A power quality evaluation method and device based on multi-source data fusion

Abstract

This invention relates to the field of power quality analysis technology, specifically providing a power quality evaluation method and apparatus based on multi-source data fusion. The method includes: acquiring power quality parameters at the power supply end, performing wavelet transform on these parameters to obtain the information entropy corresponding to the power quality parameters; using the information entropy as input to a pre-trained SVM classifier to obtain the power quality assessment level of the power supply end output by the pre-trained SVM classifier; acquiring power quality parameters at the user end, using these parameters as input to a pre-built PCNN model to obtain the power quality assessment level of the user end output by the pre-built PCNN model; and determining the power system power command assessment result based on the power supply end power quality assessment level and the user end power quality assessment level. The technical solution provided by this invention can effectively reflect the overall power quality of the power system.

Description

Technical Field

[0001] This invention relates to the field of power quality analysis technology, specifically to a power quality evaluation method and apparatus that integrates multi-source data. Background Technology

[0002] Power quality refers to the electrical performance requirements met during the transmission and utilization of electrical energy in a power supply system, and is an important indicator for evaluating power system performance. Power quality issues mainly include voltage fluctuations, frequency fluctuations, harmonics, voltage sags, and voltage flicker. In modern power systems, the stability and quality of power quality are crucial for ensuring safe equipment operation, improving production efficiency, and reducing energy waste.

[0003] Currently, the main methods for power quality analysis and evaluation include traditional measurement methods and numerical simulation methods based on computer simulation. Measurement methods involve directly collecting field data, such as voltage and current waveforms, by installing sensors, and then analyzing the data. Numerical simulation methods, on the other hand, use computer software to simulate the operating state of the power system, predict power quality problems, and analyze the power quality situation based on the simulation results.

[0004] However, existing power quality analysis and evaluation methods require significant human and material resources for measurement, are easily affected by external environmental factors, and face difficulties in data collection. Furthermore, measurement methods can only provide point data, making it difficult to comprehensively reflect the overall situation of power quality issues. Numerical simulation methods require accurate input of various parameters during modeling and have high computational complexity, making them time-consuming and labor-intensive. Therefore, existing methods need further improvement in terms of real-time performance and accuracy, especially for power quality problems in complex power systems, where current methods often fail to provide accurate analysis. Summary of the Invention

[0005] To overcome the above-mentioned shortcomings, this invention proposes a power quality evaluation method and apparatus based on multi-source data fusion.

[0006] Firstly, a power quality assessment method based on multi-source data fusion is provided, the power quality assessment method based on multi-source data fusion includes:

[0007] The power quality parameters of the power supply end are obtained, and wavelet transform is performed on the power quality parameters of the power supply end to obtain the information entropy corresponding to the power quality parameters of the power supply end.

[0008] The information entropy corresponding to the power quality parameters of the power supply terminal is used as the input of a pre-trained SVM classifier to obtain the power quality assessment level of the power supply terminal output by the pre-trained SVM classifier.

[0009] Obtain the power quality parameters of the user terminal, and use the power quality parameters of the user terminal as the input of a pre-built PCNN model to obtain the power quality assessment level of the user terminal output by the pre-built PCNN model;

[0010] The power system power command assessment result is determined based on the power quality assessment level at the power supply end and the power quality assessment level at the user end.

[0011] Preferably, the power quality parameters include at least one of the following: voltage, current, frequency, and harmonics.

[0012] Preferably, the wavelet transform is a dual-tree complex wavelet transform.

[0013] Preferably, the training process of the pre-trained SVM classifier includes:

[0014] Obtain the power quality parameters of the historical power supply terminal, and perform wavelet transform on the power quality parameters of the historical power supply terminal to obtain the information entropy corresponding to the power quality parameters of the historical power supply terminal;

[0015] The information entropy corresponding to the power quality parameters of the historical power supply terminals is manually annotated with power quality assessment levels, and training data is constructed using the information entropy corresponding to the power quality parameters of the historical power supply terminals with annotated power quality assessment levels.

