Intelligent electric meter performance test method and system

Abstract

The invention provides an intelligent electric meter performance test method and system, and relates to the technical field of data processing, and the method comprises the steps: constructing an orthogonal pseudo-random dynamic test voltage signal and a current signal; inputting the orthogonal pseudo-random dynamic test voltage signal and the current signal into the intelligent electric meter to be detected for testing, and counting a dynamic electric energy error signal; judging whether the dynamic electric energy error signal is lower than a preset electric energy error; if yes, determining that the to-be-detected intelligent electric meter is qualified; otherwise, determining that the performance of the to-be-detected intelligent electric meter is doubted, and entering the next step for fine testing; extracting signal features in the dynamic electric energy error signals through a deep belief network; inputting the signal features extracted by the deep belief network into a multi-layer perceptron; performance testing is carried out on the intelligent electric meter to be detected through the multi-layer perceptron. According to the invention, the test result can truly reflect the measurement performance of the electric meter in a real environment, and the delivery quality of the intelligent electric meter is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a smart meter performance testing method and system. BACKGROUND

[0002] Before the smart meter is shipped, it needs to undergo strict calibration testing to ensure that its electric energy measurement accuracy meets the requirements of national or industry standards, and to ensure that its measurement error under various typical working conditions is within a controllable range. In order to simulate the voltage, current fluctuation and load change in the actual use environment, test signals under various working conditions are applied to evaluate the measurement accuracy, response speed and anti-interference ability of the smart meter. After calibration, statistical analysis of the test data is required, and a qualified report is issued to ensure that each shipped meter has good long-term stability and use reliability, thereby ensuring fair user measurement, safe power use and fine management of power grid operation.

[0003] The traditional testing scheme often uses standard power supply to output steady-state sinusoidal voltage and current signals, and then superimposes certain step or slope disturbances to simulate load changes. These signals are input to the meter under test, and the dynamic energy error value is obtained by comparing with a standard energy meter. Whether the energy error is within the national standard allowable range is determined as the basis for shipment or measurement performance.

[0004] However, the real power usage scenario is not a steady-state sinusoidal voltage and current signal, and the traditional scheme cannot accurately simulate the real-life load conditions of asymmetric, random jump, periodic jitter, etc. It is not sensitive to dynamic error, easy to miss potential problems, and difficult to truly reflect the measurement performance of the meter in the real environment. Moreover, when the performance of the smart meter is in doubt after testing, it is difficult to analyze the specific fault problem, and cannot effectively guide the subsequent production of smart meters. SUMMARY

[0005] In order to solve the technical problems that the real power usage scenario is not a steady-state sinusoidal voltage and current signal, the traditional scheme cannot accurately simulate the real-life load conditions of asymmetric, random jump, periodic jitter, etc. It is not sensitive to dynamic error, easy to miss potential problems, and difficult to truly reflect the measurement performance of the meter in the real environment. Moreover, when the performance of the smart meter is in doubt after testing, it is difficult to analyze the specific fault problem, and cannot effectively guide the subsequent production of smart meters, the present application provides a smart meter performance testing method and system.

[0006] The technical scheme provided by the embodiments of the present application is as follows: First aspect: The smart meter performance testing method provided by the embodiments of the present application comprises: S1: Constructing a quadrature pseudo-random dynamic test voltage signal and a current signal; S2: inputting the orthogonal pseudo-random dynamic test voltage signal and the current signal into the smart meter to be detected for testing, and counting a dynamic electric energy error signal; S3: judging whether the dynamic electric energy error signal is lower than a preset electric energy error; if yes, determining that the smart meter to be detected is qualified; otherwise, determining that the performance of the smart meter to be detected is questionable, and entering the next step for fine testing; S4: extracting a signal feature in the dynamic electric energy error signal through a deep belief network; S5: inputting the signal feature extracted by the deep belief network into a multi-layer perception machine; S6: performing performance testing on the smart meter to be detected through the multi-layer perception machine.

[0007] The second aspect: The embodiment of the present application provides a kind of smart meter performance test system, comprising: Processor; Memory, the computer readable instruction is stored on the memory, the computer readable instruction is implemented when the processor is executed, the smart meter performance test method as described in the first aspect.

[0008] The third aspect: The embodiment of the present application provides a kind of computer readable storage medium, which stores computer program, the program is executed when processor realizes the smart meter performance test method as described in the first aspect.

