Electrical equipment impedance identification method and device, equipment and storage medium

Abstract

The invention is suitable for the technical field of power system simulation, and provides a power equipment impedance identification method and device, equipment and a storage medium, the method comprises the following steps: obtaining a first working condition parameter set of target power equipment, the first working condition parameter set comprising a target parameter set of the target power equipment under each of a plurality of working conditions; and inputting the first working condition parameter set into a trained impedance identification neural network model to obtain impedance data of the target power equipment under each working condition. According to the technical scheme provided by the invention, the obtained target parameter set of the target power equipment under each of the plurality of working conditions is input into the trained impedance identification neural network model, the impedance data of the target power equipment under each working condition is directly obtained, and the impedance identification efficiency of the power equipment is improved.

Description

Technical Field

[0001] The present application belongs to the technical field of power system simulation, and in particular, relates to a method, device, equipment and storage medium for identifying impedance of power equipment. Background Art

[0002] With the rapid development of the power industry, the scale and complexity of power system equipment are increasing. Under the circumstances of large scale and high complexity, the difficulty of testing its impedance is also increasing.

[0003] Currently, physical formulas (such as Ohm's law) are usually used to calculate impedance. However, under different working conditions, the type of working condition combination determines the number of impedance calculation tests required. Each impedance test experiment will take a lot of time and be inefficient if calculated using the current formula.

[0004] Therefore, how to improve the efficiency of impedance identification of power equipment has become an urgent problem to be solved. Summary of the invention

[0005] The embodiments of the present application provide a method, apparatus, device and storage medium for identifying impedance of an electric power device, aiming to improve the efficiency of identifying impedance of an electric power device.

[0006] In a first aspect, an embodiment of the present application provides an impedance identification method for an electrical equipment, the method comprising: obtaining a first operating condition parameter set of a target electrical equipment, the first operating condition parameter set comprising a target parameter set of the target electrical equipment under each of a plurality of operating conditions; inputting the first operating condition parameter set into a trained impedance identification neural network model to obtain impedance data of the target electrical equipment under each of the operating conditions.

[0007] In a possible implementation, the first operating condition parameter set is input into a trained impedance identification neural network model to obtain the impedance data of the target power equipment under each of the operating conditions, including: based on the impedance identification neural network model, the first operating condition parameter set is operated in batches to obtain multiple second operating condition parameter sets, each of the second operating condition parameter sets includes at least two target parameter sets; the multiple second operating condition parameter sets are input one by one into the trained impedance identification neural network model to obtain the impedance data of the target power equipment under each of the operating conditions.

[0008] In one possible implementation, the training process of the impedance identification neural network model includes: building the impedance identification neural network model based on a funnel model; obtaining a first historical data set, the first historical data set including multiple historical operating condition parameter sets and historical impedance data sets corresponding to each of the historical operating condition parameter sets, the historical operating condition parameter sets including historical parameter sets of the target power equipment under any of the operating conditions; based on the first historical data set, training the impedance identification neural network model using a gradient descent method and introducing weight decay.

[0009] In a possible implementation, before the impedance identification neural network model is trained based on the first historical data set by adopting a gradient descent method and introducing weight decay, the method further includes: standardizing or normalizing the first historical data set to obtain a first target historical data set; correspondingly, the impedance identification neural network model is trained based on the first historical data set by adopting a gradient descent method and introducing weight decay, including: based on the first target historical data set, the impedance identification neural network model is trained by adopting a gradient descent method and introducing weight decay.

[0010] In a possible implementation, the impedance identification neural network model is trained based on the first target historical data set by adopting a gradient descent method and introducing weight decay, including: constructing a data loader; based on the data loader, the first target historical data set is batched to obtain multiple second target historical data sets, each of the second target historical data sets includes at least two target historical operating parameter sets and a historical impedance data set corresponding to each of the target historical parameter sets, and the target historical operating parameter set is any of the historical operating parameter sets; adjusting hyperparameters, the hyperparameters include one or more of training rounds, learning rate, batches, and weight decay; based on the data loader, the multiple second target historical data sets are input into the impedance identification neural network model one by one, and according to the hyperparameters, the impedance identification neural network model is trained by adopting a gradient descent method and introducing weight decay.

