Full-performance test real-time monitoring method and system based on edge computing

Abstract

The invention discloses a full-performance test real-time monitoring method and system based on edge computing. The method comprises the following steps: acquiring electric energy meter data through test equipment, performing smoothing, interpolation and adaptive adjustment on the acquired data in edge calculation to supplement signal loss, performing frequency spectrum transformation on signals for multiple times, and performing weighted fusion on frequency spectrum results of each stage to obtain a composite frequency spectrum; the adaptive signal filtering algorithm based on composite limit detection is adopted to filter interference and abnormal signals, the abnormal quantification method is adopted to carry out quantitative evaluation on the filtered signals and the composite frequency spectrum, and real-time early warning is carried out on the fused signal features based on the neural-decision network model to detect performance abnormity and make a response in time. And real-time monitoring of the full-performance test is realized. According to the scheme of the invention, real-time and accurate detection and evaluation of the electric energy meter signal are realized.

Description

Technical Field

[0001] This invention belongs to the field of full-performance test monitoring, and specifically relates to a real-time monitoring method and system for full-performance tests based on edge computing. Background Technology

[0002] With the widespread application of smart grids and modern power equipment, electricity meters, as an important tool for electricity measurement, are playing an increasingly crucial role. Traditional electricity meters mainly rely on mechanical and analog signal processing technologies. While they can complete basic power measurement tasks, they still have limitations in processing complex signals, real-time monitoring, and high-precision testing. Especially in dynamically changing and complex environments, traditional electricity meter signal acquisition and processing methods struggle to cope with problems such as signal loss, noise interference, and high-frequency data variations.

[0003] Currently, the monitoring needs of power systems are trending towards higher precision, stronger real-time performance, and more complex fault detection capabilities, especially in the fields of performance testing and fault diagnosis. How to extract and analyze effective information from signals in real time while ensuring the integrity of electricity meter data has become a significant challenge in the research of electricity meters and related equipment.

[0004] Meanwhile, the rapid development of edge computing has provided a new technological approach for traditional electricity meter signal processing. Edge computing enables real-time data processing and analysis near the data acquisition point, reducing data transmission latency and bandwidth pressure, and providing more efficient support for power equipment monitoring. However, since electricity meter signals are greatly affected by environmental noise, especially in complex real-world application scenarios, how to effectively supplement missing data, remove noise interference, and efficiently extract key features from the signal remains a problem that urgently needs to be solved.

[0005] Existing technologies lack real-time analysis and adaptive adjustment of dynamic signal characteristics, making it impossible to effectively remove all interference and abnormal signals, especially in environments with large signal amplitude fluctuations or strong noise components. Traditional methods for detecting abnormal signals in electricity meters mostly rely on simple threshold judgments and empirical rules, lacking efficient model and algorithm support. This results in the inability to detect and evaluate abnormal changes in electricity meter signals in real time and accurately in complex environments, seriously affecting the accuracy and reliability of full-performance tests. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a real-time monitoring method and system for full-performance testing based on edge computing, which solves the technical problem of failing to effectively remove all interference and abnormal signals due to the lack of real-time analysis and adaptive adjustment of dynamic signal characteristics.

[0007] To solve the above-mentioned technical problems, the present invention adopts the following technical solution.

[0008] This invention first discloses a real-time monitoring method for full-performance testing based on edge computing, which includes the following steps: Step S1: Collect electricity meter data through test equipment, smooth, interpolate and adaptively adjust the collected data in edge computing to supplement signal loss, perform multiple spectrum transformations on the signal, and weight and fuse the spectrum results of each stage to obtain a composite spectrum. Step S2: An adaptive signal filtering algorithm based on composite limit detection is used to filter out interference and abnormal signals. An anomaly quantization method is used to quantify and evaluate the filtered signal and composite spectrum. A neural-decision network model is used to provide real-time early warning of the fused signal characteristics in order to detect performance anomalies and respond in a timely manner, thereby achieving real-time monitoring of the full performance test.