[0009] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects: (1) in the embodiment of the present application, orthogonal pseudo-random dynamic test voltage signal and current signal are constructed, the real situation such as asymmetric, random jump, periodic jitter of load in real life is accurately simulated, and dynamic error of smart meter is tested using orthogonal pseudo-random dynamic test voltage signal and current signal, the sensitivity of smart meter to dynamic error is improved, so that the test result can truly reflect the measurement performance of meter in real environment, and the factory quality of smart meter is improved.

[0010] (2) in the embodiment of the present application, the performance of smart meter is tested through deep belief network and multi-layer perception machine, the specific fault type of smart meter can be accurately diagnosed, effective guidance is given to subsequent smart meter production, and the production efficiency of smart meter is improved. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0012] Figure 1 A flowchart of a smart meter performance test method provided by an embodiment of the present application.

[0013] Figure 2 A schematic diagram of forming a dynamic power error signal provided by an embodiment of the present application.

[0014] Figure 3 A structural schematic diagram of a smart meter performance test model provided by an embodiment of the present application.

[0015] Figure 4 A structural schematic diagram of a smart meter performance test system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0016] The technical solutions in the present application will be described below with reference to the drawings.

[0017] In order to make the technical problems, technical solutions and advantages of the present application clearer, the following will be described in detail with reference to the drawings and specific embodiments.

[0018] Reference is made to the accompanying drawings Figure 1 , which shows a flowchart of a smart meter performance test method provided by an embodiment of the present application.

[0019] The present embodiment provides a smart meter performance test method, which can be implemented by a smart meter performance test device, which can be a terminal or a server. The processing flow of the smart meter performance test method can include the following steps: Reference is made to the accompanying drawings Figure 2 , which shows a schematic diagram of forming a dynamic power error signal provided by an embodiment of the present application.

[0020] S1: Constructing a pseudo-random orthogonal dynamic test power signal.

[0021] In a possible implementation, S1 specifically includes: S101: Constructing a distorted steady-state standard test voltage signal and a current signal: Wherein, u sdenotes the distorted steady-state standard test voltage signal, t denotes the time, k = a, b, c respectively denotes phase A, phase B, phase C, denotes the distorted steady-state standard test voltage signal of the kth phase at time t, k denotes the fundamental amplitude of the kth phase voltage signal, n1 denotes the harmonic order of the voltage signal, L k denotes the total harmonic order of the kth phase voltage signal, denotes the relative amplitude of the n1th order voltage harmonic component with respect to the fundamental amplitude, A k1 = 1, sin denotes the sine function, ω denotes the angular frequency, denotes the phase angle of the n1th order voltage harmonic component, i s denotes the distorted steady-state standard test current signal, denotes the distorted steady-state standard test current signal of the kth phase at time t, k denotes the fundamental amplitude of the kth phase current signal, n2 denotes the harmonic order of the current signal, l k denotes the total harmonic order of the kth phase current signal, denotes the relative amplitude of the n2th order current harmonic component with respect to the fundamental amplitude, B k1 = 1, denotes the relative amplitude of the n2th order current harmonic component with respect to the fundamental amplitude.

[0022] It should be noted that the distorted steady-state standard test voltage signal and the current signal are basic input signals for simulating complex non-ideal power signals in actual power consumption environment, which are usually superimposed by multiple harmonics with different amplitudes and phases to reflect the common waveform distortion phenomenon in real power grid.

[0023] S102: truncate the distorted steady-state standard test voltage signal and the current signal by multiple rectangular window functions to obtain the truncated voltage signal and the current signal of each fundamental period, and combine to form the distorted steady-state test truncated voltage signal and the current signal: wherein u c denotes the distorted steady-state test truncated voltage signal, denotes the distorted steady-state test truncated voltage signal of the kth phase at time t, denotes the distorted steady-state test truncated voltage signal of the kth phase at time t under the nth fundamental period, denotes the rectangular window function, i cIndicates the distortion steady-state test cutoff current signal, It represents the distorted steady-state test cut-off current signal of phase k at time t. It represents the distorted steady-state test cut-off current signal of the k-th phase at time t in the n-th basic cycle, where n represents the basic cycle and t0 represents the zero-crossing time of the signal.

