing the core characteristics of power grid faults (such as voltage fluctuation rate and power change rate) into cross-probability adjustment. It dynamically optimizes feature weights through historical data and real-time feedback, so that the adjustment mechanism can accurately respond to different power grid conditions, solving the problem that existing technologies cannot quickly adapt to highly volatile power grid conditions. In the experiment, the convergence speed was increased by 25%.

[0072] The dynamic fluctuation of mutation probability is achieved through sine function, so that it can be flexibly adjusted in different optimization stages to avoid falling into local optimality; the mutation probability is dynamically adjusted according to the characteristic fluctuations of different fault scenarios. For example, the mutation probability is increased in high-volatility scenarios and reduced in stable scenarios; the global search capability of the algorithm is significantly enhanced, and the detection accuracy of complex power grid faults (such as multi-point faults) is improved by 12%.

[0073] In another embodiment provided by the present invention, the fault probability prediction model is: ; in, represents the feature weight vector, represents the input variable after feature weighting, represents the feature vector, represents the optimized feature weight vector, is the output result of the flexible DC distribution network fault prediction model, which means that under a given feature weight vector In case of failure Probability of occurrence.

[0074] Represents the optimized feature weight vector, which is the final weight parameter obtained by iterative optimization of the genetic algorithm. It is used to transform each input feature ( )’s contribution values ​​are dynamically weighted to maximize the accuracy and robustness of fault prediction. The optimization combines the dynamic crossover and mutation strategies of genetic algorithms, enabling it to adapt to the changes in feature importance under multiple working conditions (high load conditions, low load conditions, etc.).

[0075] is the output of the flexible DC distribution network fault prediction model, which means that given the feature vector In case of failure The probability of occurrence. Its value range is [0,1]. The closer the value is to 1, the higher the possibility of fault occurrence. The closer the value is to 0, the normal operation of the power grid. The calculation is combined with dynamic optimization and multidimensional feature weights , significantly enhancing the model's predictive capabilities in complex fault scenarios.

[0076] Through the improved weighted optimization logistic regression model, a fault probability prediction model based on equipment status is generated. The optimized model further combines the dynamic weights of different characteristic variables to more accurately predict the probability of fault events under different power grid operating conditions.

[0077] The present invention also provides a flexible DC distribution network fault prediction device, see Figure 2 , is a schematic diagram of the structure of a flexible DC distribution network fault prediction device provided by an embodiment of the present invention, the device comprising: A collection module, used to collect grid operation data of the flexible DC distribution network; An extraction module, used for extracting features from the power grid operation data to obtain feature variables; An optimization module, used to optimize a preset initial weight vector using a preset optimization target model and adjust feature weight distribution; A calculation module is used to calculate the failure probability based on the feature weight distribution, the feature vector and a pre-built failure probability prediction model.

[0078] It should be noted that the flexible DC distribution network fault prediction device provided in the embodiment of the present invention can execute the flexible DC distribution network fault prediction method described in any of the above embodiments, and the specific functions of the flexible DC distribution network fault prediction device are not described in detail herein.

[0079] See also Figure 3 , is a schematic diagram of the structure of a terminal device provided in an embodiment of the present invention. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a flexible DC distribution network fault prediction program. When the processor executes the computer program, the steps in the above-mentioned flexible DC distribution network fault prediction method embodiments are implemented, such as Figure 1 Steps S1 to S4 are shown. Or the processor implements the functions of each module in the above-mentioned device embodiments when executing the computer program.

[0080] Exemplarily, the computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments that can complete functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device. For example, the computer program may be divided into various modules, and the specific functions of each module are not described again.

[0081] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the schematic diagram is merely an example of a terminal device and does not constitute a limitation on the terminal device. The terminal device may include more or fewer components than shown in the diagram, or may combine certain components, or different components. For example, the terminal device may also include an input / output device, a network access device, a bus, etc.

[0082] The processor may be a central processing unit (CPU), 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 general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0083] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the terminal device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can 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 can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also 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 volatile solid-state storage devices.

[0084] Wherein, if the module / unit integrated in the terminal device 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 such an 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 various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0085] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for predicting faults in a flexible DC distribution network, characterized in that: The method comprises: Collect grid operation data of flexible DC distribution network; Extracting features from the power grid operation data to obtain feature variables; The preset initial weight vector is optimized using the preset optimization target model to adjust the feature weight distribution; The failure probability is calculated according to the feature weight distribution, the feature vector and a pre-built failure probability prediction model.