[0016] The initial SVM classifier is trained using the training data to obtain the pre-trained SVM classifier.

[0017] Furthermore, the kernel function of the SVM classifier is either the Morlet wavelet function or the Mexico wavelet function.

[0018] Preferably, the method includes: adding an interference injection model to the transmission line between the power supply end and the user end to simulate harmonic interference and signal attenuation during the transmission process.

[0019] Furthermore, the interference injection model includes: a harmonic sub-model that inputs harmonics for interference into the transmission line and a length attenuation sub-model that adjusts the impedance of the transmission line.

[0020] Furthermore, the harmonic sub-model is a nonlinear load model or an odd harmonic model.

[0021] Preferably, the power system energy command evaluation results are as follows:

[0022] A3 = p*B1 + q*B2

[0023] In the above formula, A3 is the power system power command assessment result, B1 is the average power quality assessment level at the power supply end, B2 is the power quality assessment level at the user end, p is the first weight, q is the second weight, and p+1 = 1.

[0024] Secondly, a power quality assessment device based on multi-source data fusion is provided, the power quality assessment device based on multi-source data fusion includes:

[0025] The acquisition module is used to acquire the power quality parameters of the power supply end, and perform wavelet transform on the power quality parameters of the power supply end to obtain the information entropy corresponding to the power quality parameters of the power supply end.

[0026] The first analysis module is used to take the information entropy corresponding to the power quality parameters of the power supply terminal as the input of a pre-trained SVM classifier to obtain the power quality assessment level of the power supply terminal output by the pre-trained SVM classifier.

[0027] The second analysis module is used to obtain the power quality parameters of the user end, and use the power quality parameters of the user end as the input of the pre-built PCNN model to obtain the power quality assessment level of the user end output by the pre-built PCNN model.

[0028] The evaluation module is used to determine the power system power command evaluation result based on the power quality evaluation level at the power supply end and the power quality evaluation level at the user end.

[0029] Preferably, the power quality parameters include at least one of the following: voltage, current, frequency, and harmonics.

[0030] Preferably, the wavelet transform is a dual-tree complex wavelet transform.

[0031] Preferably, the training process of the pre-trained SVM classifier includes:

[0032] Obtain the power quality parameters of the historical power supply terminal, and perform wavelet transform on the power quality parameters of the historical power supply terminal to obtain the information entropy corresponding to the power quality parameters of the historical power supply terminal;

[0033] The information entropy corresponding to the power quality parameters of the historical power supply terminals is manually annotated with power quality assessment levels, and training data is constructed using the information entropy corresponding to the power quality parameters of the historical power supply terminals with annotated power quality assessment levels.

[0034] The initial SVM classifier is trained using the training data to obtain the pre-trained SVM classifier.

[0035] Furthermore, the kernel function of the SVM classifier is either the Morlet wavelet function or the Mexico wavelet function.

[0036] Preferably, the device includes: an interference module, used to add an interference injection model to the transmission line between the power supply end and the user end to simulate harmonic interference and signal attenuation during the transmission process.

[0037] Furthermore, the interference injection model includes: a harmonic sub-model that inputs harmonics for interference into the transmission line and a length attenuation sub-model that adjusts the impedance of the transmission line.

[0038] Furthermore, the harmonic sub-model is a nonlinear load model or an odd harmonic model.

[0039] Preferably, the power system energy command evaluation results are as follows:

[0040] A3 = p*B1 + q*B2

[0041] In the above formula, A3 is the power system power command assessment result, B1 is the average power quality assessment level at the power supply end, B2 is the power quality assessment level at the user end, p is the first weight, q is the second weight, and p+1 = 1.

[0042] Thirdly, a computer device is provided, comprising: one or more processors;

[0043] The processor is used to execute one or more programs;

[0044] When the one or more programs are executed by the one or more processors, the power quality assessment method based on multi-source data fusion is implemented.

[0045] Fourthly, a computer-readable storage medium is provided having a computer program stored thereon, wherein when the computer program is executed, it implements the power quality evaluation method based on multi-source data fusion.