[0024] It should be noted that the distortion steady-state test cuts off the voltage and current signals on this basis, and cuts off the entire cycle through the rectangular window function, retaining the waveform fragments of each base cycle and combining them in chronological order, so as to more accurately restore the response behavior of the meter in each cycle, laying the foundation for subsequent dynamic modulation and error analysis.

[0025] S103: Construct an orthogonal pseudo-random measurement matrix through the pseudo-random sequence diagonal matrix, the orthogonal Hadamard matrix and the downsampling matrix.

[0026] The pseudorandom sequence diagonal matrix is ​​a diagonal matrix composed of pseudorandomly generated binary sequences as the main diagonal elements, with all off-diagonal elements set to 0. This matrix is ​​used to pseudorandomly modulate the input signal, breaking its periodicity and predictability, thereby enhancing the system's ability to simulate nonstationary disturbances. It is suitable for compressed sensing and dynamic testing scenarios.

[0027] An orthogonal Hadamard matrix is ​​a square matrix composed of 1s and -1s with orthogonal rows and columns. It exhibits excellent orthogonality and fast transformation properties. In dynamic testing, the introduction of an orthogonal Hadamard matrix can achieve orthogonal transformation of the signal, preserving its energy structure while avoiding interference between dimensions, thus facilitating the construction of an uncorrelated measurement basis.

[0028] The downsampling matrix is ​​a sparse diagonal matrix with diagonal elements of either 0 or 1. Only M diagonal elements are 1, and the rest are 0. It is used to select some sampling points in the signal. This matrix enables compressed sampling of high-dimensional signals, reducing the data dimension while retaining key information, satisfying the principles of compressed sensing.

[0029] Next we will introduce how to construct an orthogonal pseudo-random measurement matrix.

[0030] Optionally, the S103 specifically includes: S1031: Select the first random sequence and the second random sequence as the preferred sequence pair to generate a Gold sequence family: Wherein, R1 represents the first random sequence, R2 represents the second random sequence, Indicates bitwise exclusive OR, D l Indicates a right-shift operation of 1 bit. , denotes the length of the Gold sequence family.

[0031] wherein the Gold sequence family is a set of pseudo-random binary sequences generated by exclusive-OR and shift combination of two m-sequences (maximum length sequences) which are a preferred pair of each other, has good autocorrelation and cross-correlation characteristics, and the Gold sequence family has been widely applied in the fields of communication, radar, spread spectrum and testing, which will not be described herein.

[0032] S1032: Select a Gold sequence with the best balance from the Gold sequence family.

[0033] The balance refers to the balance of the distribution of the number 1 and the number 0 in the sequence, and the balance can be evaluated by the ratio of the number of the number 1 and the number 0 in the sequence.

[0034] S1033: Construct a pseudo-random sequence diagonal matrix according to the Gold sequence with the best balance: wherein G represents the pseudo-random sequence diagonal matrix, g ij represents the value of the element in the i-th row and the j-th column of the pseudo-random sequence diagonal matrix, g i represents the middle vector in the i-th row of the pseudo-random sequence diagonal matrix, Z represents the Gold sequence with the best balance, and δ ij represents the assignment parameter of the i-th row and the j-th column, N represents the size of the pseudo-random sequence diagonal matrix, and at the same time represents the number of columns of the finally generated orthogonal pseudo-random measurement matrix.

[0035] In the embodiment of the application, the sequence with the best balance is selected from the Gold sequence family to construct the pseudo-random sequence diagonal matrix, which can ensure that the modulated signal has good statistical balance in time, that is, the distribution of "1" and "0" is as uniform as possible, thereby avoiding problems such as uneven energy distribution and excessive bias in signal modulation. The guarantee of such balance helps to improve the representativeness and stability of the subsequent test signal, so that the constructed orthogonal pseudo-random measurement matrix has higher robustness and accuracy in compressive sensing or fault testing, thereby improving the overall performance and reliability of dynamic error detection and smart meter performance testing.

[0036] S1034: Generate an orthogonal Hadamard matrix through Walsh-Hadamard transformation.

[0037] Wherein, Walsh-Hadamard transform (Walsh-Hadamard Transform, WHT for short) is a kind of orthogonal transform method, which uses Hadamard matrix composed of only +1 and-1 to transform signal, without multiplication operation, and high calculation efficiency.Walsh-Hadamard transform can be used to generate orthogonal Hadamard matrix by using recursive construction method.This part is very mature existing technology, and the present application will not be described again.