[0009] The present invention further includes the following preferred embodiments: The collection of electricity meter data further includes: The system collects current, voltage, power, and energy data, as well as electricity meter parameters and test conditions. Each acquisition channel is equipped with an adjustable bandwidth filter and gain regulator to accommodate signals of different intensities and frequency ranges. First, the acquired signals are preliminarily frequency-screened by the filter to remove high-frequency noise or irrelevant frequency bands, ensuring that the target signal can be accurately captured. The gain regulator is used to adjust the amplitude of the signal to meet the requirements of digital conversion. A high-precision analog-to-digital converter is used to efficiently sample the acquired signals.

[0010] The process of smoothing, interpolating, and adaptively adjusting the collected data further includes calculations:

[0011] in, This represents the energy meter signal after dynamic expansion; It is the collected electricity meter signal that, after time shifting and weighted adjustment, forms an extended signal; It is the total number of signal shift segments; It is the maximum amplitude of the signal during the i-th translation phase; It is the average amplitude of the signal during the i-th translation phase; It is the time interval during the signal translation in the i-th translation stage; It is the noise compensation coefficient; It is the exponential decay coefficient; It is the time length of the integration interval; the gaps in the signal are filled by translation and interpolation, and noise interference is reduced through compensation mechanisms.

[0012] The process of performing multiple spectral transformations on the signal and weighting and fusing the spectral results from each stage further includes: A composite spectrum is obtained through spectrum fusion. :

[0013] in, It is the dynamically expanded energy meter signal. The spectrum after k-th order Fourier transform and wavelet transform; These are the weighting coefficients of the k-th order spectrum; This represents the total number of spectrum fusion stages; It is the attenuation factor of the spectrum. It is a frequency variable.

[0014] The method of using an adaptive signal filtering algorithm based on composite limit detection to filter out interference and abnormal signals further includes calculating the filtered energy meter signal. :

[0015]

[0016] in, and These represent the lower and upper threshold values ​​of the signal, respectively. It is the minimum compensation factor; It is the maximum compensation factor.

[0017] The step of using anomaly quantization to quantize and evaluate the filtered signal and composite spectrum further includes calculating:

[0018] in, It is in time Abnormal quantization characteristics of the electricity meter signal are used to quantize the signal over a time period. Abnormal changes within; It is a signal The second derivative; It is the integration interval. These are the weighting coefficients for the spectral characteristics. It is a composite spectrum In time and frequency scope Maximum amplitude value within, These are the minimum and maximum frequencies used in spectrum analysis; It is the adjustment coefficient.

[0019] The neural-decision network-based model includes:

[0020] in, The evaluation result indicates the performance anomaly, representing the intensity or state of the disturbance at time t; It is the first The weight coefficients of each neuron; It refers to the number of neurons; It is a non-linear activation function; It is the j-th abnormal quantization feature The weights; yes Divide into time periods to obtain The j-th abnormal quantization feature in each sub-interval; It is a bias term; It is the early warning coefficient; This refers to the time delay and early warning response correction item.

[0021] This invention also discloses an edge computing-based real-time monitoring system for full-performance testing, utilizing the aforementioned edge computing-based real-time monitoring method, comprising: The composite spectrum generation module is used to collect electricity meter data through test equipment, smooth, interpolate and adaptively adjust the collected data in edge computing to supplement signal loss, and obtain the composite spectrum by performing multiple spectrum transformations on the signal and weighting and fusing the spectrum results of each stage. The real-time monitoring module is used to filter out interference and abnormal signals using an adaptive signal filtering algorithm based on composite limit detection, to quantitatively evaluate the filtered signal and composite spectrum using an anomaly quantification method, and to provide real-time early warning of the fused signal characteristics based on a neural-decision network model, so as to detect performance anomalies and respond in a timely manner, thereby realizing real-time monitoring of the full performance test.

[0022] Accordingly, this application also discloses a terminal, including a processor and a storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the aforementioned real-time monitoring method for full-performance testing based on edge computing.