[0011] In a possible implementation, after the data loader inputs the multiple second target historical data sets into the impedance identification neural network model one by one, and trains the impedance identification neural network model according to the hyperparameters by adopting a gradient descent method and introducing weight decay, the method further includes: based on the first target historical data set, using L1 norm loss to calculate the loss value of the impedance identification neural network model; based on the loss value, evaluating the credibility of the impedance identification neural network model.

[0012] In a possible implementation, the obtaining the first historical data set includes: asynchronously reading the first historical data set in a multi-threaded manner.

[0013] In a second aspect, an embodiment of the present application provides an impedance identification device for an electric power equipment, the device comprising: a parameter acquisition module, used to acquire a first operating condition parameter set of a target electric power equipment, the first operating condition parameter set comprising a target parameter set of the target electric power equipment under each of a plurality of operating conditions; an impedance identification module, used to input the first operating condition parameter set into a trained impedance identification neural network model to obtain impedance data of the target electric power equipment under each of the operating conditions.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described in the first aspect or any one of the implementation methods thereof is implemented.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect or any one of the implementation methods thereof is implemented.

[0016] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the method described in the first aspect or any one of the implementation methods thereof.

[0017] Compared with the prior art, the embodiment of the present application has the beneficial effect that the target parameter set of the target power equipment under each of the multiple working conditions is input into the trained impedance identification neural network model to directly obtain the impedance data of the target power equipment under each working condition. Compared with using physical formulas to calculate the impedance data of the power equipment under each different working condition, or using a neural network model for impedance identification, only the parameter set of the target power equipment under one working condition can be input each time to obtain the impedance data of the target power equipment under this working condition, thereby improving the efficiency of power equipment impedance identification.

[0018] It can be understood that the power equipment impedance identification device, electronic device, computer-readable storage medium and computer program product provided in the embodiments of the present application have the same beneficial effects as the above-mentioned power equipment impedance identification method, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A schematic diagram of a flow chart of a method for identifying impedance of an electric device provided in one embodiment of the present application;

[0021] Figure 2 A schematic diagram of the principle of impedance identification using an impedance identification neural network model provided in one embodiment of the present application;

[0022] Figure 3 A flowchart of a training method for an impedance identification neural network model provided in one embodiment of the present application;

[0023] Figure 4 A flowchart of another method for training an impedance identification neural network model provided in one embodiment of the present application;

[0024] Figure 5 A structural block diagram of an electrical equipment impedance identification device provided in one embodiment of the present application;

[0025] Figure 6 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0026] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0027] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0028] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0029] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0030] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0031] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0032] To facilitate understanding, some concepts involved in the embodiments of the present application are first explained.

[0033] Neural Networks: Neural network models consist of interconnected nodes, or neurons, that process and learn from data, supporting tasks such as pattern recognition and decision making in machine learning. Neural networks, also known as artificial neural networks, are a computing architecture based on a model of human brain function, consisting of a group of processing units called "nodes" that pass data to each other, similar to how neurons in the brain pass electrical impulses to each other.

[0034] Large Model: refers to the large and complex algorithmic models used in machine learning. They can process and analyze large amounts of data and are used for various tasks such as image recognition and natural language processing.

[0035] Dataset: refers to a group of related data samples, usually used in machine learning, data mining, statistical analysis and other fields. It is usually divided into three subsets: training set, validation set and test set. The training set is used to train the machine learning model, the validation set is used to select and adjust the hyperparameters and structure of the model, and the test set is used to evaluate the performance and accuracy of the model. The data sets in this application all take operating data and impedance data as examples.

[0036] Emergence: When a complex system is composed of many tiny individuals, these tiny individuals come together and interact with each other. When the number is large enough, special phenomena that cannot be explained by microscopic individuals will appear at the macro level. This is called "emergence phenomenon."

[0037] Standardization: Scale the data so that it falls into a small specific range, remove the unit restrictions of the data, and convert it into a dimensionless pure value.

[0038] Normalization: Map the data to the range of 0 to 1, turning it into a decimal between (0, 1), making subsequent data processing more convenient and faster.