[0038] S1035: constructing a down-sampling matrix: Wherein, D down represents a down-sampling matrix, represents the element value of the i-th row and the j-th column in the down-sampling matrix, the sum of the elements of 1 on the diagonal line of the down-sampling matrix is M, and M represents the number of columns of the finally generated orthogonal pseudo-random measurement matrix, represents much smaller than.

[0039] It should be noted that the down-sampling matrix can be constructed by randomly setting 1 or 0 on the diagonal line.

[0040] S1036: the pseudo-random sequence diagonal matrix, the orthogonal Hadamard matrix and the down-sampling matrix are combined to form the orthogonal pseudo-random measurement matrix: Wherein, represents an orthogonal pseudo-random measurement matrix, H U represents an orthogonal Hadamard matrix.

[0041] It should be noted that the principle of constructing the orthogonal pseudo-random measurement matrix by the above method is that high-efficiency sampling under signal sparse representation is realized by combining three kinds of matrix modules.The pseudo-random sequence diagonal matrix is used for disturbance and weighting of the original signal.The orthogonal Hadamard matrix provides a sparse orthogonal transform base, which projects the signal to the frequency domain or orthogonal space, enhances the separability and feature saliency.The signal is projected to the frequency domain or orthogonal space, and the separability and feature saliency are enhanced.The down-sampling matrix is a sparse diagonal matrix, which is used to select M observation values from the results of Hadamard transform, to achieve the dimension reduction effect in compressive sensing, reduce the data dimension, and improve the calculation efficiency.

[0042] In the embodiment of the present application, the three mechanisms of pseudo-random disturbance, orthogonal transformation and sparse sampling are combined, which not only retains the main features of the signal, but also greatly reduces the data redundancy, improves the complexity and diversity of the test signal; compared with the traditional method, it can more truly simulate the current / voltage fluctuation in the non-ideal power consumption environment, effectively improve the accuracy and coverage of the dynamic error detection of the smart meter, and provide high-quality input features for subsequent performance testing.

[0043] S1037: determining the optimal orthogonal pseudo-random measurement matrix from the generated plurality of orthogonal pseudo-random measurement matrices according to the incoherence criterion: wherein μ represents the coherence measure, represents the orthogonal pseudo-random measurement matrix, represents the sparse matrix, and max represents the maximum value, represents the row vector composed of the i-th row elements of the measurement matrix, T represents the matrix transposition operation, represents the column vector composed of the j-th column elements of the sparse matrix, represents the inner product operation.

[0044] It should be noted that in the field of signal processing and compressed sensing, the incoherence criterion is used to ensure that there is no strong correlation between the measurement matrix and the sparse matrix, thereby improving the signal recovery capability. The coherence measure refers to the maximum value of the inner product between each row of the measurement matrix and each column of the sparse matrix. The smaller this maximum value is, the weaker the correlation between the measurement matrix and the sparse matrix, and the less interference and information loss there will be during signal recovery.

[0045] Further, the orthogonal pseudo-random measurement matrix with the smallest coherence measure can be selected as the optimal orthogonal pseudo-random measurement matrix.

[0046] In the embodiment of the present application, by using the incoherence criterion, the optimal orthogonal pseudo-random measurement matrix is selected, and the target is to minimize the inner product between the measurement matrix and the sparse matrix. This ensures that the compressed sampling of the signal does not lose key information, and helps to improve the accuracy of subsequent signal recovery.

[0047] S104: modulating the distorted steady-state test cut-off voltage signal and current signal by the orthogonal pseudo-random measurement matrix to generate orthogonal pseudo-random dynamic test voltage signal, current signal and power signal.

[0048] Optionally, the S104 specifically includes: S1041: performing a vectorization operation on the orthogonal pseudo-random measurement matrix to form an orthogonal pseudo-random dynamic sequence: wherein, Vec represents vectorization operation, represents the kth phase orthogonal pseudo-random dynamic sequence, represents the Lth sequence value in the kth phase orthogonal pseudo-random dynamic sequence, L = MN.