[0046] The above-described technical solutions of the present invention have at least one or more of the following beneficial effects:

[0047] This invention provides a power quality assessment method and apparatus based on multi-source data fusion, comprising: acquiring power quality parameters at the power supply end, and performing wavelet transform on the power quality parameters at the power supply end to obtain the information entropy corresponding to the power quality parameters at the power supply end; using the information entropy corresponding to the power quality parameters at the power supply end as input to a pre-trained SVM classifier to obtain the power quality assessment level of the power supply end output by the pre-trained SVM classifier; acquiring power quality parameters at the user end, and using the power quality parameters at the user end as input to a pre-built PCNN model to obtain the power quality assessment level of the user end output by the pre-built PCNN model; and determining the power system power command assessment result based on the power quality assessment level at the power supply end and the power quality assessment level at the user end. The technical solution provided by this invention performs power quality analysis on both the power supply end and the user end. In particular, the PCNN model is used on the user end, making the results more accurate. At the same time, interference injection on the transmission line can simulate the environmental and interference changes in actual transmission, eliminating the need to retest every time a change occurs, making actual testing simpler. In addition, the final power quality level is obtained by using a weighted method, taking into account the influence of the three links in the transmission process, and effectively reflecting the overall power quality. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the main steps of the power quality evaluation method based on multi-source data fusion according to an embodiment of the present invention. Detailed Implementation

[0049] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] As disclosed in the background section, power quality refers to the electrical performance requirements met during the transmission and utilization of electrical energy in a power supply system, and is an important indicator for evaluating power system performance. Power quality issues mainly include voltage fluctuations, frequency fluctuations, harmonics, voltage sags, and voltage flicker. In modern power systems, the stability and quality of power quality are of great significance for ensuring the safe operation of equipment, improving production efficiency, and reducing energy waste.

[0052] Currently, the main methods for power quality analysis and evaluation include traditional measurement methods and numerical simulation methods based on computer simulation. Measurement methods involve directly collecting field data, such as voltage and current waveforms, by installing sensors, and then analyzing the data. Numerical simulation methods, on the other hand, use computer software to simulate the operating state of the power system, predict power quality problems, and analyze the power quality situation based on the simulation results.

[0053] However, existing power quality analysis and evaluation methods require significant human and material resources for measurement, are easily affected by external environmental factors, and face difficulties in data collection. Furthermore, measurement methods can only provide point data, making it difficult to comprehensively reflect the overall situation of power quality issues. Numerical simulation methods require accurate input of various parameters during modeling and have high computational complexity, making them time-consuming and labor-intensive. Therefore, existing methods need further improvement in terms of real-time performance and accuracy, especially for power quality problems in complex power systems, where current methods often fail to provide accurate analysis.

[0054] To address the aforementioned issues, this invention provides a power quality assessment method and apparatus based on multi-source data fusion, comprising: acquiring power quality parameters at the power supply end, and performing wavelet transform on the power quality parameters at the power supply end to obtain the information entropy corresponding to the power quality parameters at the power supply end; using the information entropy corresponding to the power quality parameters at the power supply end as input to a pre-trained SVM classifier to obtain the power quality assessment level of the power supply end output by the pre-trained SVM classifier; acquiring power quality parameters at the user end, and using the power quality parameters at the user end as input to a pre-built PCNN model to obtain the power quality assessment level of the user end output by the pre-built PCNN model; and determining the power system power command assessment result based on the power supply end power quality assessment level and the user end power quality assessment level. The technical solution provided by this invention performs power quality analysis on both the power supply end and the user end. In particular, the PCNN model is used on the user end, making the results more accurate. At the same time, interference injection on the transmission line can simulate the environmental and interference changes in actual transmission, eliminating the need to retest every time a change occurs, making actual testing simpler. In addition, the final power quality level is obtained by using a weighted method, taking into account the influence of the three links in the transmission process, and effectively reflecting the overall power quality.

[0055] The above plan will be explained in detail below.