[0023] Accordingly, this application also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned real-time monitoring method for full-performance testing based on edge computing.

[0024] The beneficial effects of this invention are as follows: Compared with the prior art, this invention provides a real-time monitoring method and system for full-performance testing based on edge computing. It introduces a dynamic signal expansion scheme, effectively compensating for data loss caused by acquisition delay or signal interference through interpolation, smoothing, and adaptive adjustment of the acquired signal, ensuring that the signal dataset is expanded in the time domain and avoiding the loss of important performance characteristics. Multiple Fourier transforms and wavelet transforms are performed on the signal, combined with weighted fusion technology, significantly improving the resolution of spectrum analysis, reducing high-frequency noise interference, accurately extracting frequency components from the signal, and removing the influence of noise on the signal frequency characteristics. An adaptive signal filtering algorithm based on composite limit detection dynamically adjusts the upper and lower thresholds of the filter to remove noise and abnormal signals from the energy meter signal in real time, ensuring signal clarity and stability. This filtering mechanism can automatically adjust according to the characteristics of the electricity meter signal in different environments, reducing the impact of interference on the signal. By quantifying the acceleration characteristics and spectral features of the electricity meter signal, it can accurately detect abnormal changes in the signal. The combination of the second derivative and the maximum value of the spectrum enables the timely capture and analysis of abnormal changes in the signal in both the time and frequency domains, thereby improving the system's anomaly detection accuracy. The early warning model based on a neural-decision network can integrate multiple signal features and output accurate early warning signals. Through nonlinear activation functions and adaptive adjustment mechanisms, it efficiently captures complex relationships in the electricity meter signal, providing support for timely response to performance anomalies. Attached Figure Description

[0025] Figure 1 This is a flowchart of the real-time monitoring method for full-performance testing based on edge computing in this invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0027] The embodiments described in this application are merely some, not all, embodiments of the present invention. Based on the spirit of the present invention, other embodiments obtained by those skilled in the art without inventive effort are all within the protection scope of the present invention.

[0028] To address the shortcomings of existing technologies, this invention proposes a real-time monitoring method and system for full-performance testing based on edge computing. (See [link to relevant documentation]). Figure 1 As shown, the real-time monitoring method for full-performance testing based on edge computing disclosed in this invention includes the following steps: Step S1: Collect electricity meter data through test equipment, smooth, interpolate and adaptively adjust the collected data in edge computing to supplement signal loss, perform multiple spectrum transformations on the signal, and weight and fuse the spectrum results of each stage to obtain a composite spectrum. Step S2: An adaptive signal filtering algorithm based on composite limit detection is used to filter out interference and abnormal signals. An anomaly quantization method is used to quantify and evaluate the filtered signal and composite spectrum. A neural-decision network model is used to provide real-time early warning of the fused signal characteristics in order to detect performance anomalies and respond in a timely manner, thereby achieving real-time monitoring of the full performance test.

[0029] In step S1, data from the electricity meter is first collected using testing equipment to ensure a comprehensive reflection of the meter's actual condition. The collected data includes current, voltage, power, energy, meter parameters, and test conditions. Each acquisition channel is equipped with an adjustable bandwidth filter and gain regulator to accommodate signals of varying intensities and frequencies. The acquired signal undergoes initial frequency filtering through the filter to remove high-frequency noise or irrelevant frequency bands, ensuring accurate capture of the target signal. Subsequently, the gain regulator adjusts the signal amplitude to meet the requirements of digital conversion, thus preventing data distortion due to excessively weak or strong signals. A high-precision analog-to-digital converter (ADC) with a conversion accuracy of 12 bits or higher is used to efficiently sample the acquired signal at a frequency of millions of bits per second, stably and accurately capturing fluctuations in the electricity meter signal.