[0049] It should be noted that the two-dimensional matrix is converted into a one-dimensional modulation sequence, which is convenient for subsequent one-to-one element-level modulation with voltage and current signals. At the same time, the orthogonal pseudo-randomness of the original matrix is retained, so that the modulation signal has excellent structure and resolution.

[0050] S1042: modulate the distortion steady-state test cut-off voltage signal and current signal by the orthogonal pseudo-random dynamic sequence to obtain an orthogonal pseudo-random dynamic test voltage signal and current signal: wherein, u d represents the orthogonal pseudo-random dynamic test voltage signal, represents the kth phase t time orthogonal pseudo-random dynamic test voltage signal, i d represents the orthogonal pseudo-random dynamic test current signal, represents the kth phase t time orthogonal pseudo-random dynamic test current signal, represents Hadamard multiplication.

[0051] It should be noted that the static cut-off signal is converted into a dynamic signal to realize "on-off" control of power transmission. Combined with subsequent pseudo-random modulation, it can simulate real complex working conditions such as mutation, jump, jitter, etc. in the power grid.

[0052] S1043: Hadamard multiplication is performed on the orthogonal pseudo-random dynamic test voltage signal and current signal to obtain an orthogonal pseudo-random dynamic test power signal: wherein, p represents the orthogonal pseudo-random dynamic test power signal, represents the kth phase t time orthogonal pseudo-random dynamic test power signal.

[0053] It should be noted that the orthogonal pseudo-random dynamic test power signal can reflect the transmission intensity of electric energy at each time, forming a dynamic power input. Consistent with the internal electric energy accumulation logic of the electric meter, it is helpful to accurately simulate real power fluctuations. Finally, a "power driving signal" with strong realism and rich disturbance is formed, which is convenient for high-precision dynamic error detection.

[0054] S2: inputting the orthogonal pseudo-random dynamic test power signal into the smart meter to be detected for testing, and counting a dynamic energy error signal.

[0055] In a possible implementation, the S2 specifically includes: S201: inputting the orthogonal pseudo-random dynamic test power signal into the smart meter to be detected for testing.

[0056] S202: generating a gate signal to control a test interval, and each test interval contains a plurality of base periods of the orthogonal pseudo-random dynamic test power signal.

[0057] S203: determining a reference dynamic energy sequence generated by inputting the orthogonal pseudo-random dynamic test power signal into the smart meter to be detected. wherein E n represents the reference dynamic energy of the nth base period, represents the nth sequence value in the k-phase orthogonal pseudo-random dynamic sequence, and is 0 or 1, which determines whether energy is transmitted in the nth base period, when , the dynamic energy is E0, when , the dynamic energy is 0, N t represents the total number of base periods contained in each test interval.

[0058] It should be noted that, according to the orthogonal pseudo-random sequence, a 0 / 1 switch type control is introduced, an expected energy transmission mode is theoretically constructed, a comparison benchmark is established, and a “gold standard” is provided for subsequent error evaluation.

[0059] It should be noted that the dynamic energy specifically includes: wherein t0 represents the starting time of a base period, T represents the base period, and p(t) represents the power at t.

[0060] S204: calculating a reference dynamic cumulative energy according to the reference dynamic energy sequence. wherein E ref represents the reference dynamic cumulative energy.

[0061] S205: measuring an actual dynamic cumulative energy generated by inputting the orthogonal pseudo-random dynamic test power signal into the smart meter to be detected.

[0062] S206: counting a dynamic energy error between the reference dynamic cumulative energy and the actual dynamic cumulative energy. wherein e represents a dynamic energy error, E act measured actual dynamic cumulative energy.

[0063] It should be noted that the dynamic energy error provides an accurate quantitative error index, which can be used to determine whether the electric meter is qualified and as a basis for further diagnosis.

[0064] S207: record the dynamic energy error in each test interval in time sequence as the dynamic energy error signal.

[0065] In the embodiment of the present application, the orthogonal pseudo-random dynamic test voltage signal and the current signal are constructed to accurately simulate the real-life conditions of asymmetric, random jump, periodic jitter and the like of the load, and the dynamic error of the smart electric meter is tested using the orthogonal pseudo-random dynamic test voltage signal and the current signal, which improves the sensitivity of the smart electric meter to the dynamic error, so that the test result can truly reflect the measurement performance of the electric meter in the real environment, and the quality of the smart electric meter at the factory is improved.