[0056] Example 1

[0057] See appendix Figure 1 , Figure 1 This is a schematic flowchart illustrating the main steps of a power quality assessment method based on multi-source data fusion, according to an embodiment of the present invention. Figure 1As shown, the power quality assessment method based on multi-source data fusion in this embodiment of the invention mainly includes the following steps:

[0058] Step S101: Obtain the power quality parameters of the power supply end, and perform wavelet transform on the power quality parameters of the power supply end to obtain the information entropy corresponding to the power quality parameters of the power supply end;

[0059] Step S102: Use the information entropy corresponding to the power quality parameters of the power supply terminal as the input of the pre-trained SVM classifier to obtain the power quality assessment level of the power supply terminal output by the pre-trained SVM classifier.

[0060] Step S103: Obtain the power quality parameters of the user terminal, and use the power quality parameters of the user terminal as the input of the pre-built PCNN model to obtain the power quality assessment level of the user terminal output by the pre-built PCNN model;

[0061] Step S104: Determine the power system power command assessment result based on the power quality assessment level at the power supply end and the power quality assessment level at the user end.

[0062] The power quality parameters include at least one of the following: voltage, current, frequency, and harmonics.

[0063] In one specific embodiment, the power quality assessment method based on multi-source data fusion provided by the present invention includes the following steps:

[0064] S1. A first analysis system is set up at the power supply end. The first analysis system includes a signal acquisition unit, an ADC converter, and a CPU. The signal acquisition unit collects the signals transmitted from the power supply end. The collected signals are converted into digital signals by the ADC converter and then transmitted to the CPU. The CPU can perform wavelet transform on the digital signals to obtain the characteristic parameters of power quality. Based on the characteristic parameters, the corresponding power quality level A1 of the power supply end is determined.

[0065] Specifically, the process of performing wavelet transform and obtaining power quality level A1 in step S1 includes the following steps: performing dual-tree complex wavelet decomposition on the digital signal. The dual-tree complex wavelet uses a binary tree structure for discrete wavelet transform, with one tree representing the real part and the other the imaginary part. The real and imaginary parts form a Hilbert transform pair, which is a 90-degree phase difference in the frequency domain. Filtering the two trees with a filter yields the information entropy of multiple wavelet coefficients. These multiple information entropies are used as feature parameters in the form of feature vectors. An SVM classifier is used to identify each feature parameter, thereby obtaining the power quality level A1. For example, with 10 feature parameters, the SVM classifier can identify 10 feature vectors. Then, the kernel function inside the SVM classifier is used to transform the feature vectors into wavelet function form, so that the 10 feature vectors are used as training samples to obtain the power quality level A1. Here, a commonly used SVM classifier can be used for ease of use. The specific kernel function can be either the Morlet wavelet function or the Mexico wavelet function. The transformation methods for these two wavelet functions are highly accurate and publicly available, so they will not be elaborated further here. Using dual-tree complex wavelet decomposition and an SVM classifier, the power quality level can be effectively obtained. Dual-tree complex wavelet decomposition is a complex form of wavelet transform, capable of handling non-stationary signals and possessing high time-frequency locality properties, which closely matches the characteristics of circuit transmission.

[0066] It's important to note that wavelet transform is a signal processing technique used to decompose a signal into wavelet functions of different scales for analysis in both the time and frequency domains. Wavelet transform helps capture instantaneous changes and frequency features in signals, providing more comprehensive information across different time scales and frequencies. An SVM classifier is a supervised learning model-based instrument used for pattern recognition, classification, and regression analysis. Here, wavelet transform is used for signal processing and feature parameter extraction. The extracted feature parameters can be used as input data to train the SVM classifier. This combination fully utilizes the feature extraction capabilities of wavelet transform and the classification performance of SVM, resulting in relatively simple classification, i.e., identifying power quality levels.

[0067] S2. After determining the power quality level A1, a second analysis system is set up at the user end. This second analysis system includes a multi-terminal monitoring instrument and a user power monitoring platform. The multi-terminal monitoring instrument, also a type of signal acquisition instrument, is installed between the transmission line and the user's input terminal. It can connect to multiple user devices, collecting signals from each device and transmitting them to the user power monitoring platform for processing. The user power monitoring platform uses a PCNN model to analyze the power quality of the signals transmitted by the multi-terminal monitoring instrument and determine the corresponding power quality level A2 at the user end.