[0030] To further improve the real-time performance and accuracy of data acquisition, edge computing is used to process the acquired data in real time. In edge computing, due to computation and transmission delays, signal acquisition may experience time-domain gaps. To compensate for these missing signal data, step S1 proposes a dynamic signal expansion scheme. By supplementing and reconstructing the acquired electricity meter signals, a complete and continuous signal set is generated, thereby enhancing the accuracy and real-time performance of subsequent analysis. The dynamic signal expansion scheme expands the signal dataset through smoothing, interpolation, and adaptive adjustment, ensuring that the signal is expanded in the time domain and preventing the loss of important performance characteristics. The specific mathematical model can be expressed as follows:

[0031] in, This represents the energy meter signal after dynamic expansion, with the aim of generating a complete and continuous signal sequence; It is the collected electricity meter signal that, after time shifting and weighted adjustment, forms an extended signal; It represents the total number of signal shift segments, indicating the number of positions the signal has shifted to at different times; It is the maximum amplitude of the signal during the i-th translation phase; It is the average amplitude of the signal during the i-th translation phase; It is the time interval during the signal translation in the i-th translation stage, representing the amount of translation of the signal on the time axis; It is the noise compensation coefficient, used to reduce the impact of noise during the interpolation process. It is calculated by the ratio of the standard deviation of the ambient noise to the signal strength. It is the exponential decay coefficient, which controls the speed of noise compensation and is obtained through the bandwidth of the signal and the frequency distribution characteristics of the noise; It is the time length of the integration interval. By filling the gaps in the signal through translation and interpolation, and reducing noise interference through compensation mechanisms, the time-domain characteristics of the signal can be recovered.

[0032] Even after dynamic expansion, the frequency characteristics of the electricity meter signal may still be affected by noise. Spectrum analysis is commonly used to extract frequency components from a signal, but traditional spectrum analysis methods may be insufficient to capture subtle features in complex signals. Therefore, this invention proposes a multi-stage, multi-spectral fusion processing method. By performing multiple spectrum transformations on the signal and weighting and fusing the spectral results from each stage, the accuracy and stability of spectrum analysis are significantly improved. Specifically, the dynamically expanded electricity meter signal undergoes multiple Fourier transforms (FTs) and wavelet transforms (WTs) to extract the signal's spectral information. Then, by weighting and fusing the results from different spectral stages, a comprehensive spectral output is obtained to further remove high-frequency noise components.

[0033] The process of spectrum fusion is represented by the following formula:

[0034] in, It is a composite spectrum obtained from time-frequency analysis, which describes the common characteristics of the signal in the time and frequency domains; It is the dynamically expanded energy meter signal. The spectrum after k-th order transformation (Fourier transform, wavelet transform); These are the weighting coefficients of the k-th order spectrum, representing the contribution of different spectral analysis stages to the composite spectrum result, obtained through experiments; This is the total number of spectrum fusion stages, which is set according to the specific signal analysis needs, and is usually selected as 3 to 5 stages; It is the attenuation factor of the spectrum, used to suppress high-frequency noise. Its magnitude is usually determined based on the bandwidth of the signal spectrum and the intensity of the noise. It is a frequency variable, representing the frequency components of the signal in the frequency domain. (Exponential decay function) This helps reduce interference from high-frequency signals, ensuring that only effective low-frequency information is retained in the spectrum. Through multiple spectrum transformations and weighted fusion, the resolution of spectrum analysis can be significantly improved, reducing the impact of noise on signal frequency components, thereby providing more accurate data for subsequent processing.