[0066] S3: determine whether the dynamic energy error signal is lower than the preset energy error. If yes, it is determined that the smart electric meter to be detected is qualified. Otherwise, it is determined that the performance of the smart electric meter to be detected is questionable, and the next step of fine testing is entered.

[0067] wherein the size of the preset energy error threshold can be set by a person skilled in the art according to the actual situation, and the present application is not limited.

[0068] In the embodiment of the present application, by setting the preset energy error threshold, rapid screening and preliminary determination of the smart electric meter can be realized. When the dynamic energy error signal is lower than the threshold, it indicates that the electric meter still has good measurement accuracy under the simulated actual complex load conditions, and can be directly determined as qualified. If the threshold is exceeded, it automatically enters a deeper feature analysis and performance test stage, avoiding misjudgment and omission. The hierarchical detection mechanism not only improves the test efficiency and reduces resource waste, but also enables targeted fine processing of abnormal electric meters, thereby enhancing the accuracy and intelligent level of the entire quality inspection process.

[0069] Reference is made to the accompanying drawings Figure 3 , which shows a structural schematic diagram of a smart electric meter performance test model provided by an embodiment of the present application.

[0070] S4: extract signal features in the dynamic energy error signal through a deep belief network.

[0071] The deep belief network is composed of multiple layers of restricted Boltzmann machines, and each layer of restricted Boltzmann machine is composed of a visible layer and a hidden layer. The deep belief network is a very mature prior art, and will not be described herein.

[0072] S5: inputting the signal feature extracted by the deep belief network into a multi-layer perception.

[0073] In a possible implementation, the S5 specifically includes: S501: evaluating feature dispersion of the signal feature extracted by the deep belief network: wherein J represents the feature dispersion, S B represents an inter-class feature dispersion matrix, S W represents an intra-class feature dispersion matrix, and tr represents a trace of a matrix.

[0074] The greater the inter-class dispersion matrix and the smaller the intra-class dispersion matrix, the stronger the distinguishing ability of the feature to different classes, and the higher the value of the feature dispersion. Using the index as an evaluation criterion helps to quantify the feature extraction effect and serve as a basis for judging whether to trigger the feature enhancement mechanism, thereby improving the accuracy and reliability of performance testing and avoiding false classification caused by feature aliasing.

[0075] S502: determining whether the feature dispersion is greater than a preset dispersion. If yes, directly entering S6. Otherwise, triggering the feature enhancement mechanism.

[0076] wherein the size of the preset dispersion can be set by a person skilled in the art according to actual conditions, and the present application is not limited.

[0077] In the embodiment of the present application, by judging whether the feature dispersion is greater than the preset threshold, it can be dynamically evaluated whether the current extracted feature has good class distinguishing ability. If the feature dispersion is high, it means that the existing feature is sufficient for accurate classification, and the performance testing stage can be directly entered to improve the efficiency. Otherwise, the feature enhancement mechanism is triggered to introduce more dimensions or more discriminative features to improve the recognition ability of the model to complex or boundary fuzzy samples. This strategy takes into account the diagnostic accuracy and computational efficiency, and realizes an adaptive and refined intelligent meter fault recognition process.

[0078] Optionally, the feature enhancement mechanism specifically includes: introducing wavelet transform to extract time-frequency domain features of the dynamic power error signal, and taking the time-frequency domain features as inputs of the deep belief network.

[0079] In the embodiments of the present application, the wavelet transform is introduced to extract the time-frequency domain features of the dynamic power error signal, which can effectively capture the transient changes and frequency components in the signal at different time scales, thereby revealing more abundant and discriminative feature information. Taking these time-frequency domain features as the supplementary input of the deep belief network can help improve the model's understanding ability of complex and non-stationary signals, enhance the accuracy and robustness of performance testing, especially in the case of dealing with dynamic error fluctuations or local features.

[0080] Optionally, the feature enhancement mechanism specifically includes: introducing a dynamic response voltage signal, a dynamic response voltage signal and / or a dynamic response power signal as input of the deep belief network together with the dynamic power error signal.

[0081] The dynamic response voltage signal refers to the signal generated by the smart meter under test when the orthogonal pseudo-random dynamic test voltage signal is input into the smart meter under test for testing, and the dynamic response current signal refers to the signal generated by the smart meter under test when the orthogonal pseudo-random dynamic test current signal is input into the smart meter under test for testing.