[0068] In this embodiment, in step S2, the PCNN model uses multiple neurons to perform power quality analysis on the signal transmitted by the multi-terminal monitoring device. Each neuron includes a receiving part, a modulation part, and a pulse part. The receiving part is equivalent to the input end of the neuron and is used to receive the transmitted signal. The modulation part modulates the signal received by the receiving part using offset and multiplication modulation. This part is equivalent to adjusting the strength and form of the signal for subsequent processing. The pulse part is equivalent to the output end of the neuron and is used to determine the power quality of the modulated signal and output power quality level A2.

[0069] Specifically, for each neuron in a PCNN, the receiving section has an information threshold. If the received signal is less than this threshold, the neuron will not proceed with further information transmission. Once the received signal reaches the threshold, the neuron will ignite and generate a potential, which is then integrated with information from other neurons and transmitted together through two or more channels to the corresponding modulation section. The modulation section adds an offset to the signal of each channel, multiplying and modulating different channels to improve the signal representation and obtain a modulated signal that is easier to process. The pulse section consists of a threshold regulator, a comparator, and a pulse generator. The threshold regulator adjusts the ignition threshold at the neuron's output to ensure that a pulse can be generated. The comparator compares the adjusted signal with the ignition threshold set by the threshold regulator to determine whether a pulse should be generated, which is equivalent to determining whether the power quality is acceptable. The pulse generator generates a pulse signal based on the comparison result of the comparator.

[0070] The threshold settings mentioned above correspond to the power quality levels. There can be multiple power quality levels, and thus multiple thresholds can be set accordingly. This allows the pulse generator to output a corresponding pulse signal at the corresponding threshold, which in turn indicates the power quality level.

[0071] It's important to note that the PCNN model is a pulse-coupled neural network model, a self-supervised learning network model that can extract effective information from complex data without prior training. This characteristic makes the PCNN model more efficient and flexible in real-time evaluation. On the user side, not only are there numerous devices and varied environments, but the specifications of the devices used by users may also change constantly. Therefore, using the PCNN model allows for real-time evaluation of power quality and rapid response to changes in various data within the power grid, aligning with actual user scenarios. Conversely, on the power supply side, equipment and lines are generally relatively stable, so relatively simple wavelet transforms and SVM classifiers can be used to complete the power quality level testing and analysis.

[0072] S3. An interference injection model is added to the transmission line to simulate harmonic interference and signal attenuation that may occur during transmission. The interference injection model includes at least a harmonic sub-model and a length attenuation sub-model. The harmonic sub-model is used to input harmonics for interference into the transmission line, and the length attenuation sub-model is used to adjust the impedance of the transmission line. Steps S1 and S2 are repeated multiple times to obtain multiple power quality levels A1 and A2. The harmonic sub-model is used to simulate various types of harmonic interference that frequently occur in actual transmission; the length attenuation sub-model is used to simulate the attenuation problem of the transmission line. Attenuation may be caused by impedance changes due to changes in transmission length, leading to signal attenuation, or by attenuation caused by skin effect and transmission medium loss, all of which can lead to changes in power quality.

[0073] Specifically, in step S3, the harmonic sub-model adopts a nonlinear load model or an odd harmonic model, and the length attenuation sub-model also includes a composite noise module, which consists of white noise, pink noise, and impulse noise. The composite noise module can also simulate noise problems in actual transmission, making the results more accurate. Nonlinear load models are commonly used for interference injection, capable of inputting harmonic currents or voltages, both of which are frequently encountered in power electronic equipment. Odd-order harmonic models are used to input homogeneous harmonics, which are very common in actual transmission and are one of the important factors affecting power quality. White noise is random noise with the same energy density at all frequencies. In power systems, white noise may originate from various random factors, such as thermal noise and shot noise. Pink noise is a type of noise with a relatively uniform frequency distribution, whose energy density gradually decreases with increasing frequency. Pink noise is common in nature, such as wind and rain sounds. In power systems, pink noise may originate from electromagnetic radiation from certain nonlinear components in the power grid or transmission lines. Impulse noise is a sudden, short-duration noise pulse. Impulse noise in power systems may be caused by various transient processes, such as switching operations and short-circuit faults. Impulse noise has a significant impact on the stability of the power grid and power quality.