[0039] Learning rate: The learning rate in deep learning is a crucial hyperparameter that determines the speed and direction of the model's weight parameter updates during training. The learning rate plays a core role when using the gradient descent optimization algorithm model. If it is set too large, it may cause oscillation or divergence; conversely, the speed of obtaining the optimal solution will be very slow or the local minimum, rather than the global optimum.

[0040] The technical solution of the present application will be described in detail below with reference to the accompanying drawings.

[0041] Figure 1 This is a flow chart of a method for identifying impedance of an electric device provided in an embodiment of the present application. For the sake of convenience, only the part related to the present embodiment is shown. The method provided in the present embodiment includes the following steps:

[0042] S110, obtaining a first operating condition parameter set of a target electric power device, where the first operating condition parameter set includes a target parameter set of the target electric power device in each operating condition of a plurality of operating conditions.

[0043] Specifically, the target power equipment is any power equipment in the power system, and the target parameter set corresponding to each operating condition of the target power equipment under multiple operating conditions in the first operating condition parameter set includes but is not limited to the effective value of the line voltage, three-phase voltage phase, effective reference value of the line voltage, three-phase active power, three-phase reactive power, fundamental frequency and disturbance frequency of the target power equipment under the corresponding operating condition.

[0044] In a specific implementation, various sensors in the target power equipment are used to collect the operating parameters of the target power equipment under various operating conditions in real time, and all the operating parameters corresponding to each operating condition are used as the target parameter set of the target power equipment under the operating condition, and all the target parameter sets corresponding to the target power equipment under multiple different operating conditions are used as the first operating parameter set.

[0045] S120, inputting the first operating condition parameter set into a trained impedance identification neural network model to obtain impedance data of the target power equipment under each operating condition.

[0046] Specifically, the impedance data of the target power equipment under each operating condition includes the amplitude phases of four sequences of impedance / admittance of the target power equipment under each operating condition, which are positive sequence_positive sequence, positive sequence_negative sequence, negative sequence_positive sequence and negative sequence_negative sequence.

[0047] In a specific implementation, firstly, based on the impedance identification neural network model, the first operating condition parameter set is operated in batches to obtain multiple second operating condition parameter sets, each of which includes at least two target parameter sets, and then the multiple second operating condition parameter sets are input one by one into the trained impedance identification neural network model to obtain the impedance data of the target power equipment under each operating condition. Among them, the first operating condition parameter set of the target power equipment is input one by one in batches into the trained impedance identification neural network model for calculation, which improves the calculation efficiency.

[0048] As an example, the first operating condition parameter set of the target power equipment is stored in the form of an array, and each target parameter set occupies one row of the array. When performing batch operations on the first operating condition parameter set based on the impedance identification neural network model, the first target operating condition parameter set is read in a row-by-row manner to ensure that all parameters of the target power equipment under one operating condition are grouped into the same batch.

[0049] As an example, before inputting the first operating condition parameter set into the trained impedance identification neural network model, the first operating condition parameter set is first standardized or normalized before performing subsequent operations.

[0050] Exemplarily, the standardization process is (original data - mean) / standard deviation; the normalization process is (original value - minimum value) / (maximum value - minimum value).

[0051] As an example, Figure 2 As shown, the operating data X of the target power equipment under any operating condition 1 , X 2 ...X n Input to the input layer of the trained impedance identification neural network model, after being processed by multiple hidden layers such as Layer1 and Layer2, the impedance data Y of the target power equipment under the working condition is output in the output layer LayerL. 1 , Y 2 ...Y n .

[0052] The technical solution provided by the present application inputs the acquired target parameter set of the target power equipment under each working condition in multiple working conditions into the trained impedance identification neural network model, and directly obtains the impedance data of the target power equipment under each working condition. Compared with using physical formulas to calculate the impedance data of the power equipment under each different working condition, or using a neural network model for impedance identification, only the parameter set of the target power equipment under one working condition can be input each time to obtain the impedance data of the target power equipment under this working condition, which improves the efficiency of power equipment impedance identification.

