CN-121978612-A - Full-performance test real-time monitoring method and system based on edge calculation

CN121978612ACN 121978612 ACN121978612 ACN 121978612ACN-121978612-A

Abstract

A full-performance test real-time monitoring method and system based on edge calculation. The method comprises the steps of collecting electric energy meter data through test equipment, smoothing, interpolating and self-adapting to the collected data in edge calculation to supplement signal deficiency, carrying out frequency spectrum transformation on signals for many times, carrying out weighted fusion on frequency spectrum results of each stage to obtain a composite frequency spectrum, adopting a self-adapting signal filtering algorithm based on composite limit detection to filter interference and abnormal signals, adopting an abnormal quantification method to carry out quantitative assessment on the filtered signals and the composite frequency spectrum, carrying out real-time early warning on fusion signal characteristics based on a neural-decision network model so as to detect performance abnormality and respond in time, and realizing real-time monitoring of a full-performance test. The scheme of the invention realizes real-time and accurate detection and evaluation of the electric energy meter signal.

Inventors

CAO XIAODONG
ZHAO SHUANGSHUANG
CHEN WENGUANG
Wang siyun
YI YONGXIAN
BAO JIN
XIA GUOFANG

Assignees

国网江苏省电力有限公司营销服务中心
国网江苏省电力有限公司

Dates

Publication Date: 20260505
Application Date: 20260403

Claims (10)

1. The full-performance test real-time monitoring method based on edge calculation is characterized by comprising the following steps of: step S1, acquiring electric energy meter data through test equipment, smoothing, interpolating and self-adapting adjusting the acquired data in edge calculation to supplement signal loss, carrying out spectrum transformation on the signals for a plurality of times, and carrying out weighted fusion on spectrum results of each stage to obtain a composite spectrum; And S2, filtering interference and abnormal signals by adopting a self-adaptive signal filtering algorithm based on composite limit detection, quantitatively evaluating the filtered signals and the composite spectrum by adopting an abnormal quantization method, and carrying out real-time early warning on the characteristics of the fusion signals based on a neural-decision network model so as to detect performance abnormality and respond in time, thereby realizing real-time monitoring of a full-performance test.
2. The method for monitoring the full-performance test based on the edge calculation in real time according to claim 1, wherein the collecting the electric energy meter data further comprises: The method comprises the steps of collecting current, voltage, power data, electric energy meter parameters and test conditions, configuring a filter and a gain adjuster with adjustable bandwidth for each collecting channel to adapt to signals with different intensities and frequency ranges, firstly, carrying out preliminary frequency screening on the collected signals through the filter, removing high-frequency noise or irrelevant frequency bands, ensuring that target signals can be accurately captured, adjusting the amplitude of the signals through the gain adjuster to adapt to the requirements of digital conversion, and adopting a high-precision analog-digital converter to sample the collected signals with high efficiency.
3. The method for real-time monitoring of full performance test based on edge calculation according to claim 2, wherein the smoothing, interpolation and adaptive adjustment of the collected data further comprises the steps of calculating: Wherein, the Representing the electric energy meter signal after dynamic expansion; The acquired electric energy meter signals are subjected to time translation and weighting adjustment to form signal expansion; is the total signal translation segment number; Is the maximum amplitude of the signal at the ith translation stage; Is the average amplitude of the signal during the ith translation stage; is the time interval when the signal translates in the ith translation stage; is a noise compensation coefficient; Is an exponential decay coefficient; Is the time length of the integration interval, fills the signal gap by translation and interpolation, and reduces noise interference by a compensation mechanism.
4. The method for real-time monitoring of full performance test based on edge calculation according to claim 3, wherein the performing spectrum transformation on the signal for multiple times and performing weighted fusion on the spectrum result of each stage further comprises: obtaining a composite spectrum through spectrum fusion : Wherein, the Is an electric energy meter signal after dynamic expansion The frequency spectrum after K-order Fourier transform and wavelet transform; is the weighting coefficient of the k-th order spectrum; is the total number of spectrum fusion stages; Is the attenuation factor of the frequency spectrum, Is a frequency variable.
5. The method for real-time monitoring full-performance test based on edge calculation as set forth in claim 4, wherein said adaptive signal filtering algorithm based on composite limit detection is used to filter out interference and abnormal signals, further comprising calculating a filtered electric energy meter signal : Wherein, the And Representing the lower and upper thresholds of the signal, respectively; Is the minimum compensation factor; is the maximum compensation factor.
6. The method for real-time monitoring of full performance test based on edge calculation according to claim 5, wherein said performing quantization evaluation on the filtered signal and the composite spectrum by using an anomaly quantization method further comprises calculating: Wherein, the Is at the time of Abnormal quantization characteristics of the electric energy meter signal for quantizing the signal in a time period Abnormal changes in; Is a signal Is a second derivative of (2); is the integration interval of the light source, Is a weighting coefficient of the spectral feature, Is a composite spectrum At the time of Sum frequency Range of The maximum amplitude value within the range, Is the minimum and maximum frequency used in the spectral analysis; is the adjustment coefficient.
7. The edge-computing-based full performance test real-time monitoring method of claim 6, wherein the neural-decision network model comprises: Wherein, the The evaluation result of the performance abnormality is represented, and the interference intensity or state at the time t is represented; Is the first The weighting coefficients of the individual neurons; Is the number of neurons; A nonlinear activation function; is the j-th anomaly quantized feature Weights of (2); Is that Dividing according to time period to obtain The j-th abnormal quantization feature in the subinterval; Is a bias term; Is an early warning coefficient; is a time delay and early warning response correction term.
8. The utility model provides a full performance test real-time monitoring system based on edge calculation which characterized in that includes: The composite spectrum generation module is used for acquiring data of the electric energy meter through test equipment, smoothing, interpolating and self-adaptively adjusting the acquired data in edge calculation to supplement signal loss, carrying out spectrum transformation on the signals for a plurality of times, and carrying out weighted fusion on spectrum results of each stage to obtain composite spectrum; The real-time monitoring module is used for filtering interference and abnormal signals by adopting a self-adaptive signal filtering algorithm based on composite limit detection, carrying out quantitative evaluation on the filtered signals and the composite spectrum by adopting an abnormal quantization method, and carrying out real-time early warning on the characteristics of the fusion signals based on a neural-decision network model so as to detect performance abnormality and respond in time, thereby realizing real-time monitoring of a full-performance test.
9. A terminal comprises a processor and a storage medium, and is characterized in that: The storage medium is used for storing instructions; The processor is operative according to the instructions to perform the steps of the edge calculation based full performance test real time monitoring method according to any one of claims 1-7.
10. A computer readable storage medium having stored thereon a computer program, characterized in that the program when executed by a processor implements the steps of the edge computing based full performance test real time monitoring method according to any of claims 1-7.

