Electric energy subentry metering error detection model training method and metering error detection method

Abstract

The invention relates to an electric energy subentry metering error detection model training method and a metering error detection method. The electric energy subentry metering error detection model training method comprises the steps of obtaining an electric energy subentry metering signal sequence; performing wavelet packet transformation on the electric energy subentry metering signal sequence, and extracting a plurality of target electric energy subentry metering signals in a target characteristic frequency band from the electric energy subentry metering signal sequence; determining a model training sample based on the plurality of target electric energy sub-metering signals; and taking the model training sample as input, calling a pre-constructed initial metering error detection model, iteratively training the initial metering error detection model, and adjusting model parameters of the initial metering error detection model until a preset training ending condition is met, thereby obtaining a trained metering error detection model. By adopting the method, the accuracy of electric energy subentry metering error detection can be improved.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for training an electric energy sub-item metering error detection model, and a method, device, computer equipment, computer-readable storage medium and computer program product for detecting electric energy sub-item metering errors. Background Art

[0002] The accuracy of electric energy measurement is crucial to ensure the stable operation of the power grid, achieve fair transactions in the power market, and improve energy efficiency. Electric energy sub-item measurement refers to the classification and measurement of different types of electric energy consumption, which is of great significance in energy saving, energy management, and optimization of power distribution.

[0003] However, the sub-item metering of electric energy may be affected by multiple factors. For example, during operation, the sub-item metering device of electric energy may be affected by factors such as the external ambient temperature, current size and voltage size, which may lead to metering errors.

[0004] At present, the traditional electric energy metering error detection method has the problem of low detection accuracy, which makes it difficult to meet the needs of high-precision electric energy measurement. Summary of the invention

[0005] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for training an electric energy sub-metering error detection model that can improve the accuracy of electric energy sub-metering error detection in distribution networks, as well as a method, device, computer equipment, computer-readable storage medium and computer program product for detecting electric energy sub-metering errors in response to the above-mentioned technical problems.

[0006] In a first aspect, the present application provides a method for training an electric energy sub-item metering error detection model, comprising:

[0007] Obtain the electric energy sub-item metering signal sequence of the distribution network;

[0008] Performing wavelet packet transform on the electric energy sub-item metering signal sequence, and extracting multiple target electric energy sub-item metering signals in the target characteristic frequency band from the electric energy sub-item metering signal sequence;

[0009] Determine the model training samples based on multiple target electric energy sub-item metering signals;

[0010] Taking the model training samples as input, the pre-built initial metering error detection model is called to obtain the metering error prediction value of each target electric energy sub-item metering signal in the model training samples;

[0011] For each target electric energy sub-item metering signal, determine the difference between the metering error prediction value and the metering error actual value of the target electric energy sub-item metering signal;

[0012] Based on the difference value, the model parameters of the initial measurement error detection model are adjusted until the preset training end condition is reached to obtain a trained measurement error detection model.

[0013] In one embodiment, wavelet packet transform is performed on the electric energy sub-item metering signal sequence, and multiple target electric energy sub-item metering signals in a target characteristic frequency band are extracted from the electric energy sub-item metering signal sequence, including:

[0014] For each electric energy item metering signal in the electric energy item metering signal sequence, the following processing is performed:

[0015] Performing wavelet packet transform on the electric energy sub-item metering signal to decompose the electric energy sub-item metering signal into sub-band signals of multiple frequency bands;

[0016] Determine the energy value corresponding to each decomposed sub-band signal;

[0017] Determine the frequency band of the sub-band signal with the largest energy value as the target characteristic frequency band;

[0018] The target electric energy sub-item metering signal under the target characteristic frequency band is extracted from the electric energy sub-item metering signal.

[0019] In one embodiment, wavelet packet transform is performed on the electric energy sub-item metering signal to decompose the electric energy sub-item metering signal into sub-band signals of multiple frequency bands, including:

[0020] Based on a preset wavelet packet basis function, the electric energy sub-item metering signal is decomposed into sub-band signals of multiple frequency bands according to a preset wavelet packet transform decomposition layer number.

[0021] In one embodiment, adjusting the model parameters of the initial metrology error detection model based on the difference value includes:

[0022] Based on the difference value, the weight vector of the output layer, the center vector of the hidden layer and the width vector of the Gaussian basis function of the initial measurement error detection model are adjusted through the gradient descent algorithm.

[0023] In one embodiment, determining a model training sample based on a plurality of target electric energy sub-item metering signals includes:

[0024] For each target electric energy sub-item metering signal, perform the following steps:

[0025] Obtaining the measured value of the target electric energy sub-item metering signal;

[0026] Determine the difference between the measured value and the expected value of the electric energy sub-item metering signal to obtain the metering error value;

[0027] Based on the metering error value, an error label is added to the target electric energy sub-item metering signal;

[0028] Multiple target electricity sub-item metering signals with error labels added are integrated to obtain model training samples.

[0029] In one embodiment, obtaining a signal sequence of electric energy item metering of a distribution network includes:

[0030] Adjust the effective number of bits of the analog-to-digital converter of the electric energy sub-metering device;

[0031] The electric energy sub-item metering signal sequence collected by the adjusted electric energy sub-item metering device is obtained.

[0032] In a second aspect, the present application also provides a method for detecting an error in electric energy sub-item metering, comprising:

[0033] Obtain electric energy sub-item metering signal sequence;

[0034] Taking the electric energy sub-item metering signal sequence as input, calling the trained metering error detection model, detecting the metering error in the electric energy sub-item metering signal sequence, and obtaining the metering error value in the electric energy sub-item metering signal sequence, the trained metering error detection model is trained based on the above-mentioned electric energy sub-item metering error detection model training method.