[0053] Figure 3 A flowchart of a training method for an impedance identification neural network model provided in one embodiment of the present application. Figure 3 As shown, based on the above embodiment, this embodiment further illustrates and optimizes the technical solution. Specifically, in this embodiment, the training process of the impedance identification neural network model includes:

[0054] S310, building an impedance identification neural network model based on the funnel model.

[0055] In the specific implementation, according to the large model thinking and model emergence phenomenon, a relatively large-scale neural network model is built based on the funnel model as an impedance identification neural network model. The impedance identification neural network model provided in this application has many neurons at the entrance, and then decreases step by step.

[0056] S320, obtaining a first historical data set, the first historical data set including a plurality of historical operating condition parameter sets and historical impedance data sets corresponding to each historical operating condition parameter set, the historical operating condition parameter set including a historical parameter set of the target power equipment under any operating condition.

[0057] In a possible implementation, the first historical data set is read asynchronously in a multi-threaded manner.

[0058] In a specific implementation, the first historical data set is stored in the form of an array, and each historical operating parameter set and a corresponding historical impedance data set occupy a row of the array; the number of cores of the central processing unit (CPU) of the computing device is obtained, an asynchronous thread pool is created, and data is read according to each row in the array storing the first historical data set until all data in the first historical data set is read and the process is terminated.

[0059] S330, based on the first historical data set, adopting a gradient descent method and introducing weight decay to train the impedance identification neural network model.

[0060] Specifically, the first historical data set is divided into a training set, a validation set and a test set according to a preset ratio. The training set is used to train the impedance identification neural network model, the validation set is used to select and adjust the hyperparameters and structure of the impedance identification neural network model, and the test set is used to evaluate the performance and accuracy of the impedance identification neural network model, so that each layer of neurons in the impedance identification neural network model after training has appropriate weights and biases.

[0061] In a possible implementation, the first historical data set is first standardized or normalized to obtain a first target historical data set, and then based on the first target historical data set, the impedance identification neural network model is trained by using a gradient descent method and introducing weight decay.

[0062] In one possible implementation, a data loader is constructed, and based on the data loader, a first target historical data set is operated in batches to obtain multiple second target historical data sets, each of which includes at least two target historical operating parameter sets and historical impedance data sets corresponding to each target historical parameter set, and the target historical operating parameter set is any historical operating parameter set; then, hyperparameters are adjusted, and the hyperparameters include one or more of training rounds, learning rate, batches, and weight decay; finally, based on the data loader, the multiple second target historical data sets are input one by one into the impedance identification neural network model, and according to the hyperparameters, the impedance identification neural network model is trained by adopting a gradient descent method and introducing weight decay.

[0063] As an example, if all the training set data is put into the impedance identification neural network model at once to update the parameters during model training, not only will the cycle of each parameter update be extended, there is a risk of memory overflow, but the trained impedance identification neural network model will be more affected by the noise in the data set itself. To this end, the embodiment of the present application constructs a data loader DataLoader to perform batch operations on the first target historical data set to obtain multiple second target historical data sets, and then send the multiple second target historical data sets one by one into the impedance identification neural network model for training. Among them, the data loader DataLoader is mainly used to perform two key tasks: data loading and batch processing. Its main advantages are as follows:

[0064] Improve computing efficiency: When the amount of data in the training set is very large, loading all the data in the training set into memory at once for model training may cause memory overflow. By dividing the training set into multiple batches, part of the data can be used for model training in each batch, thus saving memory resources;

[0065] Reduce the impact of noise: For noisy data, if all of them are processed at once, it may have a great impact on the model, resulting in reduced performance of the impedance identification neural network model. Processing data in batches can reduce the impact of noise data on the model.

[0066] Correct back propagation: During the training of the impedance identification neural network model, a back propagation algorithm is required to adjust the weights of the neurons, which requires calculating the gradient of the loss function with respect to the weights. If batch training is adopted, the gradient can be calculated and the weights can be updated in each batch, which can speed up the training process. At the same time, the gradient direction of each batch can also provide a certain degree of randomness, which helps the model to jump out of the local minimum and find better optimization results.

[0067] Better fit to data patterns: Some potential patterns learned by the impedance identification neural network model in one batch can be used in the training of the next batch, so as to achieve better results in the training of the next batch.