[0035] Preferably, in step S2, to further purify the signal, an adaptive signal filtering algorithm based on composite limit detection is proposed. This algorithm dynamically sets upper and lower threshold values ​​and adaptively adjusts their values ​​to identify and filter out interference and abnormal signals. It utilizes the acceleration characteristics and amplitude changes of the signal to adjust the filter parameters in real time to adapt to the signal characteristics of the energy meter under different environments. The adaptive signal filtering process based on composite limit detection is as follows:

[0036]

[0037] in, It is the filtered signal from the electricity meter; and These represent the lower and upper thresholds of the signal, respectively, used to determine whether the signal is within the valid range; It is the minimum compensation factor, used to suppress excessive amplification when the signal exceeds the effective range; It is the maximum compensation factor, used to enhance the amplitude of the signal when it exceeds the effective range. By analyzing the rate of change of the signal, the threshold is dynamically calculated, and the effective range of the signal is automatically adjusted, thereby removing interference components that exceed this range. Through filter adjustment, interference signals are further removed, thus ensuring the clarity of the electricity meter signal. Through an adaptive signal filtering mechanism based on composite limit detection, abnormal noise components in the signal can be suppressed in real time, ensuring the stability and accuracy of the signal.

[0038] Anomalies in electricity meter signals are typically manifested in changes in their acceleration characteristics and spectral features. In particular, when the signal is subjected to external interference, the energy distribution in the frequency domain and the variation characteristics in the time domain will exhibit significant anomalies. To capture these anomalies, step S2 processes the second derivative of the filtered electricity meter signal to reflect the instantaneous rate of change and acceleration of the signal, thereby uncovering potential anomalies. Furthermore, the composite spectrum in the frequency domain is used as auxiliary information, and the maximum value in the frequency domain is extracted to capture abnormal fluctuations in the signal frequency components. For the joint characteristics in the time and frequency domains, the influence of the signal strength is adjusted through an exponential decay function to reflect the severity of different anomalies.

[0039] Preferably, the mathematical model for anomaly quantification is expressed as:

[0040] in, It is in time Abnormal quantization characteristics of the electricity meter signal are used to quantize the signal over a time period. Abnormal changes within; It is a signal The second derivative of the signal represents the second derivative of the signal. The acceleration reflects the rapid changes in signal fluctuations; It is the integration interval, representing the time window for anomaly quantization, and defining the duration of anomaly quantization; These are the weighting coefficients of the spectral characteristics, which adjust the importance of the composite spectrum in anomaly quantization; It is a composite spectrum In time and frequency scope Maximum amplitude value within, These are the minimum and maximum frequencies used in spectrum analysis; This is an adjustment coefficient that modulates the effect of signal oscillation amplitude on the quantization process; it was obtained experimentally. By quantifying abnormal features, rapid detection and accurate assessment of signal anomalies can be achieved.

[0041] Preferably, in order to provide real-time early warning based on anomaly quantification features, step S2 designs an early warning model based on a neural-decision network. Through a multi-layered network structure, various features of the signal are fused to output an early warning signal. The activation function of the neural-decision network adopts nonlinear functions such as Sigmoid or Tanh to enhance the model's ability to capture complex signal relationships. The early warning output model formula is as follows:

[0042] in, The evaluation result indicates the performance anomaly, representing the intensity or state of the disturbance at time t; It is the first The weight coefficients of each neuron are used to adjust the signal strength input to the neuron; It refers to the number of neurons; It is a non-linear activation function that enhances the network's ability to capture complex signal relationships; It is the j-th abnormal quantization feature The weights; yes Divide into time periods to obtain The j-th abnormal quantization feature in each sub-interval; It is a bias term that adjusts the output of each neural network node; It is the early warning coefficient, used to adjust the output of the neural-decision network; This is a time delay and early warning response correction term, used to adjust the system's response latency. Through a neural-decision network model, it can output accurate early warning results based on real-time signal data, enabling timely detection and response to performance anomalies.

[0043] Early warning output The anomaly of the electricity meter signal was evaluated using a neural-decision network model, providing an intuitive basis for judging whether the signal exceeds performance standards. The warning output can be represented as a binary value (such as 0 or 1) to indicate whether the signal meets performance requirements, or it may be a continuous value indicating the degree of anomaly. The output results can trigger subsequent response mechanisms, such as adjusting the working status of the test equipment, enabling additional filtering or compensation measures, or issuing alarm notifications so that relevant personnel can take timely intervention measures.

