Substation GIS equipment state evaluation model based on multi-frequency signal processing

Abstract

The invention discloses a transformer substation GIS equipment state evaluation model based on multi-frequency signal processing, which can comprehensively utilize a multi-frequency signal processing technology, improve the detection precision and the anti-interference capability, comprehensively evaluate the state of GIS equipment and provide a scientific basis for maintenance and management of the equipment. Multi-frequency signal real-time acquisition is carried out on the GIS equipment, and the acquired data is transmitted to the signal preprocessing module for processing; extracting a frequency characteristic parameter, a time-frequency characteristic parameter or a statistical characteristic parameter of the signal; training a machine learning algorithm by adopting the training sample, and establishing a GIS equipment state evaluation model; inputting the extracted multi-frequency signal characteristic parameters into an evaluation model, and outputting a state evaluation result in a visual mode, so that a user can intuitively know the state of the GIS equipment; according to a state evaluation result, diagnosis and early warning are carried out on faults of the GIS equipment, and a user is reminded to take measures in time; the method can be widely applied to the field of network security operation systems.

Description

Technical Field

[0001] The invention relates to a substation GIS equipment state assessment model for multi-frequency signal processing, belonging to the technical field of power grid safety operation systems. Background Art

[0002] With the development of society and the advancement of science and technology, the complexity and importance of power systems are increasing. In modern power networks, gas-insulated switchgear (GIS) is widely used in substations and distribution systems due to its advantages such as small footprint, high reliability and low maintenance cost. However, GIS equipment may still fail during operation, resulting in power outages and safety hazards. Therefore, it is particularly important to detect GIS equipment failures in a timely manner and eliminate them in the bud.

[0003] At present, the status monitoring of GIS equipment mainly adopts traditional electrical detection methods, such as partial discharge detection, ultra-high frequency detection, etc. Although these methods can detect the failure of GIS equipment to a certain extent, they have problems such as low detection accuracy and poor anti-interference ability. In addition, traditional detection methods can often only detect signals of a single frequency and cannot fully reflect the status of GIS equipment. Summary of the invention

[0004] The present invention overcomes the shortcomings of the prior art and provides a substation GIS equipment status assessment model based on multi-frequency signal processing. The model can comprehensively utilize multi-frequency signal processing technology to improve detection accuracy and anti-interference ability, comprehensively evaluate the status of GIS equipment, and provide a scientific basis for equipment maintenance and management.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a substation GIS equipment status assessment model based on multi-frequency signal processing, including signal acquisition, signal preprocessing, status assessment and result output; wherein: Signal acquisition: Use a variety of sensors to collect multi-frequency signals from GIS equipment in real time, and transmit the collected data to the signal preprocessing module for processing; Signal preprocessing: Use digital filters or wavelet transforms to denoise the collected multi-frequency signals to remove noise and interference in the signals, then use digital filters or bandpass filters to filter the denoised signals to remove high-frequency noise and low-frequency interference in the signals; then use amplifiers to amplify the filtered signals to increase the amplitude of the signals, and finally use fast Fourier transforms, wavelet transforms or statistical analysis methods to analyze the preprocessed signals to extract the frequency characteristic parameters, time-frequency characteristic parameters or statistical characteristic parameters of the signals.

[0006] Condition assessment: Use training samples to train the machine learning algorithm and establish a GIS equipment condition assessment model; input the extracted multi-frequency signal characteristic parameters into the assessment model and automatically output the condition assessment results of the GIS equipment; Result output: The status assessment results are output in a visual way so that users can intuitively understand the status of GIS equipment; and based on the status assessment results, GIS equipment faults are diagnosed and warned, reminding users to take timely measures.

[0007] Preferably, the sensor includes a combination of two or three of an ultra-high frequency sensor, an ultrasonic sensor and an ultra-high frequency sensor.

[0008] Preferably, the machine learning algorithms used for status assessment of GIS equipment include but are not limited to support vector machines and artificial neural networks.

[0009] Preferably, the visual output includes but is not limited to reports and graphics.

[0010] Compared with the prior art, the present invention has the following beneficial effects.

[0011] 1. Improve detection accuracy and anti-interference ability The present invention adopts multi-frequency signal processing technology, which can comprehensively utilize multiple signals such as ultra-high frequency, ultrasonic wave, and ultra-high frequency to improve detection accuracy and anti-interference ability; it adopts digital signal processing technology, such as fast Fourier transform, wavelet transform, etc., to analyze and process the signal, extract the characteristic parameters of the signal, and further improve the detection accuracy and anti-interference ability.

[0012] 2. Comprehensively evaluate the status of GIS equipment The present invention uses a machine learning algorithm to evaluate the status of GIS equipment based on the characteristic parameters of multi-frequency signals, which can fully reflect the operating status of GIS equipment. Moreover, the status evaluation module can automatically adjust the parameters of the evaluation model according to different signal characteristics and fault types to improve the accuracy and reliability of the evaluation. Among them, the machine learning algorithm used for the status evaluation of GIS equipment includes but is not limited to support vector machines and artificial neural networks.