[0035] In a third aspect, the present application also provides a device for training an electric energy sub-item metering error detection model, comprising:

[0036] A data acquisition module is used to obtain the electric energy sub-item metering signal sequence of the distribution network;

[0037] A signal extraction module is used to perform wavelet packet transform on the electric energy sub-item metering signal sequence, and extract multiple target electric energy sub-item metering signals in a target characteristic frequency band from the electric energy sub-item metering signal sequence;

[0038] A training sample determination module, used to determine a model training sample based on multiple target electric energy sub-item metering signals;

[0039] The metering error prediction module is used to take the model training sample as input, call the pre-built initial metering error detection model, and obtain the metering error prediction value of each target electric energy sub-item metering signal in the model training sample;

[0040] A difference value determination module is used to determine the difference between the metering error prediction value and the metering error actual value of the target electric energy sub-item metering signal for each target electric energy sub-item metering signal;

[0041] The parameter adjustment module is used to adjust the model parameters of the initial measurement error detection model based on the difference value until the preset training end condition is reached to obtain the trained measurement error detection model.

[0042] In a fourth aspect, the present application further provides an electric energy sub-item metering error detection device, comprising:

[0043] A signal acquisition module is used to obtain the electric energy sub-item metering signal sequence of the distribution network;

[0044] The metering error detection module is used to take the electric energy sub-item metering signal sequence as input, call the trained metering error detection model, detect the metering error in the electric energy sub-item metering signal sequence, and obtain the metering error value in the electric energy sub-item metering signal sequence. The trained metering error detection model is trained based on the above-mentioned electric energy sub-item metering error detection model training method.

[0045] In a fifth aspect, the present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the steps in any one of the above-mentioned embodiments of the training method for the electric energy sub-item metering error detection model, and the steps in an embodiment of the electric energy sub-item metering error detection method.

[0046] In a sixth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in any one of the above-mentioned embodiments of the training model for detecting errors in sub-item electric energy metering, and the steps in an embodiment of the method for detecting errors in sub-item electric energy metering.

[0047] In the seventh aspect, the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps in any one of the above-mentioned embodiments of the training model for electric energy sub-item metering error detection method, and the steps in an embodiment of the electric energy sub-item metering error detection method.

[0048] The above-mentioned electric energy sub-item metering error detection model training method, device, computer equipment, computer-readable storage medium and computer program product, on the one hand, obtain the electric energy sub-item metering signal sequence of the distribution network, and perform wavelet packet transform on the electric energy sub-item metering signal sequence, so as to be able to analyze the frequency characteristics in the metering signal at multiple scales. Through the characteristics of high and low frequency separation during processing, noise information is eliminated during signal analysis and processing, so as to accurately extract the electric energy sub-item metering signal in the characteristic frequency band. According to the electric energy sub-item metering signal in the characteristic frequency band, a model training sample is constructed, and the training sample is used as input to iteratively train the pre-constructed initial metering error detection model, which is conducive to improving the quality of model training and error detection performance. On the other hand, the model outputs a metering error prediction value according to the electric energy sub-item metering signal in the characteristic frequency band extracted by wavelet packet transform, and adjusts the model parameters according to the difference between the metering error prediction value and the actual metering error value until the preset training end condition is reached to obtain a trained metering error detection model. The whole scheme processes complex and non-stationary electric energy sub-item metering signals through wavelet packet transform and trains the metering error model, which can make the model better adapt to the influence of various interferences and harmonics, thereby helping to improve the accuracy of metering error detection.

[0049] The above-mentioned electric energy sub-item metering error detection method, device, computer equipment, computer-readable storage medium and computer program product obtain the electric energy sub-item metering signal sequence of the distribution network, input it into the trained metering error detection model obtained by the above-mentioned electric energy sub-item metering error detection model training method, and detect the metering error in the electric energy sub-item metering signal. On the one hand, the accuracy and efficiency of electric energy sub-item metering error detection are improved through model detection. On the other hand, based on the metering error detection result, it is conducive to analyzing the deviation of each power indicator in the actual measurement, and then it is conducive to taking corresponding measures according to the metering error, so as to realize the accurate measurement and error analysis of each power indicator in the distribution network, and can provide strong support for the calibration and optimization of the power metering system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions 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 related drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A diagram showing an application environment of a method for training an electric energy sub-item metering error detection model in an embodiment;

[0052] Figure 2It is a flowchart of a method for training an electric energy sub-item metering error detection model in one embodiment;

[0053] Figure 3 It is a flowchart of a method for training an electric energy sub-item metering error detection model in another embodiment;

[0054] Figure 4 It is a flowchart of a method for training an electric energy sub-item metering error detection model in another embodiment;

[0055] Figure 5 A schematic diagram of a flow chart of a method for training an electric energy sub-item metering error detection model in yet another embodiment;

[0056] Figure 6 It is a structural block diagram of a device for training an electric energy sub-item metering error detection model in one embodiment;

[0057] Figure 7 It is a structural block diagram of an electric energy sub-item metering error detection device in one embodiment;

[0058] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0060] The electric energy sub-item metering error detection model training method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the terminal 102 communicates with the server 104 through a network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers.

