Transformer partial discharge on-line monitoring system based on artificial intelligence

Abstract

The invention relates to the field of transformer monitoring, is used for solving the problem that the monitoring result is single in the partial discharge phenomenon monitoring process and the partial discharge phenomenon change dynamic monitoring is lacked, and particularly relates to an artificial intelligence-based transformer partial discharge on-line monitoring system. According to the invention, the partial discharge phenomenon of the transformer is obtained through the acoustic emission instrument, diversified data analysis is carried out on the partial discharge phenomenon, and the discharge intensity, the discharge fluctuation condition, the discharge position movement condition and the development and change trend of the partial discharge phenomenon of the transformer are obtained. Then combined analysis is carried out according to the monitored partial discharge intensity and the discharge position intensity change phenomenon of the transformer, a breakdown low-risk alarm or a breakdown high-risk alarm is obtained according to a combined analysis result, and corresponding early warning and automatic control measures are generated according to the breakdown low-risk alarm or the breakdown high-risk alarm, so that safe operation of the transformer is guaranteed, and the safety of the transformer is improved. And the monitoring effect of the partial discharge phenomenon is improved.

Description

Technical Field

[0001] The present invention relates to the field of transformer online monitoring, and in particular to an artificial intelligence-based transformer partial discharge online monitoring system. Background Art

[0002] Partial discharge refers to the discharge that occurs in a local area of ​​the insulation of power equipment under the action of a sufficiently strong electric field. This type of discharge is limited to causing only a local short circuit in the insulation between conductors without forming a conductive channel. The partial discharge phenomenon mainly occurs in high-voltage electrical equipment. Each partial discharge will have some impact on the insulating medium. A slight partial discharge has little effect on the insulation of power equipment, and the insulation strength decreases slowly; while a strong partial discharge will cause the insulation strength to decrease rapidly. Partial discharge is an important cause of the final insulation breakdown of high-voltage electrical equipment and an important sign of insulation degradation. As a common high-voltage electrical equipment, the partial discharge phenomenon of the transformer will interfere with the operation of the transformer and cause damage to the transformer. Therefore, it is necessary to monitor the partial discharge phenomenon of the transformer to ensure the normal use of the transformer.

[0003] At present, the transformer partial discharge monitoring system in the prior art still has shortcomings. When the transformer has partial discharge, it will be accompanied by electric pulses, ultrasound, electromagnetic radiation, light, and chemical reactions. Therefore, the prior art mostly uses acoustic emission instruments to monitor the partial discharge phenomenon. In the existing monitoring system, the monitored partial discharge phenomenon is basically directly output, lacking in-depth analysis and combined use of partial discharge phenomenon data, resulting in a single output monitoring result. Although partial discharge monitoring can be achieved, the development law and occurrence trend of the transformer partial discharge phenomenon cannot be analyzed, thereby reducing the comprehensiveness of the monitoring results in the transformer partial discharge phenomenon monitoring operation;

[0004] In view of the above technical problems, this application proposes a solution. Summary of the invention

[0005] In the present invention, the partial discharge phenomenon of the transformer is obtained by an acoustic emission instrument, and a diversified data analysis is performed on the partial discharge phenomenon to obtain the discharge intensity, discharge fluctuation, discharge position movement and development and change trend of the partial discharge phenomenon of the transformer, thereby realizing comprehensive monitoring of the partial discharge phenomenon of the transformer, improving the monitoring effect of the partial discharge phenomenon, solving the problem that the monitoring result is single and lacks monitoring of the dynamic changes of the partial discharge phenomenon during the partial discharge phenomenon monitoring process, and proposing a transformer partial discharge online monitoring system based on artificial intelligence.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The transformer partial discharge online monitoring system based on artificial intelligence includes an information acquisition unit, which is used to collect monitoring information of the transformer;

[0008] A processing and generating unit, wherein the processing and generating unit can analyze the monitoring information acquired by the information acquisition unit, and generate a partial discharge position and a partial discharge intensity according to the monitoring information;

[0009] A discharge evaluation unit, wherein the discharge evaluation unit can perform a comprehensive analysis based on the local discharge position and the local discharge intensity, generate a discharge exceeding standard signal or a discharge not exceeding standard signal according to the local discharge intensity judgment result, analyze according to the local discharge intensity change factor, generate a discharge continuation signal, a discharge random signal, a discharge aggravation signal or a discharge stable signal, and analyze according to the local discharge position to generate a position fixed signal or a position random signal;

[0010] a discharge statistics unit, wherein the discharge statistics unit obtains a discharge evaluation signal through the discharge evaluation unit, and performs combined analysis according to the discharge evaluation signal to generate different warning signals;

[0011] The monitoring and early warning unit obtains the early warning signal and the discharge evaluation signal through the discharge statistics unit, and generates a corresponding alarm reminder.

[0012] As a preferred embodiment of the present invention, when the information acquisition unit monitors the transformer, a plurality of groups of acquisition ports of acoustic emitters are arranged at different positions on the transformer, an acoustic wave acquisition matrix is ​​constructed on the transformer, and monitoring information of the transformer is collected using the acquisition matrix of the plurality of groups of acoustic emitters, wherein the monitoring information includes acoustic wave waveform, acquisition time and acquisition port position.