Description

Full-performance test real-time monitoring method and system based on edge calculation Technical Field The invention belongs to the field of full-performance test monitoring, and particularly relates to a full-performance test real-time monitoring method and system based on edge calculation. Background With the wide application of smart grids and modern power devices, electric energy meters play an increasingly critical role as an important tool for power metering. Traditional electric energy meters mainly rely on mechanical and analog signal processing techniques, and although basic electric power measurement tasks can be completed, the traditional electric energy meters still have some limitations in terms of complex signal processing, real-time monitoring and high-precision testing. Particularly, under the dynamic change and complex environment, the traditional electric energy meter signal acquisition and processing method is difficult to solve the problems of signal loss, noise interference, high-frequency data change and the like. Currently, the monitoring requirements of power systems tend to be higher precision, stronger real-time and more complex fault detection capabilities, especially in the fields of performance testing and fault diagnosis. How to extract and analyze effective information of signals in real time on the premise of ensuring the data integrity of the electric energy meter has become an important challenge in the research of the electric energy meter and related equipment. Meanwhile, the rapid development of edge calculation provides a new technical approach for signal processing of the traditional electric energy meter. The edge calculation can perform real-time data processing and analysis near the data acquisition point, reduces data transmission delay and bandwidth pressure, and provides more efficient support for monitoring the power equipment. However, because the electric energy meter signal is greatly affected by environmental noise, especially in a complex practical application scenario, how to effectively supplement missing data, remove noise interference, and efficiently extract key features in the signal is still a problem to be solved. The prior art lacks real-time analysis and self-adaptive adjustment of dynamic signal characteristics, so that all interference and abnormal signals cannot be effectively removed, particularly in an environment with large signal amplitude fluctuation or strong noise components, and the traditional electric energy meter signal abnormality detection method mostly relies on simple threshold judgment and experience rules, lacks efficient model and algorithm support, so that abnormal changes of the electric energy meter signals cannot be accurately detected and estimated in real time in a complex environment, and the accuracy and reliability of a full-performance test are seriously affected. Disclosure of Invention In order to solve the defects in the prior art, the invention provides a full-performance test real-time monitoring method and system based on edge calculation, which are used for solving the technical problem that all interference and abnormal signals cannot be effectively removed due to the lack of real-time analysis and self-adaptive adjustment of dynamic signal characteristics. In order to solve the technical problems, the invention adopts the following technical scheme. The invention discloses a full-performance test real-time monitoring method based on edge calculation, which comprises the following steps: step S1, acquiring electric energy meter data through test equipment, smoothing, interpolating and self-adapting adjusting the acquired data in edge calculation to supplement signal loss, carrying out spectrum transformation on the signals for a plurality of times, and carrying out weighted fusion on spectrum results of each stage to obtain a composite spectrum; And S2, filtering interference and abnormal signals by adopting a self-adaptive signal filtering algorithm based on composite limit detection, quantitatively evaluating the filtered signals and the composite spectrum by adopting an abnormal quantization method, and carrying out real-time early warning on the characteristics of the fusion signals based on a neural-decision network model so as to detect performance abnormality and respond in time, thereby realizing real-time monitoring of a full-performance test. The invention further comprises the following preferable schemes: the collecting electric energy meter data further comprises: The method comprises the steps of collecting current, voltage, power data, electric energy meter parameters and test conditions, configuring a filter and a gain adjuster with adjustable bandwidth for each collecting channel to adapt to signals with different intensities and frequency ranges, firstly, carrying out preliminary frequency screening on the collected signals through the filter, removing high-frequency noise or irrelevant freq