[0061] Specifically, the operator may upload the collected electric energy sub-item metering signal sequence of the distribution network to the server 104 through the terminal 102, and then send the model training message to the server 104 through the terminal 102, and the server 104 obtains the electric energy sub-item metering signal sequence of the distribution network. Secondly, the electric energy sub-item metering signal sequence is subjected to wavelet packet transform, and multiple target electric energy sub-item metering signals in the target characteristic frequency band are extracted from the electric energy sub-item metering signal sequence. Based on the multiple target electric energy sub-item metering signals, the model training samples are determined, and then, the pre-built initial metering error detection model is called with the model training samples as input to obtain the metering error prediction value of each target electric energy sub-item metering signal in the model training sample, and for each target electric energy sub-item metering signal, the difference value between the metering error prediction value and the actual value of the metering error of the target electric energy sub-item metering signal is determined. Finally, based on the difference value, the model parameters of the initial metering error detection model are adjusted until the preset training end condition is reached to obtain the trained metering error detection model.

[0062] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. Head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server 104 may be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0063] In an exemplary embodiment, Figure 2 As shown in the figure, a method for training an electric energy sub-item metering error detection model is provided. Figure 1 The server 104 in the example is used for explanation, and includes the following S100 to S600. Among them:

[0064] S100, obtaining a signal sequence of electric energy item metering of a distribution network.

[0065] Among them, the electric energy sub-item metering signals are nonlinear voltage and current signals.

[0066] In practical applications, the electric energy sub-item metering device can collect the electric energy sub-item metering signals of electrical equipment within a historical period (such as half a year) to obtain the electric energy sub-item metering signal sequence. Among them, the electric energy sub-item metering device is a device used to separately record and measure the electric energy consumption of different equipment or different purposes, and can monitor and record the electric energy consumption in real time.

[0067] S200, performing wavelet packet transform on the electric energy sub-item metering signal sequence, and extracting a plurality of target electric energy sub-item metering signals in a target characteristic frequency band from the electric energy sub-item metering signal sequence.

[0068] Among them, wavelet packet transform is a multi-resolution analysis method that can decompose the signal in a finer granularity in the frequency domain. The target characteristic frequency band represents the frequency band containing the characteristic information of the electric energy sub-item metering.

[0069] In practical applications, the wavelet basis function can be selected according to the type and characteristics of the analyzed signal, such as wavelet basis functions including but not limited to Haar wavelet basis, Daubechies wavelet basis, Symlets wavelet basis and Coiflets wavelet basis. The number of layers of wavelet packet decomposition is determined according to the signal characteristics and the frequency range to be extracted. The more decomposition layers, the higher the frequency resolution, but the amount of calculation will also increase. The electric energy sub-item metering signal sequence in the electric energy sub-item signal sequence is decomposed by wavelet packet decomposition, and the electric energy sub-item metering signal is decomposed into signals of different frequency bands. The target characteristic frequency band can be determined based on prior knowledge, so as to extract multiple target electric energy sub-item metering signals under the target characteristic frequency band through inverse wavelet packet transform.

[0070] S300, determining a model training sample based on a plurality of target electric energy sub-item metering signals.

[0071] In practical applications, the corresponding metering error can be calculated for each target electric energy sub-item metering signal, and a metering error label can be added to the corresponding target electric energy sub-item metering signal to obtain a model training sample. The metering error is the difference between the actual measurement value and the true value caused by various factors when measuring electric energy for each electrical device or branch in the distribution network. The metering error can be determined by comparing the difference between the actual measurement value and the standard reference value, such as generating a standard reference value through power simulation software, or using the value measured by a high-precision standard electric energy sub-item metering device as the standard reference value, obtaining the actual measurement value of the electric energy sub-item metering device, and comparing the difference between the actual measurement value and the standard reference value to determine the metering error.

[0072] S400, taking the model training sample as input, calling the pre-built initial metering error detection model, and obtaining the metering error prediction value of each target electric energy sub-item metering signal in the model training sample.

[0073] In practical applications, an initial metering error detection model is pre-built based on a radial basis neural network, and the model parameters are initialized. Specifically, the number of nodes in the input layer, output layer, and hidden layer of the radial basis neural network is defined, and the network parameters are initialized. The hidden layer uses the radial basis function as the activation function. The pre-built initial metering error detection model is called with the model training sample as input. Assume that the target electric energy sub-item metering signal in the input model training sample , , the radial basis function is a Gaussian basis function, then the hidden layer output of the jth unit is:

[0074]

[0075] Where c represents the center vector of the jth hidden layer neuron, represents the Euclidean distance, b represents the width vector of the Gaussian basis function, and m represents the number of nodes in the hidden layer.

[0076] The predicted value of the metering error of each target electric energy sub-item metering signal is obtained:

[0077]

[0078] In the formula, Represents the weight vector of the output layer.

[0079] S500, for each target electric energy sub-item metering signal, determining a difference between a metering error prediction value and an actual metering error value of the target electric energy sub-item metering signal.

[0080] The actual value of the metering error may be determined according to the metering error label of the target electric energy sub-item metering signal.

[0081] In practical applications, based on the performance indicators of the initial metering error detection model, the difference between the metering error prediction value and the actual value of the metering error of the target electric energy sub-item metering signal is determined:

[0082]

[0083] S600, based on the difference value, adjusting the model parameters of the initial measurement error detection model until a preset training end condition is reached, thereby obtaining a trained measurement error detection model.

[0084] In practical applications, after determining the difference value, the center vector can be optimized by a clustering algorithm, the optimal width parameter can be found by grid search, and the output weight vector can be solved by the least square method to minimize the difference value. The preset training end condition can be that the error is continuously less than the preset error threshold within the preset number of iterations, and the training of the initial metrology error detection model is stopped to obtain the trained metrology error detection model.