[0044] The beneficial effects of this invention are as follows: Compared with the prior art, this invention provides a real-time monitoring method and system for full-performance testing based on edge computing. It introduces a dynamic signal expansion scheme, effectively compensating for data loss caused by acquisition delay or signal interference through interpolation, smoothing, and adaptive adjustment of the acquired signal, ensuring that the signal dataset is expanded in the time domain and avoiding the loss of important performance characteristics. Multiple Fourier transforms and wavelet transforms are performed on the signal, combined with weighted fusion technology, significantly improving the resolution of spectrum analysis, reducing high-frequency noise interference, accurately extracting frequency components from the signal, and removing the influence of noise on the signal frequency characteristics. An adaptive signal filtering algorithm based on composite limit detection dynamically adjusts the upper and lower thresholds of the filter to remove noise and abnormal signals from the energy meter signal in real time, ensuring signal clarity and stability. This filtering mechanism can automatically adjust according to the characteristics of the electricity meter signal in different environments, reducing the impact of interference on the signal. By quantifying the acceleration characteristics and spectral features of the electricity meter signal, it can accurately detect abnormal changes in the signal. The combination of the second derivative and the maximum value of the spectrum enables the timely capture and analysis of abnormal changes in the signal in both the time and frequency domains, thereby improving the system's anomaly detection accuracy. The early warning model based on a neural-decision network can integrate multiple signal features and output accurate early warning signals. Through nonlinear activation functions and adaptive adjustment mechanisms, it efficiently captures complex relationships in the electricity meter signal, providing support for timely response to performance anomalies.

[0045] This invention can be a system, method, and / or computer program product. This invention also discloses an edge computing-based real-time monitoring system for full-performance testing, based on the aforementioned edge computing-based real-time monitoring method, comprising: The composite spectrum generation module is used to collect electricity meter data through test equipment, smooth, interpolate and adaptively adjust the collected data in edge computing to supplement signal loss, and obtain the composite spectrum by performing multiple spectrum transformations on the signal and weighting and fusing the spectrum results of each stage. The real-time monitoring module is used to filter out interference and abnormal signals using an adaptive signal filtering algorithm based on composite limit detection, to quantitatively evaluate the filtered signal and composite spectrum using an anomaly quantification method, and to provide real-time early warning of the fused signal characteristics based on a neural-decision network model, so as to detect performance anomalies and respond in a timely manner, thereby realizing real-time monitoring of the full performance test.

[0046] Based on the spirit of this invention, those skilled in the art will readily conceive of a computer program product derived from the aforementioned edge computing-based real-time monitoring method for full-performance testing. The computer program product may include a computer-readable storage medium on which computer-readable program instructions are loaded to enable a processor to implement various aspects of this disclosure. Specifically, this application also includes a terminal comprising a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the aforementioned edge computing-based real-time monitoring method for full-performance testing.

[0047] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0048] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0049] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0050] 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 protection scope of the claims of the present invention.

Claims

1. A real-time monitoring method for full-performance testing based on edge computing, characterized in that, Includes the following steps: Step S1: Collect electricity meter data through test equipment, smooth, interpolate and adaptively adjust the collected data in edge computing to supplement signal loss, perform multiple spectrum transformations on the signal, and weight and fuse the spectrum results of each stage to obtain a composite spectrum. Step S2: An adaptive signal filtering algorithm based on composite limit detection is used to filter out interference and abnormal signals. An anomaly quantization method is used to quantify and evaluate the filtered signal and composite spectrum. A neural-decision network model is used to provide real-time early warning of the fused signal characteristics in order to detect performance anomalies and respond in a timely manner, thereby achieving real-time monitoring of the full performance test.

2. The real-time monitoring method for full-performance testing based on edge computing according to claim 1, characterized in that, The collection of electricity meter data further includes: The system collects current, voltage, power, and energy data, as well as electricity meter parameters and test conditions. Each acquisition channel is equipped with an adjustable bandwidth filter and gain regulator to accommodate signals of different intensities and frequency ranges. First, the acquired signals are preliminarily frequency-screened by the filter to remove high-frequency noise or irrelevant frequency bands, ensuring that the target signal can be accurately captured. The gain regulator is used to adjust the amplitude of the signal to meet the requirements of digital conversion. A high-precision analog-to-digital converter is used to efficiently sample the acquired signals.