[0068] As an example, the hyperparameters are divided into training round epoch, learning rate learning_rate, batch batch_size, weight decay weight_decay. Use OneCycleLR to automatically adjust the learning rate, set the initial learning rate div_factor, the final learning rate final_div_factor, and the proportion of the number of iterations (steps) of the learning rate increase pct_start.

[0069] Specifically, OneCycleLR is a learning rate scheduler that dynamically adjusts the learning rate during training to improve model convergence and accuracy. It speeds up model training and prevents the model from falling into a local minimum by gradually increasing and then decreasing the learning rate during training. OneCycleLR takes into account the rising and falling trends of the learning rate, so as to better control the learning rate during training.

[0070] As an example, the AdamW optimization algorithm is used to train the impedance identification neural network model.

[0071] Specifically, the AdamW optimization algorithm adds weight decay regularization to the Adam optimization algorithm. In the Adam optimization algorithm, weight decay is directly added to the gradient:

[0072] g t =g t +λθ t-1 ,

[0073] Among them, g tis the gradient of the current step, θ t-1 is the model weight in the previous step, and λ is the regularization coefficient.

[0074] In the AdamW optimization algorithm, regularization becomes:

[0075] θ t =θ t-1 -γλθ t-1 ,

[0076] Among them, θ t is the model weight in the current step, θ t-1 is the model weight in the previous step, λ is the regularization coefficient, and γ is the attenuation coefficient.

[0077] In a possible implementation, multiple second target historical data sets are input one by one into the impedance identification neural network model based on a data loader, and the impedance identification neural network model is trained according to hyperparameters by using a gradient descent method and introducing weight decay. Then, based on the first target historical data set, an L1 norm loss is used to calculate the loss value of the impedance identification neural network model; based on the loss value, the credibility of the impedance identification neural network model is evaluated.

[0078] In a specific implementation, based on a test set divided from the first target historical data set according to a preset ratio, the L1 norm loss is used to calculate the loss value of the impedance identification neural network model. The smaller the calculated loss value, the higher the credibility of the trained impedance identification neural network model.

[0079] As an example, the L1 norm loss is used to evaluate the credibility of the impedance identification neural network model, that is, the absolute value loss of the predicted result and the actual result of the impedance identification neural network model is calculated based on the following formula:

[0080]

[0081] Among them, loss(x, y) represents the loss value of the impedance identification neural network model, n represents the number of samples, x represents the predicted value of the sample, and y represents the true value of the sample.

[0082] The technical solution provided in this embodiment builds an impedance identification neural network model based on the funnel model according to the big model thinking and model emergence phenomenon, and trains the impedance identification neural network model based on the acquired first historical data set of the target power equipment by adopting the gradient descent method and introducing weight decay, thereby improving the accuracy of the trained impedance identification neural network model.

[0083] Figure 4 A flowchart of another method for training an impedance identification neural network model provided in an embodiment of the present application. Figure 4 As shown, based on the above embodiment, this embodiment further illustrates and optimizes the technical solution. Specifically, in this embodiment, the training process of the impedance identification neural network model includes:

[0084] Step 1: Build an impedance identification neural network model.

[0085] Specifically, according to the big model thinking and model emergence phenomenon, a relatively large-scale neural network model is built based on the funnel model as the impedance identification neural network model.

[0086] Step 2: Read the data set.

[0087] Specifically, a data set of a target power device is asynchronously read in a multi-threaded manner, where the data set includes historical operating data of the target power device under different operating conditions and corresponding historical impedance data.

[0088] Step three: standardize the data set.

[0089] Specifically, the data set is standardized or normalized.

[0090] Step 4: Build the data loader.

[0091] Specifically, data loading and batch processing operations are performed by building a data loader.

[0092] Step 5: Adjust hyperparameters.

[0093] Specifically, the hyperparameters are divided into training round epoch, learning rate learning_rate, batch batch_size, weight decay weight_decay. Use OneCycleLR to automatically adjust the learning rate, set the initial learning rate div_factor, the final learning rate final_div_factor, and the proportion of the number of iterations (steps) of the learning rate increase pct_start.

[0094] Step 6: Input the data in the data loader batch by batch into the impedance identification neural network model.