[0085] In the above-mentioned electric energy sub-item metering error detection model training method, on the one hand, the electric energy sub-item metering signal sequence is obtained, and the electric energy sub-item metering signal sequence is subjected to wavelet packet transformation, so that the frequency characteristics in the metering signal can be analyzed at multiple scales. Through the characteristics of high and low frequency separation during processing, the noise information is eliminated during signal analysis and processing, so that the electric energy sub-item metering signal under the characteristic frequency band is accurately extracted. According to the electric energy sub-item metering signal under the characteristic frequency band, the model training sample is constructed, and the training sample is used as input to iteratively train the pre-constructed initial metering error detection model, which is conducive to improving the quality of model training and error detection performance. On the other hand, the model outputs the metering error prediction value according to the electric energy sub-item metering signal under the characteristic frequency band extracted by wavelet packet transformation, and adjusts the model parameters according to the difference between the metering error prediction value and the actual value of the metering error until the preset training end condition is reached, and the trained metering error detection model is obtained. The whole scheme processes complex and non-stationary electric energy sub-item metering signals through wavelet packet transformation and trains the metering error model, which can make the model better adapt to the influence of various interferences and harmonics, thereby helping to improve the accuracy of metering error detection.

[0086] In order to improve the accuracy of the metering error analysis of the electric energy sub-item metering signal, in an exemplary embodiment, Figure 3 As shown, S200 includes S220 to S280. Among them:

[0087] S220, for each electric energy sub-item metering signal in the electric energy sub-item metering signal sequence, perform wavelet packet transform on the electric energy sub-item metering signal, and decompose the electric energy sub-item metering signal into sub-band signals of multiple frequency bands.

[0088] Among them, the sub-band signal is a signal of several sub-bands (sub-bands) formed by dividing the electric energy sub-item metering signal in the frequency domain.

[0089] In practical applications, the signal decomposition of the electric energy sub-item metering signal can be performed based on the wavelet packet theory, which can be based on the preset wavelet packet basis function, which satisfies the orthogonal wavelet packet basis. Defined as an orthogonal wavelet packet basis, the closed space of the function is , satisfy:

[0090]

[0091]

[0092] In the formula, k is the decomposition order, is the high-pass filter coefficient, is the low-pass filter coefficient, , that is, the two coefficients are orthogonal.

[0093] Consider the spatial decomposition of electric energy sub-item metering signals:

[0094] , ,

[0095] set up , .

[0096] make , then the decomposition of electric energy sub-item metering signal based on wavelet packet is expressed as:

[0097]

[0098]

[0099] in, and They respectively represent the decomposition sequence of multi-scale analysis of electric energy sub-item metering signals.

[0100] The closed space may refer to a space generated by a set of orthogonal wavelet packet basis functions. The orthogonal basis function set can ensure that the decomposition and reconstruction of the signal are lossless.

[0101] Set the number of decomposition layers so that the electric energy sub-item measurement signals can be in the square integrable space The orthogonal decomposition is performed in the above equations. The bandwidth of the electric energy sub-item metering signal can be divided into equal width subbands, among which The signal decomposition of wavelet packet transform is performed by using the preset wavelet packet basis function and decomposition layer number, and the electric energy sub-item metering signal is decomposed into sub-band signals, each sub-band signal corresponds to a specific frequency interval.

[0102] S240: Determine the energy value corresponding to each decomposed sub-band signal.

[0103] In practical applications, the energy value corresponding to each decomposed sub-band signal can be determined by the following formula:

[0104]

[0105] In the formula, represents the number of wavelet packet coefficients, express Wavelet packet decomposition coefficients of the subband.

[0106] S260: Determine the frequency band of the sub-band signal with the maximum energy value as the target characteristic frequency band.

[0107] In practical applications, considering that the sub-band signal in the frequency band with larger energy contains more complete electric energy sub-item metering information, the frequency band of the sub-band signal with the largest energy value is determined as the target characteristic frequency band.

[0108] In other embodiments, after obtaining sub-band signals of multiple frequency bands, not all sub-band signals are meaningful for analysis. Considering the sampling frequency of the electric energy sub-item metering signal, further parts of the frequency bands containing the main signal components are screened out. Specifically, according to the preset number of frequency bands and the frequency bands where the main signal components of the electric energy sub-item metering error signal are concentrated, the preset number of target frequency bands can be screened out from the frequency bands corresponding to the multiple sub-band signals. Exemplarily, 16 sub-band signals are obtained after decomposition. When the sampling frequency of the electric energy sub-item metering signal of the distribution network is 10kHz (kilohertz), the preset number of frequency bands is 5. Considering that the frequency bands from low frequency to medium frequency contain the main signal components, 5 target frequency bands that meet the low frequency to medium frequency are screened out from the frequency bands corresponding to the 16 sub-band signals. Determine the energy value of the sub-band signal corresponding to the target frequency band, and determine the frequency band of the sub-band signal with the largest energy value as the target characteristic frequency band.

[0109] S280, extracting a target electric energy sub-item metering signal in a target characteristic frequency band from the electric energy sub-item metering signal.

[0110] In practical applications, after determining the target characteristic frequency band, the target electric energy sub-item metering signal under the target characteristic frequency band is extracted from the electric energy sub-item metering signal, and the signal under the selected target characteristic frequency band can be reconstructed by inverse wavelet packet transform.

[0111] In this embodiment, wavelet packet transform is performed on each electric energy sub-item metering signal, and based on the energy values ​​of the decomposed sub-band signals of each frequency band, the target characteristic frequency band containing the most signal information is screened out, thereby extracting the electric energy sub-item metering signal under the target frequency band, which is beneficial to improving the accuracy and efficiency of signal analysis.