[0074] In this embodiment, step S3, the interference injection model may further include a load sub-model. The load sub-model is used to add multiple different loads to the transmission line. Each load sub-model contains multiple specific circuits, each a hybrid model combining various electronic devices such as capacitors, resistors, inductors, field-effect transistors, integrated circuits, and chips. These circuits are primarily connected in parallel to the transmission circuit. The specific circuits of the load sub-model need to be simulated based on the actual equipment used. Each specific circuit corresponds to an equivalent circuit of an actual device. For example, if a rechargeable battery is to be connected, a specific circuit for the rechargeable battery is set up. The added loads may introduce new noise and may also affect the impedance of the transmission line. The load sub-model can effectively simulate the impact of various external devices on power quality during actual transmission.

[0075] S4. Use the weighting method to process the multiple power quality levels A1 and A2 in step S3 to obtain the final power quality level A3.

[0076] In this embodiment, step S4, the specific process of obtaining the final power quality level A3 using the weighting method includes the following steps: setting a first weighting coefficient p and a second weighting coefficient q, where p and q are both non-negative numbers, calculating the mean B1 of multiple power quality levels A1 and the mean B2 of multiple power quality levels A2, and A3 = p*B1 + q*B2. The weighted method can effectively determine the overall power quality level. The two weighting coefficients can be set by the tester according to the needs and the overall transmission situation. For example, in the first scenario, the power supply end has fewer devices and the power supply and power quality levels remain stable, while the number of devices used on the user end far exceeds that of the power supply end. At the same time, the devices on the user end are constantly changing, and the power quality level on the user end is also frequently changing. In this case, the first weighting coefficient p can be lowered and the second weighting coefficient q can be increased to more accurately reflect the overall power quality level. In the second scenario, the number of devices used on the power supply end and the user end is similar, and the devices at both ends are frequently replaced, resulting in a high degree of device change. In this case, both the first weighting coefficient p and the second weighting coefficient q can be set to 0.5 to better reflect the overall power quality level.

[0077] Furthermore, it should be noted that since both the average power quality level B1 of the power supply end and the power quality level B2 of the user end are obtained, and the overall power quality level A3 is also calculated, testers can take into account both local and overall situations. When it is necessary to perform power quality analysis on a specific transmission link, the first weighting coefficient p or the second weighting coefficient q can be set to 0 directly. The result obtained in this way reflects the power quality level of the power supply end or the power quality level of the user end.

[0078] In summary, this invention performs power quality analysis on both the power supply end and the user end, with the PCNN model used at the user end for more accurate results. Furthermore, interference injection on the transmission line simulates environmental and interference changes during actual transmission, eliminating the need for retesting every time a change occurs and simplifying practical testing. Moreover, the weighted method used to obtain the final power quality level takes into account the influence of all three stages of transmission, effectively reflecting the overall power quality and demonstrating significant advancements.

[0079] Example 2

[0080] Based on the same inventive concept, the present invention also provides a power quality assessment device for multi-source data fusion, the power quality assessment device for multi-source data fusion comprising:

[0081] The acquisition module is used to acquire the power quality parameters of the power supply end, and perform wavelet transform on the power quality parameters of the power supply end to obtain the information entropy corresponding to the power quality parameters of the power supply end.

[0082] The first analysis module is used to take the information entropy corresponding to the power quality parameters of the power supply terminal as the input of a pre-trained SVM classifier to obtain the power quality assessment level of the power supply terminal output by the pre-trained SVM classifier.

[0083] The second analysis module is used to obtain the power quality parameters of the user end, and use the power quality parameters of the user end as the input of the pre-built PCNN model to obtain the power quality assessment level of the user end output by the pre-built PCNN model.