[0095] Step seven: train the impedance identification neural network model.

[0096] Specifically, the AdamW optimization algorithm is used to train the impedance identification neural network model through gradient descent and the introduction of weight decay.

[0097] Step eight, determine whether the accuracy of the impedance identification neural network model meets the requirements, if yes, execute step nine; if not, execute step five.

[0098] Specifically, the L1 norm loss is used to calculate the loss value of the impedance identification neural network model. When the calculated loss value is less than the preset threshold, it is determined that the accuracy of the impedance identification neural network model meets the requirements, and step nine is executed; when the calculated loss value is greater than or equal to the preset threshold, it is determined that the accuracy of the impedance identification neural network model does not meet the requirements, and step five is re-executed.

[0099] Step nine, save the impedance identification neural network model.

[0100] The technical solution provided in this embodiment reduces data differences by standardizing or normalizing the data set; automatically adjusts the learning rate, AdamW optimization algorithm and L1 norm loss to improve the accuracy of the impedance identification results, thereby ensuring the accuracy of the impedance identification results while improving the efficiency of impedance identification. It should be noted that the present application can ensure that the mean square error accuracy of the impedance identification results is below 0.005.

[0101] In summary, the technical solution provided by this application has the following beneficial effects:

[0102] 1. Quickly identify impedance.

[0103] Under large-scale working conditions, it is only necessary to input all the working parameters under different working conditions into the trained impedance identification neural network model, and the impedance data under each working condition can be obtained in a very short time. Especially when the computer contains a graphics processing unit (GPU), the efficiency will be further improved, but this application does not limit the use of a GPU.

[0104] 2. The required technical threshold is low.

[0105] It is only necessary to know the input and output data of the target power equipment to train the constructed neural network and obtain the impedance identification neural network model, without the need for professional skills related to power electronics.

[0106] 3. Strong scalability

[0107] When there are other factors that may affect the output results, the factors can be further trained based on the existing impedance identification neural network model, so that the new impedance identification neural network model has the ability to identify new influencing factors.

[0108] Figure 5 This is a structural block diagram of an electrical equipment impedance identification device provided by an embodiment of the present application. For the sake of convenience, only the part related to the embodiment of the present application is shown. Figure 5 The power equipment impedance identification device 500 may include a parameter acquisition module 501 and an impedance identification module 502 .

[0109] The parameter acquisition module 501 is used to acquire a first operating condition parameter set of a target power device, where the first operating condition parameter set includes a target parameter set of the target power device in each operating condition among a plurality of operating conditions.

[0110] The impedance identification module 502 is used to input the first operating condition parameter set into the trained impedance identification neural network model to obtain the impedance data of the target power equipment under each operating condition.

[0111] An electrical equipment impedance identification device provided in an embodiment of the present application has the same beneficial effects as the above-mentioned electrical equipment impedance identification method.

[0112] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0113] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0114] Figure 6 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 6 As shown, the electronic device 6 of this embodiment includes: at least one processor 60 ( Figure 6 Only one is shown in the figure), a memory 61, and a computer program 62 stored in the memory 61 and executable on at least one processor 60, the processor 60 executes the computer program 62 to implement the above Figure 1 , Figure 3 , Figure 4 The steps in the method embodiment, or the implementation of the above Figure 5 Functions of each module / unit in the device embodiment.

[0115] The electronic device 6 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 6 may include but is not limited to a processor 60 and a memory 61. Those skilled in the art will appreciate that Figure 6 It is only an example of the electronic device 6 and does not constitute a limitation on the electronic device 6. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0116] The processor 60 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. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0117] In some embodiments, the memory 61 may be an internal storage unit of the electronic device 6, such as a hard disk or memory of the electronic device 6. In other embodiments, the memory 61 may also be an external storage device of the electronic device 6, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 6. Further, the memory 61 may also include both an internal storage unit and an external storage device of the electronic device 6. The memory 61 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as program codes of a computer program. The memory 61 may also be used to temporarily store data that has been output or is to be output.

[0118] The embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0119] If the integrated unit 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 application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and 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. 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. The computer-readable medium may include at least: any entity or device capable of carrying the computer program code to an electronic device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electric carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk.