[0112] In order to improve the accuracy of characteristic frequency band extraction, in an exemplary embodiment, S220 includes S222.

[0113] S222, based on a preset wavelet packet basis function, decompose the electric energy sub-item metering signal into sub-band signals of multiple frequency bands according to a preset wavelet packet transform decomposition layer number.

[0114] Among them, the wavelet packet basis function is the core of the wavelet packet transform, which performs multi-scale refinement analysis on the signal through scaling and offset operations. The number of decomposition layers of the wavelet packet transform is the number of times the signal is decomposed in the wavelet packet transform.

[0115] In practical applications, wavelet packet decomposition is the process of filtering the electric energy sub-item metering signal through a bandpass filter. The bandwidth of each filter is .in, is the number of layers of wavelet packet decomposition, are the k nodes of wavelet packet decomposition. It is the frequency of the input electric energy sub-item metering signal.

[0116] The preset wavelet packet basis function may be the db4 wavelet in the Daubechies wavelet. The db4 wavelet is used to perform wavelet packet decomposition on the electric energy sub-item metering signal. The number of decomposition layers is set to 4, and a total of 16 sub-band signals are decomposed.

[0117] In this embodiment, based on the wavelet packet basis function, the electric energy sub-item metering signal is decomposed according to a preset wavelet packet transform decomposition layer number, which is conducive to signal analysis according to the decomposed sub-band signals.

[0118] In order to accelerate the convergence of the model, in one exemplary embodiment, S600 includes S620. In which:

[0119] S620, based on the difference value, adjusting the weight vector of the output layer, the center vector of the hidden layer, and the width vector of the Gaussian basis function of the initial measurement error detection model through a gradient descent algorithm.

[0120] In practical applications, based on the difference value, the weight vector of the output layer, the center vector of the hidden layer, and the width vector of the Gaussian basis function of the initial measurement error detection model are updated through the gradient descent algorithm:

[0121]

[0122] In the formula, represents the weight vector of the output layer, represents the center vector of the jth hidden layer neuron, Represents the width vector of the Gaussian basis function.

[0123] After updating, the weight vector of the initial measurement error detection model is determined and the weight vector is updated:

[0124]

[0125] in, represents the learning rate, ; represents the momentum coefficient, .

[0126] In this embodiment, the weight vector of the output layer, the center vector of the hidden layer and the width vector of the Gaussian basis function of the model are adjusted based on the difference value by gradient descent, which accelerates the convergence of the model and helps to improve the measurement error detection performance of the model.

[0127] Add data labels to the input data of the model. In one exemplary embodiment, Figure 4 As shown, S300 includes S320 to S380. Among them:

[0128] S320: For each target electric energy sub-item metering signal, obtain a measurement value of the target electric energy sub-item metering signal.

[0129] S340, determining the difference between the measured value and the expected value of the electric energy sub-item metering signal to obtain a metering error value.

[0130] S360: Add an error label to the target electric energy sub-item metering signal based on the metering error value.

[0131] S380, integrating multiple target electric energy sub-item metering signals with error labels added thereto to obtain model training samples.

[0132] In practical applications, for each target electric energy sub-item metering signal, the electric energy measurement value of the target electric energy sub-item metering signal can be obtained by comparing the voltage signal and the current signal in the target electric energy sub-item metering signal.

[0133] After obtaining the measurement of the target electric energy sub-item metering signal, the difference between the measured value and the expected value of the electric energy sub-item metering signal is determined to obtain a metering error value.

[0134] Considering that the sampling frequency of power sub-metering is limited, the waveform changes in the power grid meet the conditions of Fourier series expansion. The power sub-metering signal is decomposed by Fourier transform to obtain multiple sinusoidal solutions with different frequencies. If there is an nth harmonic, the expression of the average power of the nth harmonic is:

[0135]

[0136] In the formula, and They represent harmonic voltage and harmonic current respectively.

[0137] Assume that the current supplied by the power grid to the load is , where the superimposed AC component is , the current of the electric energy meter after filtering Defined as:

[0138]

[0139] in, represents the grid angular frequency, Indicates the number of load operation cycles.

[0140] If the current state of the grid bus voltage is ideal, the power bus voltage Equal to its voltage amplitude , the instantaneous power of the circuit is The double law is used to convert thermal energy into electrical energy, and the formula for the expected value of grid electricity sub-item measurement is as follows:

[0141]

[0142] Instantaneous power per unit cycle The average value is the active power , power bus voltage amplitude With superimposed AC components The average value of the product is 0. Define the active power of the electric energy sub-metering device The definition formula is as follows:

[0143]

[0144] in, Represents the independent variable of the integrand.

[0145] Based on the active power, i.e. the loss part, of the grid bus voltage under ideal conditions, the superimposed AC component is regarded as a reactive component that does not generate active power and is not accumulated in the sub-item metering of electric energy.

[0146] Combination The impact of sub-power harmonics on the power grid, perform harmonic related setting, set is the voltage amplitude, is the angular frequency, is the initial phase, then the bus voltage The formula is:

[0147]

[0148] The instantaneous power p of the electric energy sub-item metering circuit under subharmonics is:

[0149]

[0150] in, Represents a segment of energy measurement.

[0151] If within a unit period, the average power of the electric energy sub-item measurement is , then the active power based on harmonics for:

[0152]

[0153] It can be seen that the harmonic frequency of the grid voltage under non-ideal conditions is equal to the frequency of the reactive current. At this time, there will be The extra kilowatt-hour of electricity is the error in the electricity sub-item measurement. The calculation formula for the error in the electricity sub-item measurement is:

[0154]

[0155] in, Indicates the time length for energy metering.