[0084] The evaluation module is used to determine the power system power command evaluation result based on the power quality evaluation level at the power supply end and the power quality evaluation level at the user end.

[0085] Preferably, the power quality parameters include at least one of the following: voltage, current, frequency, and harmonics.

[0086] Preferably, the wavelet transform is a dual-tree complex wavelet transform.

[0087] Preferably, the training process of the pre-trained SVM classifier includes:

[0088] Obtain the power quality parameters of the historical power supply terminal, and perform wavelet transform on the power quality parameters of the historical power supply terminal to obtain the information entropy corresponding to the power quality parameters of the historical power supply terminal;

[0089] The information entropy corresponding to the power quality parameters of the historical power supply terminals is manually annotated with power quality assessment levels, and training data is constructed using the information entropy corresponding to the power quality parameters of the historical power supply terminals with annotated power quality assessment levels.

[0090] The initial SVM classifier is trained using the training data to obtain the pre-trained SVM classifier.

[0091] Furthermore, the kernel function of the SVM classifier is either the Morlet wavelet function or the Mexico wavelet function.

[0092] Preferably, the device includes: an interference module, used to add an interference injection model to the transmission line between the power supply end and the user end to simulate harmonic interference and signal attenuation during the transmission process.

[0093] Furthermore, the interference injection model includes: a harmonic sub-model that inputs harmonics for interference into the transmission line and a length attenuation sub-model that adjusts the impedance of the transmission line.

[0094] Furthermore, the harmonic sub-model is a nonlinear load model or an odd harmonic model.

[0095] Preferably, the power system energy command evaluation results are as follows:

[0096] A3 = p*B1 + q*B2

[0097] In the above formula, A3 is the power system power command assessment result, B1 is the average power quality assessment level at the power supply end, B2 is the power quality assessment level at the user end, p is the first weight, q is the second weight, and p+1 = 1.

[0098] Example 3

[0099] Based on the same inventive concept, this invention also provides a computer device, which includes a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to implement corresponding method flows or corresponding functions, thereby realizing the steps of the multi-source data fusion power quality evaluation method in the above embodiments.

[0100] Example 4

[0101] Based on the same inventive concept, this invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the steps of the power quality assessment method for multi-source data fusion in the above embodiments.

[0102] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0104] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0105] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A power quality assessment method based on multi-source data fusion, characterized in that, The method includes: The power quality parameters of the power supply end are obtained, and wavelet transform is performed on the power quality parameters of the power supply end to obtain the information entropy corresponding to the power quality parameters of the power supply end. The information entropy corresponding to the power quality parameters of the power supply terminal is used as the input of a pre-trained SVM classifier to obtain the power quality assessment level of the power supply terminal output by the pre-trained SVM classifier. Obtain the power quality parameters of the user terminal, and use the power quality parameters of the user terminal as the input of a pre-built PCNN model to obtain the power quality assessment level of the user terminal output by the pre-built PCNN model; The power system power command assessment result is determined based on the power quality assessment level at the power supply end and the power quality assessment level at the user end.

2. The method as described in claim 1, characterized in that, The power quality parameters include at least one of the following: voltage, current, frequency, and harmonics.

3. The method as described in claim 1, characterized in that, The wavelet transform is a dual-tree complex wavelet transform.

4. The method as described in claim 1, characterized in that, The training process of the pre-trained SVM classifier includes: Obtain the power quality parameters of the historical power supply terminal, and perform wavelet transform on the power quality parameters of the historical power supply terminal to obtain the information entropy corresponding to the power quality parameters of the historical power supply terminal; The information entropy corresponding to the power quality parameters of the historical power supply terminals is manually annotated with power quality assessment levels, and training data is constructed using the information entropy corresponding to the power quality parameters of the historical power supply terminals with annotated power quality assessment levels. The initial SVM classifier is trained using the training data to obtain the pre-trained SVM classifier.

5. The method as described in claim 4, characterized in that, The kernel function of the SVM classifier is either the Morlet wavelet function or the Mexico wavelet function.