[0082] In the embodiments of the present application, the dynamic response voltage signal, current signal and / or power signal are introduced as supplementary input, which can comprehensively reflect the real response behavior of the smart meter under the excitation of the orthogonal pseudo-random dynamic test signal. These response signals contain the dynamic characteristics of the meter internal measurement and processing mechanism, which helps to mine potential abnormal patterns or minor fault features. Inputting them into the deep belief network together with the dynamic power error signal not only enriches the feature dimension, but also improves the model's recognition ability of complex fault types, enhances the accuracy and robustness of performance testing.

[0083] S6: performing performance testing on the smart meter under test by the multi-layer perception.

[0084] In one possible implementation, the S6 specifically includes sub-steps S601 to S603: S601: inputting the multiple signal features extracted by the deep belief network in the input layer of the multi-layer perception.

[0085] S602: extracting the hidden state in the signal features in the hidden layer of the multi-layer perception. S603: performing performance testing on the smart meter under test according to the hidden state in the output layer of the multi-layer perception.

[0086] The multi-layer perception is a very mature existing technology, and the specific use process and principles of the multi-layer perception will not be described herein.

[0087] In the embodiment of the present application, the performance of the smart meter is tested by the deep belief network and the multi-layer perception machine, the specific fault type of the smart meter can be accurately diagnosed, the subsequent smart meter production is effectively guided, and the production efficiency of the smart meter is improved.

[0088] With reference to the accompanying drawings Figure 4 The accompanying drawings show a structure schematic diagram of a smart meter performance testing system provided by the present application.

[0089] The present application further provides a smart meter performance testing system 20 applied to the smart meter performance testing method, and comprising: A processor 201.

[0090] A memory 202, the memory 202 stores computer readable instructions, and the computer readable instructions are executed by the processor 201 to realize the smart meter performance testing method of the method embodiment.

[0091] The smart meter performance testing system 20 provided by the present application can execute the smart meter performance testing method and realize the same or similar technical effects, and the present application will not be repeated here.

[0092] It should be understood that the processor in the embodiment of the present application can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0093] It should also be understood that the memory in the embodiments of the present application can be volatile or nonvolatile memory, or can include both volatile and nonvolatile memory. Nonvolatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0094] The above-described embodiments can be implemented, in whole or in part, by software, hardware (such as a circuit), firmware or any combination thereof. When implemented by software, the above-described embodiments can be implemented, in whole or in part, in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through a wired (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing a set of one or more available media. The available media can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid state disk.

[0095] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the technical solutions that essentially contribute to the prior art or the part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0096] The embodiment of the present application provides a computer-readable storage medium, which stores a computer program. The program is executed by a processor to implement the intelligent electric meter performance test method described in the method embodiment.

[0097] The computer-readable storage medium provided by the present application can implement the steps and effects of the intelligent electric meter performance test method of the above-mentioned method embodiment. To avoid repetition, the present application will not be described again.

Claims

1. A smart meter performance testing method, characterized in that: include: S1: Construct an orthogonal pseudo-random dynamic test power signal; S2: Inputting the orthogonal pseudo-random dynamic test power signal into the smart meter to be tested for testing, and calculating a dynamic power error signal; S3: Determine whether the dynamic power error signal is lower than a preset power error; if so, determine that the smart meter to be tested is qualified; Otherwise, it is determined that the performance of the smart meter to be tested is questionable, and the process proceeds to the next step of performing detailed testing; S4: extracting signal features from the dynamic power error signal through a deep belief network; S5: Inputting the signal features extracted by the deep belief network into a multi-layer perceptron; S6: Performing a performance test on the smart meter to be tested by the multi-layer perceptron; Wherein, the S1 specifically includes: S101: Construct distorted steady-state standard test voltage and current signals; S102: truncating the distorted steady-state standard test voltage signal and current signal using multiple rectangular window functions to obtain truncated voltage signals and current signals of each base cycle, and combining them to form distorted steady-state test truncated voltage signals and current signals; S103: constructing an orthogonal pseudo-random measurement matrix through the pseudo-random sequence diagonal matrix, the orthogonal Hadamard matrix and the downsampling matrix; S104: Modulating the distorted steady-state test truncated voltage signal and current signal by using the orthogonal pseudo-random measurement matrix to generate an orthogonal pseudo-random dynamic test voltage signal, current signal, and power signal.