[0156] It can be seen that when the harmonic voltage amplitude When it is low, the error of electric energy sub-item measurement It is almost small enough to be ignored. However, when the load is large, the error of electric energy sub-item measurement Likewise, the phase difference between voltage and current will also produce a measurement error that cannot be ignored.

[0157] After obtaining the metering error value corresponding to the target electric energy sub-item metering signal, an error label is added to the corresponding target electric energy sub-item metering signal based on the metering error value. Multiple target electric energy sub-item metering signals with error labels added are integrated to obtain model training samples.

[0158] In this embodiment, by obtaining the measurement value of the target electric energy sub-item metering signal, calculating the metering error value, adding an error label to the target electric energy sub-item metering signal, obtaining a model training sample, and training the metering error detection model with the model training sample, it is beneficial to improve the error detection accuracy of the metering error detection model.

[0159] In an exemplary embodiment, Figure 5 As shown, S100 includes S120 to S140. Among them:

[0160] S120, adjusting the effective number of bits of the analog-to-digital converter of the electric energy sub-item metering device.

[0161] S140, obtaining an electric energy sub-item metering signal sequence collected by the adjusted electric energy sub-item metering device.

[0162] Among them, the effective number of bits (ENOB) of the analog-to-digital converter (ADC) refers to the effective resolution that the ADC can achieve in practical applications. It takes into account the effects of quantization noise and dynamic range, as well as the effective number of bits that the ADC can provide in actual use.

[0163] The error of electric energy sub-item metering may be affected by the following factors: the phase error of the sampling transformer of the electric energy sub-item metering device, the phase error of the voltage sampling value during the interpolation process, and the error between the delay compensation value of the measured data processing and the actual delay. For the analog-to-digital converter of the electric energy sub-item metering device in the working state, its error is related to the change of the signal-to-noise ratio. The ideal signal-to-noise ratio of the analog-to-digital converter is:

[0164]

[0165] Where N represents the number of conversion bits.

[0166] However, in practice, it is difficult for the analog-to-digital converter of the electric energy sub-item metering combination unit to achieve an ideal signal-to-noise ratio. In order to accurately reflect the electric energy sub-item metering data, the effective number of bits of the analog-to-digital converter can be adjusted, and its expression is:

[0167]

[0168] Where S represents the signal-to-noise ratio.

[0169] After adjusting the effective number of bits of the analog-to-digital converter of the electric energy sub-item metering device, the electric energy sub-item metering signal sequence collected by the adjusted electric energy sub-item metering device is obtained.

[0170] In this embodiment, by adjusting the effective number of bits of the analog-to-digital converter of the electric energy sub-item metering device and collecting the electric energy sub-item metering signal sequence, the accuracy of the acquired electric energy sub-item metering data can be improved and the metering error can be reduced.

[0171] In order to make a clearer description of the electric energy sub-item metering error detection model training method provided in the present application, a specific embodiment is described below, and the specific embodiment includes the following steps:

[0172] S1, adjusting the effective number of bits of the analog-to-digital converter of the electric energy sub-item metering device to obtain the electric energy sub-item metering signal sequence collected by the adjusted electric energy sub-item metering device.

[0173] S2, for each electric energy sub-item metering signal in the electric energy sub-item metering signal sequence, based on a preset wavelet packet basis function and according to a preset number of wavelet packet transform decomposition layers, decomposing the electric energy sub-item metering signal into sub-band signals of multiple frequency bands.

[0174] S3, determining the energy value corresponding to each decomposed sub-band signal, determining the frequency band of the sub-band signal with the largest energy value as the target characteristic frequency band, and extracting the target electric energy sub-item metering signal under the target characteristic frequency band from the electric energy sub-item metering signal.

[0175] S4, for each target electric energy sub-item metering signal, obtain the measured value of the target electric energy sub-item metering signal, determine the difference between the measured value and the expected value of the electric energy sub-item metering signal, obtain the metering error value, add an error label to the target electric energy sub-item metering signal based on the metering error value, integrate multiple target electric energy sub-item metering signals with error labels added, and obtain model training samples.

[0176] S5, taking the model training sample as input, calling the pre-built initial metering error detection model, and obtaining the metering error prediction value of each target electric energy sub-item metering signal in the model training sample.

[0177] S6, for each target electric energy sub-item metering signal, determining a difference between a metering error prediction value and an actual metering error value of the target electric energy sub-item metering signal.

[0178] S7, based on the difference value, adjust the weight vector of the output layer, the center vector of the hidden layer and the width vector of the Gaussian basis function of the initial measurement error detection model through the gradient descent algorithm until the preset training end condition is reached to obtain the trained measurement error detection model.

[0179] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0180] The low voltage harmonic detection method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the terminal 102 communicates with the server 104 through a network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers.

[0181] Specifically, the operator may send an electric energy sub-item metering error detection message to the server 104 through the terminal, and the server receives the low electric energy sub-item metering error detection message. Secondly, the electric energy sub-item metering signal sequence is obtained, and the electric energy sub-item metering signal sequence is used as input to call the trained metering error detection model to detect the electric energy sub-item metering error and obtain the metering error value in the electric energy sub-item metering signal sequence.

[0182] In an exemplary embodiment, the present application further provides a method for detecting an error in electric energy sub-item metering, comprising the following S700 to S800.

[0183] S700: Obtain a signal sequence for electric energy item-by-item metering.