2. The method for predicting faults in a flexible DC distribution network according to claim 1, characterized in that: Feature extraction is performed on the power grid operation data to obtain feature variables, including: Performing missing value processing and outlier processing on the power grid operation data; According to the processed power grid operation data, key features that affect the fault detection of the flexible DC distribution network are extracted to obtain the characteristic variables.

3. The method for predicting faults in a flexible DC distribution network according to claim 2, wherein: The key characteristics include voltage fluctuation frequency, power mutation rate and load change rate; Among them, the voltage fluctuation frequency , the power mutation rate , the load change rate , is the voltage at the i-th time point, is the average value of voltage, N is the number of time points, and are the power values ​​at the current time point and the previous time point respectively, and are the load values ​​at the current time point and the previous time point respectively.

4. The method for predicting faults in a flexible DC distribution network according to claim 2, wherein: The preset initial weight vector is optimized using the preset optimization target model to adjust the feature weight distribution, including: Determine the initial weight vector of the characteristic variable based on the acquired historical power grid operation data; Using a gradient descent algorithm to minimize a preset optimization target model, the initial weight vector is optimized in real time to adjust the feature weight distribution; Wherein, the optimization target model includes: ; in, is the loss function, i.e. the optimization objective, represents the feature weight vector, represents the input variable after feature weighting, w represents the weight vector of the optimization target model, represents the input feature vector of the i-th sample, represents the true label value of the i-th sample, Indicates the total number of samples of input data.

5. The method for predicting faults in a flexible DC distribution network according to claim 4, characterized in that: According to the acquired historical power grid operation data, the initial weight vector of the characteristic variables is determined, including: When the load change rate calculated according to the historical power grid operation data exceeds a preset first threshold, it is identified that the load is in a high load condition, and a preset first vector is set as the initial weight vector; When the proportion of the voltage fluctuation frequency in the preset abnormal index range monitored according to the historical power grid operation data exceeds the preset second threshold, it is identified that it is in a low load condition and the preset second vector is set as the initial weight vector.

6. The method for predicting faults in a flexible DC distribution network according to claim 3, characterized in that: The construction process of the fault probability prediction model specifically includes: Divide the pre-collected feature data into a training set and a test set; Initialize the population of the genetic algorithm and randomly generate a number of initial solutions as initial individuals; Performance parameters for fault prediction through pre-built fitness functions; According to the performance index calculated by the fitness function, a preset selection method is used to select individuals with better fitness as the next generation; The preset crossover probability adjustment model is used to dynamically adjust the crossover probability of the population; Use the preset mutation probability adjustment model to dynamically adjust the population's mutation probability; Calculate the parameters of the individuals of the current generation, and in the optimization process of each generation, calculate the standard deviation change rate and fitness gain rate of the power grid data in real time as the input of the fitness function, and adjust the crossover probability and mutation probability in real time to iterate until the calculation result of the fitness function reaches a preset threshold or the iteration reaches a preset maximum number of iterations; Output the latest individual as a parameter of the Logistic regression model to obtain the fault probability prediction model; Among them, the crossover probability adjustment model is ; The mutation probability adjustment model is: ; The performance parameters include precision and recall; The accuracy ; The recall rate is ; in, is the crossover probability, and the rate of change of the standard deviation of the power grid operation data , and are the standard deviations of the current generation population data and the previous generation population data, is the preset value, fitness gain rate , and is the average fitness of the current generation population data and the previous generation population data; is the mutation probability, represents the nonlinear dynamics of enhanced mutation probability, TP, TN, FP, and FN represent the number of true positive examples, true negative examples, false positive examples, and false negative examples, respectively. and are the minimum and maximum crossover probability, respectively. and are the minimum and maximum values ​​of the mutation probability, respectively.

7. The method for predicting faults in a flexible DC distribution network according to claim 1, wherein: The failure probability prediction model is: ; in, represents the feature weight vector, represents the input variable after feature weighting, represents the feature vector, represents the optimized feature weight vector, is the output result of the flexible DC distribution network fault prediction model, which means that under a given feature weight vector In case of failure Probability of occurrence.

8. A flexible DC distribution network fault prediction device, characterized in that: The device comprises: A collection module, used to collect grid operation data of the flexible DC distribution network; An extraction module, used for extracting features from the power grid operation data to obtain feature variables; An optimization module, used to optimize a preset initial weight vector using a preset optimization target model and adjust feature weight distribution; A calculation module is used to calculate the failure probability based on the feature weight distribution, the feature vector and a pre-built failure probability prediction model.

9. A terminal device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for predicting faults in a flexible direct current distribution network as claimed in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the flexible DC distribution network fault prediction method according to any one of claims 1 to 7.

Images