[0013] 3. Provide scientific basis for equipment maintenance and management The present invention outputs the status assessment results in a visual manner, providing users with intuitive equipment status information. Moreover, the result output module can also provide fault diagnosis and early warning functions, timely discover and handle equipment failures, and provide a scientific basis for equipment maintenance and management. DETAILED DESCRIPTION

[0014] The present invention will be further described below in conjunction with specific embodiments.

[0015] The present invention provides a substation GIS equipment status assessment model based on multi-frequency signal processing, including signal acquisition, signal preprocessing, status assessment and result output; wherein: Signal acquisition: A variety of sensors are used to collect multi-frequency signals of GIS equipment in real time, and the collected data is transmitted to the signal preprocessing module for processing; wherein the sensors include two or three combinations of ultra-high frequency sensors, ultrasonic sensors, and ultra-high frequency sensors.

[0016] Signal preprocessing: Use digital filters or wavelet transforms to denoise the collected multi-frequency signals to remove noise and interference in the signals, then use digital filters or bandpass filters to filter the denoised signals to remove high-frequency noise and low-frequency interference in the signals; then use amplifiers to amplify the filtered signals to increase the amplitude of the signals, and finally use fast Fourier transforms, wavelet transforms or statistical analysis methods to analyze the preprocessed signals to extract the frequency characteristic parameters, time-frequency characteristic parameters or statistical characteristic parameters of the signals.

[0017] Condition assessment: Use training samples to train the machine learning algorithm and establish a GIS equipment condition assessment model; input the extracted multi-frequency signal characteristic parameters into the assessment model and automatically output the condition assessment results of the GIS equipment; Result output: The status assessment results are output in a visual manner so that users can intuitively understand the status of the GIS equipment; and based on the status assessment results, the faults of the GIS equipment are diagnosed and warned, reminding users to take timely measures, wherein the visual output includes but is not limited to reports and graphics.

[0018] 1. Improve detection accuracy and anti-interference ability The present invention adopts multi-frequency signal processing technology, which can comprehensively utilize multiple signals such as ultra-high frequency, ultrasonic wave, and ultra-high frequency to improve detection accuracy and anti-interference ability; it adopts digital signal processing technology, such as fast Fourier transform, wavelet transform, etc., to analyze and process the signal, extract the characteristic parameters of the signal, and further improve the detection accuracy and anti-interference ability.

[0019] 2. Comprehensively evaluate the status of GIS equipment The present invention uses a machine learning algorithm to evaluate the status of GIS equipment based on the characteristic parameters of multi-frequency signals, which can fully reflect the operating status of GIS equipment. Moreover, the status evaluation module can automatically adjust the parameters of the evaluation model according to different signal characteristics and fault types, thereby improving the accuracy and reliability of the evaluation.

[0020] 3. Provide scientific basis for equipment maintenance and management The present invention outputs the status assessment results in a visual manner, providing users with intuitive equipment status information. Moreover, the result output module can also provide fault diagnosis and early warning functions, timely discover and handle equipment failures, and provide a scientific basis for equipment maintenance and management.

[0021] The substation GIS equipment status assessment model based on multi-frequency signal processing provided by the present invention can comprehensively utilize multi-frequency signal processing technology to improve detection accuracy and anti-interference ability, comprehensively assess the status of GIS equipment, and provide a scientific basis for equipment maintenance and management.

[0022] The present invention has been described in detail above in conjunction with the embodiments, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in the field without departing from the purpose of the present invention.

Claims

1. Substation GIS equipment status assessment model based on multi-frequency signal processing, It is characterized in that It includes signal acquisition, signal preprocessing, state evaluation and result output; among which: Signal acquisition: Use a variety of sensors to collect multi-frequency signals from GIS equipment in real time, and transmit the collected data to the signal preprocessing module for processing; Signal preprocessing: Use digital filters or wavelet transforms to denoise the collected multi-frequency signals to remove noise and interference in the signals, then use digital filters or bandpass filters to filter the denoised signals to remove high-frequency noise and low-frequency interference in the signals; then use amplifiers to amplify the filtered signals to increase the amplitude of the signals, and finally use fast Fourier transforms, wavelet transforms or statistical analysis methods to analyze the preprocessed signals to extract the frequency characteristic parameters, time-frequency characteristic parameters or statistical characteristic parameters of the signals.

2. Status evaluation: Use training samples to train the machine learning algorithm and establish a GIS equipment status evaluation model; input the extracted multi-frequency signal characteristic parameters into the evaluation model and automatically output the status evaluation results of the GIS equipment; Result output: Output the status assessment results in a visual way so that users can intuitively understand the status of GIS equipment; Based on the status assessment results, GIS equipment failures are diagnosed and warned, reminding users to take timely measures.

3. The substation GIS equipment status assessment model based on multi-frequency signal processing according to claim 1, It is characterized in that The sensor includes a combination of two or three of an ultra-high frequency sensor, an ultrasonic sensor, and an ultra-high frequency sensor.

4. The substation GIS equipment status assessment model based on multi-frequency signal processing according to claim 1, It is characterized in that The machine learning algorithms used for status assessment of GIS equipment include but are not limited to support vector machines and artificial neural networks.

5. The substation GIS equipment status assessment model based on multi-frequency signal processing according to claim 1, It is characterized in that The visualization output includes but is not limited to reports and graphics.