[0184] In practical applications, the electric energy sub-item metering signal sequence may be obtained through an electric energy sub-item metering device.

[0185] S800, taking the electric energy sub-item metering signal sequence as input, calling the trained metering error detection model, detecting the metering error in the electric energy sub-item metering signal sequence, and obtaining the metering error value in the electric energy sub-item metering signal sequence, wherein the trained metering error detection model is trained based on the above-mentioned electric energy sub-item metering error detection model training method.

[0186] In practical applications, the obtained electric energy sub-item metering signal sequence is input into the trained metering error detection model to detect the metering error in the electric energy sub-item metering signal sequence, and obtain the metering error in the electric energy sub-item metering signal sequence. The metering error detection model is obtained by training using the steps in any of the above-mentioned electric energy sub-item metering error detection model training method embodiments, and the specific model training process is not repeated here.

[0187] In this embodiment, by acquiring the electric energy sub-item metering signal sequence and inputting it into the trained metering error detection model obtained by the above-mentioned electric energy sub-item metering error detection model training method, the metering error in the electric energy sub-item metering signal is detected. On the one hand, the accuracy and efficiency of the electric energy sub-item metering error detection are improved by the model detection method. On the other hand, based on the metering error detection result, it is helpful to analyze the deviation of each power index in the actual measurement, and then it is helpful to take corresponding measures according to the metering error, so as to realize the accurate measurement and error analysis of each power index of the power system, and can provide strong support for the calibration and optimization of the power metering system.

[0188] In an exemplary embodiment, Figure 6 As shown, a device 600 for training an electric energy sub-item metering error detection model is provided, comprising: a data acquisition module 610, a signal extraction module 620, a training sample determination module 630, a metering error prediction module 640, a difference value determination module 650 and a parameter adjustment module 660, wherein:

[0189] The data acquisition module 610 is used to acquire the electric energy item metering signal sequence;

[0190] The signal extraction module 620 is used to perform wavelet packet transform on the electric energy sub-item metering signal sequence, and extract multiple target electric energy sub-item metering signals in the target characteristic frequency band from the electric energy sub-item metering signal sequence;

[0191] A training sample determination module 630 is used to determine a model training sample based on a plurality of target electric energy sub-item metering signals;

[0192] The metering error prediction module 640 is used to take the model training sample as input, call the pre-built initial metering error detection model, and obtain the metering error prediction value of each target electric energy sub-item metering signal in the model training sample;

[0193] The difference value determination module 650 is used to determine the difference value between the metering error prediction value and the metering error actual value of the target electric energy sub-item metering signal for each target electric energy sub-item metering signal;

[0194] The parameter adjustment module 660 is used to adjust the model parameters of the initial metrology error detection model based on the difference value until a preset training end condition is reached to obtain a trained metrology error detection model.

[0195] In an exemplary embodiment, the signal extraction module 620 is also used to perform wavelet packet transform on the electric energy sub-item metering signal in each electric energy sub-item metering signal sequence, decompose the electric energy sub-item metering signal into sub-band signals of multiple frequency bands; determine the energy value corresponding to each sub-band signal after decomposition; determine the frequency band of the sub-band signal with the largest energy value as the target characteristic frequency band; and extract the target electric energy sub-item metering signal under the target characteristic frequency band from the electric energy sub-item metering signal.

[0196] In an exemplary embodiment, the signal extraction module 620 is further configured to decompose the electric energy sub-item metering signal into sub-band signals of multiple frequency bands based on a preset wavelet packet basis function and according to a preset number of wavelet packet transform decomposition layers.

[0197] In an exemplary embodiment, the parameter adjustment module 660 is further used to adjust the weight vector of the output layer, the center vector of the hidden layer and the width vector of the Gaussian basis function of the initial measurement error detection model based on the difference value through a gradient descent algorithm.

[0198] In an exemplary embodiment, the training sample determination module 630 is also used to obtain the measurement value of the target electric energy sub-item metering signal for each target electric energy sub-item metering signal; determine the difference between the measurement value and the expected value of the electric energy sub-item metering signal to obtain a metering error value; based on the metering error value, add an error label to the target electric energy sub-item metering signal; and integrate multiple target electric energy sub-item metering signals with error labels added to obtain a model training sample.

[0199] In an exemplary embodiment, the electric energy sub-item metering error detection model training device 600 further includes an effective number of bits adjustment module 670 for adjusting the effective number of bits of the analog-to-digital converter of the electric energy sub-item metering device.

[0200] The data acquisition module 610 is further used to acquire the electric energy sub-item metering signal sequence collected by the adjusted electric energy sub-item metering device.

[0201] Each module in the above-mentioned electric energy sub-item metering error detection model training device 600 can be implemented in whole or in part by software, hardware, or a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0202] In an exemplary embodiment, Figure 7 As shown, a device 700 for detecting an error in measuring electric energy sub-items is provided, comprising a signal acquisition module 710 and a measuring error detection module 720, wherein:

[0203] The signal acquisition module 710 is used to acquire a sequence of electric energy item metering signals.

[0204] The metering error detection module 720 is used to take the electric energy sub-item metering signal sequence as input, call the trained metering error detection model, detect the metering error in the electric energy sub-item metering signal sequence, and obtain the metering error value in the electric energy sub-item metering signal sequence. The trained metering error detection model is trained based on the steps in any one of the above-mentioned electric energy sub-item metering error detection model training method embodiments.

[0205] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 8As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for training an electric energy sub-item metering error detection model is implemented.