2. The smart meter performance testing method according to claim 1, characterized in that: The S103 specifically includes: S1031: Selecting a first random sequence and a second random sequence as a preferred sequence pair to generate a Gold sequence family; S1032: Selecting a Gold sequence with the best balance from the Gold sequence family; S1033: Constructing the pseudo-random sequence diagonal matrix according to the Gold sequence with the best balance; S1034: Generate the orthogonal Hadamard matrix through Walsh-Hadamard transformation; S1035: Construct the downsampling matrix; S1036: Combining the pseudo-random sequence diagonal matrix, the orthogonal Hadamard matrix, and the downsampling matrix into the orthogonal pseudo-random measurement matrix; S1037: Determine an optimal orthogonal pseudo-random measurement matrix from the generated multiple orthogonal pseudo-random measurement matrices according to the incoherence criterion.

3. The smart meter performance testing method according to claim 1, characterized in that: The S104 specifically includes: S1041: Performing a vectorization operation on the orthogonal pseudo-random measurement matrix to form an orthogonal pseudo-random dynamic sequence; S1042: Modulating the distorted steady-state test truncated voltage signal and current signal by using the orthogonal pseudo-random dynamic sequence to obtain the orthogonal pseudo-random dynamic test voltage signal and current signal; S1043: Perform Hadamard multiplication on the orthogonal pseudo-random dynamic test voltage signal and the current signal to obtain the orthogonal pseudo-random dynamic test power signal.

4. The smart meter performance testing method according to claim 1, characterized in that: The S2 specifically includes: S201: Inputting the orthogonal pseudo-random dynamic test power signal into the smart meter to be tested for testing; S202: Generate a gating signal to control a test interval, each of the test intervals including a plurality of base periods of the orthogonal pseudo-random dynamic test power signals; S203: Determine a reference dynamic power sequence generated by inputting the orthogonal pseudo-random dynamic test power signal into the smart meter to be tested; S204: Calculating reference dynamic accumulated electric energy according to the reference dynamic electric energy sequence; S205: measuring the actual dynamic accumulated electric energy generated by inputting the orthogonal pseudo-random dynamic test power signal into the smart meter to be tested; S206: Counting the dynamic energy error between the reference dynamic accumulated energy and the actual dynamic accumulated energy; S207: Record the dynamic power error in each test interval in chronological order as the dynamic power error signal.

5. The smart meter performance testing method according to claim 1, characterized in that: The S5 specifically includes: S501: Evaluate the feature discreteness of the signal feature extracted by the deep belief network; S502: Determine whether the feature discreteness is greater than a preset discreteness; if so, directly proceed to S6; otherwise, trigger a feature enhancement mechanism.

6. The smart meter performance testing method according to claim 5, characterized in that: The feature enhancement mechanism specifically includes: Wavelet transform is introduced to extract the time-frequency domain features of the dynamic electric energy error signal, and the time-frequency domain features are used as the input of the deep belief network.

7. The smart meter performance testing method according to claim 5, characterized in that: The feature enhancement mechanism specifically includes: Introducing a dynamic response voltage signal, a dynamic response voltage signal and / or a dynamic response power signal as inputs of the deep belief network together with the dynamic electric energy error signal; Among them, the dynamic response voltage signal refers to the signal generated by the smart meter to be tested when the orthogonal pseudo-random dynamic test voltage signal is input into the smart meter to be tested for testing, and the dynamic response current signal refers to the signal generated by the smart meter to be tested when the orthogonal pseudo-random dynamic test current signal is input into the smart meter to be tested for testing.

8. The smart meter performance testing method according to claim 5, characterized in that: The S6 specifically includes: S601: Inputting a plurality of signal features extracted by the deep belief network into the input layer of the multilayer perceptron; S602: Extracting a hidden state from the signal feature in a hidden layer of the multilayer perceptron; S603: In the output layer of the multi-layer perceptron, a performance test is performed on the smart meter to be detected according to the hidden state.

9. A smart meter performance test system, characterized in that: include: processor; A memory, wherein computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by the processor, the smart meter performance testing method according to any one of claims 1 to 8 is implemented.

Images