[0120] A computer-readable storage medium provided in an embodiment of the present application has the same beneficial effects as the above-mentioned method for identifying impedance of an electric power device.

[0121] An embodiment of the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0122] A computer program product provided in an embodiment of the present application has the same beneficial effects as the above-mentioned method for identifying impedance of electric power equipment.

[0123] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0124] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0125] In the embodiments provided in the present application, it should be understood that the disclosed devices / equipment and methods can be implemented in other ways. For example, the device / equipment embodiments described above are merely schematic, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0126] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0127] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for identifying impedance of electric power equipment, characterized in that: The method comprises: Acquire a first operating condition parameter set of a target electric device, wherein the first operating condition parameter set includes a target parameter set of the target electric device under each of a plurality of operating conditions; The first operating condition parameter set is input into a trained impedance identification neural network model to obtain impedance data of the target power equipment under each of the operating conditions.

2. The method according to claim 1, characterized in that The step of inputting the first operating condition parameter set into a trained impedance identification neural network model to obtain impedance data of the target power equipment under each of the operating conditions includes: Based on the impedance identification neural network model, the first operating condition parameter set is operated in batches to obtain a plurality of second operating condition parameter sets, each of which includes at least two of the target parameter sets; The plurality of second operating condition parameter sets are input one by one into the trained impedance identification neural network model to obtain impedance data of the target power equipment under each of the operating conditions.

3. The method according to claim 1, characterized in that The training process of the impedance identification neural network model includes: Building the impedance identification neural network model based on the funnel model; Acquire a first historical data set, the first historical data set comprising a plurality of historical operating condition parameter sets and a historical impedance data set corresponding to each of the historical operating condition parameter sets, the historical operating condition parameter set comprising a historical parameter set of the target power equipment under any of the operating conditions; Based on the first historical data set, the impedance identification neural network model is trained by adopting a gradient descent method and introducing weight decay.

4. The method according to claim 3, characterized in that Before training the impedance identification neural network model based on the first historical data set by adopting a gradient descent method and introducing weight decay, the method further includes: Standardizing or normalizing the first historical data set to obtain a first target historical data set; Accordingly, based on the first historical data set, the impedance identification neural network model is trained by adopting a gradient descent method and introducing weight decay, including: Based on the first target historical data set, the impedance identification neural network model is trained by adopting a gradient descent method and introducing weight decay.

5. The method according to claim 4, characterized in that: The method of training the impedance identification neural network model based on the first target historical data set by adopting a gradient descent method and introducing weight decay includes: Build the data loader; Based on the data loader, the first target historical data set is operated in batches to obtain multiple second target historical data sets, each of which includes at least two target historical operating parameter sets and historical impedance data sets corresponding to each of the target historical parameter sets, and the target historical operating parameter set is any of the historical operating parameter sets; Adjusting hyperparameters, wherein the hyperparameters include one or more of training rounds, learning rate, batch, and weight decay; Based on the data loader, the multiple second target historical data sets are input into the impedance identification neural network model one by one, and the impedance identification neural network model is trained according to the hyperparameters by adopting a gradient descent method and introducing weight decay.

6. The method according to claim 5, characterized in that After the plurality of second target historical data sets are input one by one into the impedance identification neural network model based on the data loader, and the impedance identification neural network model is trained according to the hyperparameters by adopting a gradient descent method and introducing weight decay, the method further includes: Based on the first target historical data set, using L1 norm loss, calculating the loss value of the impedance identification neural network model; Based on the loss value, the credibility of the impedance identification neural network model is evaluated.

7. The method according to any one of claims 3 to 6, characterized in that: The obtaining of the first historical data set includes: The first historical data set is read asynchronously in a multi-threaded manner.

8. An impedance identification device for electric power equipment, characterized in that: The device comprises: A parameter acquisition module, configured to acquire a first operating condition parameter set of a target electric power device, wherein the first operating condition parameter set includes a target parameter set of the target electric power device under each of a plurality of operating conditions; The impedance identification module is used to input the first operating condition parameter set into a trained impedance identification neural network model to obtain the impedance data of the target power equipment under each of the operating conditions.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images