[0206] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0207] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps in any one of the above-mentioned embodiments of the training method for electric energy sub-item metering error detection model and the steps in an embodiment of the electric energy sub-item metering error detection method are implemented.

[0208] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in any one of the above-mentioned electric energy sub-item metering error detection model training methods and the steps in an electric energy sub-item metering error detection method embodiment are implemented.

[0209] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps in any one of the above-mentioned electric energy sub-item metering error detection model training methods and the steps in an electric energy sub-item metering error detection method embodiment.

[0210] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0211] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0212] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0213] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for training an electric energy sub-item metering error detection model, characterized in that: The method comprises: Obtain electric energy sub-item metering signal sequence; Performing wavelet packet transform on the electric energy sub-item metering signal sequence, and extracting a plurality of target electric energy sub-item metering signals in a target characteristic frequency band from the electric energy sub-item metering signal sequence; Determine the model training samples based on multiple target electric energy sub-item metering signals; Taking the model training sample as input, calling the pre-built initial metering error detection model, and obtaining the metering error prediction value of each target electric energy sub-item metering signal in the model training sample; For each of the target electric energy sub-item metering signals, determining a difference between a metering error prediction value and an actual metering error value of the target electric energy sub-item metering signal; Based on the difference value, the model parameters of the initial measurement error detection model are adjusted until a preset training end condition is reached to obtain a trained measurement error detection model.

2. The method according to claim 1, characterized in that: The step of performing wavelet packet transform on the electric energy sub-item metering signal sequence to extract a plurality of target electric energy sub-item metering signals in a target characteristic frequency band from the electric energy sub-item metering signal sequence includes: For each electric energy item metering signal in the electric energy item metering signal sequence, the following processing is performed: Performing wavelet packet transformation on the electric energy sub-item metering signal to decompose the electric energy sub-item metering signal into sub-band signals of multiple frequency bands; Determine the energy value corresponding to each decomposed sub-band signal; Determining the frequency band of the subband signal with the maximum energy value as the target characteristic frequency band; A target electric energy sub-item metering signal in the target characteristic frequency band is extracted from the electric energy sub-item metering signal.

3. The method according to claim 2, characterized in that The step of performing wavelet packet transformation on the electric energy sub-item metering signal to decompose the electric energy sub-item metering signal into sub-band signals of multiple frequency bands includes: Based on a preset wavelet packet basis function, the electric energy sub-item metering signal is decomposed into sub-band signals of multiple frequency bands according to a preset wavelet packet transform decomposition layer number.

4. The method according to claim 3, characterized in that: The adjusting the model parameters of the initial measurement error detection model based on the difference value includes: Based on the difference value, the weight vector of the output layer, the center vector of the hidden layer and the width vector of the Gaussian basis function of the initial measurement error detection model are adjusted through a gradient descent algorithm.

5. The method according to claim 4, characterized in that The step of determining a model training sample based on a plurality of target electric energy sub-item metering signals comprises: For each target electric energy sub-item metering signal, perform the following steps: Obtaining the measured value of the target electric energy sub-item metering signal; Determine the difference between the measured value and the expected value of the electric energy sub-item metering signal to obtain a metering error value; Based on the metering error value, adding an error label to the target electric energy sub-item metering signal; Multiple target electricity sub-item metering signals with error labels added are integrated to obtain model training samples.

6. The method according to any one of claims 1 to 5, characterized in that: The step of obtaining a signal sequence for item-by-item measurement of electric energy comprises: Adjust the effective number of bits of the analog-to-digital converter of the electric energy sub-metering device; The electric energy sub-item metering signal sequence collected by the adjusted electric energy sub-item metering device is obtained.

7. A method for detecting errors in electric energy sub-item measurement, characterized in that: The method comprises: Obtain electric energy sub-item metering signal sequence; Taking the electric energy sub-item metering signal sequence as input, calling the trained metering error detection model, detecting the metering error in the electric energy sub-item metering signal sequence, and obtaining the metering error value in the electric energy sub-item metering signal sequence, the trained metering error detection model is trained based on the electric energy sub-item metering error detection model training method according to any one of claims 1 to 6.

8. A device for training an electric energy sub-item metering error detection model, characterized in that: The device comprises: A data acquisition module is used to obtain a signal sequence of electric energy sub-item metering; A signal extraction module, used for performing wavelet packet transform on the electric energy sub-item metering signal sequence, and extracting a plurality of target electric energy sub-item metering signals in a target characteristic frequency band from the electric energy sub-item metering signal sequence; A training sample determination module, used to determine a model training sample based on multiple target electric energy sub-item metering signals; A metering error prediction module, used to take the model training sample as input, call a pre-built initial metering error detection model, and obtain a metering error prediction value of each target electric energy sub-item metering signal in the model training sample; A difference value determination module, used to determine, for each of the target electric energy sub-item metering signals, a difference value between a metering error prediction value and an actual metering error value of the target electric energy sub-item metering signal; The parameter adjustment module is used to adjust the model parameters of the initial measurement error detection model based on the difference value until a preset training end condition is reached to obtain a trained measurement error detection model.

9. An electric energy sub-item metering error detection device, characterized in that: The device comprises: A signal acquisition module is used to obtain a signal sequence of electric energy sub-item metering; A metering error detection module is used to take the electric energy sub-item metering signal sequence as input, call a trained metering error detection model, detect the metering error in the electric energy sub-item metering signal sequence, and obtain the metering error value in the electric energy sub-item metering signal sequence, wherein the trained metering error detection model is trained based on the electric energy sub-item metering error detection model training method according to any one of claims 1 to 6.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

Images