6. The method as described in claim 1, characterized in that, The method includes adding an interference injection model to the transmission line between the power supply end and the user end to simulate harmonic interference and signal attenuation during the transmission process.

7. The method as described in claim 6, characterized in that, The interference injection model includes: a harmonic sub-model that inputs harmonics for interference into the transmission line and a length attenuation sub-model that adjusts the impedance of the transmission line.

8. The method as described in claim 7, characterized in that, The harmonic sub-model is a nonlinear load model or an odd harmonic model.

9. The method as described in claim 1, characterized in that, The evaluation results of the power system energy command are as follows: A3 = p*B1 + q*B2 In the above formula, A3 is the power system power command assessment result, B1 is the average power quality assessment level at the power supply end, B2 is the power quality assessment level at the user end, p is the first weight, q is the second weight, and p+1 = 1.

10. A power quality assessment device based on multi-source data fusion, characterized in that, The device includes: The acquisition module is used to acquire the power quality parameters of the power supply end, and perform wavelet transform on the power quality parameters of the power supply end to obtain the information entropy corresponding to the power quality parameters of the power supply end. The first analysis module is used to take the information entropy corresponding to the power quality parameters of the power supply terminal as the input of a pre-trained SVM classifier to obtain the power quality assessment level of the power supply terminal output by the pre-trained SVM classifier. The second analysis module is used to obtain the power quality parameters of the user end, and use the power quality parameters of the user end as the input of the pre-built PCNN model to obtain the power quality assessment level of the user end output by the pre-built PCNN model. The evaluation module is used to determine the power system power command evaluation result based on the power quality evaluation level at the power supply end and the power quality evaluation level at the user end.

11. The apparatus as claimed in claim 9, characterized in that, The power quality parameters include at least one of the following: voltage, current, frequency, and harmonics.

12. The apparatus as claimed in claim 9, characterized in that, The wavelet transform is a dual-tree complex wavelet transform.

13. The apparatus as claimed in claim 9, characterized in that, The training process of the pre-trained SVM classifier includes: Obtain the power quality parameters of the historical power supply terminal, and perform wavelet transform on the power quality parameters of the historical power supply terminal to obtain the information entropy corresponding to the power quality parameters of the historical power supply terminal; The information entropy corresponding to the power quality parameters of the historical power supply terminals is manually annotated with power quality assessment levels, and training data is constructed using the information entropy corresponding to the power quality parameters of the historical power supply terminals with annotated power quality assessment levels. The initial SVM classifier is trained using the training data to obtain the pre-trained SVM classifier.

14. The apparatus as claimed in claim 13, characterized in that, The kernel function of the SVM classifier is either the Morlet wavelet function or the Mexico wavelet function.

15. The apparatus as claimed in claim 9, characterized in that, The device includes an interference module, used to add an interference injection model to the transmission line between the power supply end and the user end to simulate harmonic interference and signal attenuation during the transmission process.

16. The apparatus as claimed in claim 15, characterized in that, The interference injection model includes: a harmonic sub-model that inputs harmonics for interference into the transmission line and a length attenuation sub-model that adjusts the impedance of the transmission line.

17. The apparatus as claimed in claim 16, characterized in that, The harmonic sub-model is a nonlinear load model or an odd harmonic model.

18. The apparatus as claimed in claim 9, characterized in that, The evaluation results of the power system energy command are as follows: A3 = p*B1 + q*B2 In the above formula, A3 is the power system power command assessment result, B1 is the average power quality assessment level at the power supply end, B2 is the power quality assessment level at the user end, p is the first weight, q is the second weight, and p+1 = 1.

19. A computer device, characterized in that, include: One or more processors; The processor is used to execute one or more programs; When the one or more programs are executed by the one or more processors, the power quality assessment method based on multi-source data fusion as described in any one of claims 1 to 9 is implemented.

20. A computer-readable storage medium, characterized in that, It contains a computer program, which, when executed, implements the power quality evaluation method based on multi-source data fusion as described in any one of claims 1 to 9.

Images