3. The real-time monitoring method for full-performance testing based on edge computing according to claim 2, characterized in that, The process of smoothing, interpolating, and adaptively adjusting the collected data further includes calculations: in, This represents the energy meter signal after dynamic expansion; It is the collected electricity meter signal that, after time shifting and weighted adjustment, forms an extended signal; It is the total number of signal shift segments; It is the maximum amplitude of the signal during the i-th translation phase; It is the average amplitude of the signal during the i-th translation phase; It is the time interval during the signal translation in the i-th translation stage; It is the noise compensation coefficient; It is the exponential decay coefficient; It is the time length of the integration interval; the gaps in the signal are filled by translation and interpolation, and noise interference is reduced through compensation mechanisms.

4. The real-time monitoring method for full-performance testing based on edge computing according to claim 3, characterized in that, The process of performing multiple spectral transformations on the signal and weighting and fusing the spectral results from each stage further includes: A composite spectrum is obtained through spectrum fusion. : in, It is the dynamically expanded energy meter signal. The spectrum after k-th order Fourier transform and wavelet transform; These are the weighting coefficients of the k-th order spectrum; This represents the total number of spectrum fusion stages; It is the attenuation factor of the spectrum. It is a frequency variable.

5. The real-time monitoring method for full-performance testing based on edge computing according to claim 4, characterized in that, The method of using an adaptive signal filtering algorithm based on composite limit detection to filter out interference and abnormal signals further includes calculating the filtered energy meter signal. : in, and These represent the lower and upper threshold values ​​of the signal, respectively. It is the minimum compensation factor; It is the maximum compensation factor.

6. The real-time monitoring method for full-performance testing based on edge computing according to claim 5, characterized in that, The step of using anomaly quantization to quantize and evaluate the filtered signal and composite spectrum further includes calculating: in, It is in time Abnormal quantization characteristics of the electricity meter signal are used to quantize the signal over a time period. Abnormal changes within; It is a signal The second derivative; It is the integration interval. These are the weighting coefficients for the spectral characteristics. It is a composite spectrum In time and frequency scope Maximum amplitude value within, These are the minimum and maximum frequencies used in spectrum analysis; It is the adjustment coefficient.

7. The real-time monitoring method for full-performance testing based on edge computing according to claim 6, characterized in that, The neural-decision network-based model includes: in, The evaluation result indicates the performance anomaly, representing the intensity or state of the disturbance at time t; It is the first The weight coefficients of each neuron; It refers to the number of neurons; It is a non-linear activation function; It is the j-th abnormal quantization feature The weights; yes Divide into time periods to obtain The j-th abnormal quantization feature in each sub-interval; It is a bias term; It is the early warning coefficient; This refers to the time delay and early warning response correction item.

8. A real-time monitoring system for full-performance testing based on edge computing, characterized in that, include: The composite spectrum generation module is used to collect electricity meter data through test equipment, smooth, interpolate and adaptively adjust the collected data in edge computing to supplement signal loss, and obtain the composite spectrum by performing multiple spectrum transformations on the signal and weighting and fusing the spectrum results of each stage. The real-time monitoring module is used to filter out interference and abnormal signals using an adaptive signal filtering algorithm based on composite limit detection, to quantitatively evaluate the filtered signal and composite spectrum using an anomaly quantification method, and to provide real-time early warning of the fused signal characteristics based on a neural-decision network model, so as to detect performance anomalies and respond in a timely manner, thereby realizing real-time monitoring of the full performance test.

9. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the edge computing-based real-time monitoring method for full-performance testing as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the real-time monitoring method for full-performance testing based on edge computing as described in any one of claims 1-7.

Images