[0013] As a preferred embodiment of the present invention, after receiving the monitoring information, the processing and generating unit deletes the noise interference waveform and the waveform that does not reach the threshold value in the sound wave waveform to obtain the discharge action waveform. The processing and generating unit processes the monitoring information obtained by all the acquisition ports to obtain multiple groups of discharge action waveforms. The processing and generating unit performs spatial algorithm simulation on the multiple groups of discharge action waveforms and the time when the sound wave waveform is collected to obtain the generation position of the discharge action waveform in three-dimensional space, which is recorded as the local discharge position. The processing and generating unit performs algorithm simulation based on the discharge waveform to obtain the local discharge intensity when the discharge waveform is generated. The processing and generating unit sends the local discharge position and the local discharge intensity to the discharge evaluation unit.

[0014] As a preferred embodiment of the present invention, after the discharge evaluation unit obtains the local discharge position and the local discharge intensity, the local discharge intensity is compared with a preset discharge intensity. If the local discharge intensity is greater than or equal to the preset discharge intensity, a discharge exceeding standard signal is generated; if the local discharge intensity is less than the preset discharge intensity, a discharge not exceeding standard signal is generated.

[0015] As a preferred embodiment of the present invention, after the discharge evaluation unit generates the discharge exceeding standard signal, the local discharge intensity that generates the discharge exceeding standard signal is recorded. When the discharge evaluation unit obtains the local discharge intensity again subsequently, the discharge evaluation unit performs difference calculation on the local discharge intensities obtained multiple times, and calculates the result of the difference calculation by using a variance formula to obtain the discharge intensity fluctuation. The discharge evaluation unit compares the discharge intensity fluctuation with a preset fluctuation value. If the discharge intensity fluctuation is greater than a preset fluctuation threshold, a discharge random signal is generated. If the discharge intensity fluctuation is less than or equal to the preset fluctuation threshold, a discharge continuation signal is generated.

[0016] After generating the discharge not exceeding standard signal, the discharge evaluation unit records the local discharge intensity that generates the discharge not exceeding standard signal, and records the new local discharge intensity in chronological order each time after obtaining the discharge not exceeding standard signal subsequently. The discharge evaluation unit selects the n+1th group of local discharge intensity to the 2nth group of local discharge intensity, calculates the discharge mean, obtains multiple groups of discharge mean values, and calculates the difference between two adjacent groups of discharge mean values, and records it as the discharge change rate. The discharge evaluation unit compares the discharge change rate with a preset threshold value, generates a discharge aggravation signal if the discharge change rate is greater than the preset threshold value, and generates a discharge stable signal if the discharge change rate is less than or equal to the preset threshold value.

[0017] As a preferred embodiment of the present invention, the discharge evaluation unit counts the local discharge positions obtained each time, compares the statistically obtained local discharge positions, and generates a fixed position signal if the distance dispersion degree between the local discharge positions obtained multiple times is within a preset range; if the distance dispersion degree between the local discharge positions obtained multiple times is outside the preset range, a random position signal is generated.

[0018] As a preferred embodiment of the present invention, the method for the discharge evaluation unit to obtain the distance dispersion degree is:

[0019] Step 1: The discharge assessment unit marks all partial discharge locations in the 3D model;

[0020] Step 2: Select any partial discharge position as the starting point, and connect lines from the starting point to all other partial discharge positions to obtain the total distance of the discharge positions;

[0021] Step 3: Repeat step 2 multiple times, so that each local discharge position is used as a starting point to obtain a total distance of the discharge position, and select the smallest group of total distances of all discharge positions;

[0022] Step 4: Divide the total distance of the selected discharge positions by the total number of all partial discharge positions to obtain the distance dispersion degree.

[0023] As a preferred embodiment of the present invention, the discharge statistics unit does not respond after obtaining the discharge exceeding standard signal through the discharge evaluation unit. After the discharge statistics unit obtains the discharge not exceeding standard signal through the discharge evaluation unit:

[0024] If the discharge statistics unit obtains the position fixed signal and the discharge aggravation signal at the same time, a breakdown high risk alarm is generated and sent to the monitoring and early warning unit;

[0025] If the discharge statistics unit simultaneously obtains the position fixed signal and the discharge stable signal, a breakdown low risk alarm is generated; if the discharge statistics unit simultaneously obtains the position random signal and the discharge aggravation signal, a breakdown low risk alarm is generated. The discharge statistics unit sends the breakdown low risk alarm to the monitoring and early warning unit. After receiving the breakdown low risk alarm, the monitoring and early warning unit generates a breakdown low risk reminder.

[0026] If the discharge statistics unit obtains both the position random signal and the discharge stable signal at the same time, it will not respond;

[0027] The discharge assessment unit sends the discharge exceeding standard signal to the monitoring and early warning unit. After receiving the discharge exceeding standard signal and the discharge continuous signal, the monitoring and early warning unit generates a partial discharge early warning. After acquiring the discharge exceeding standard signal and the discharge random signal, the discharge assessment unit generates an abnormal discharge early warning.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. In the present invention, after the partial discharge phenomenon of the transformer is obtained by an acoustic emission instrument, a diversified data analysis is performed on the partial discharge phenomenon, so as to obtain the discharge intensity, discharge fluctuation, discharge position movement and development and change trend of the partial discharge phenomenon of the transformer, thereby realizing comprehensive monitoring of the partial discharge phenomenon of the transformer and improving the monitoring effect of the partial discharge phenomenon.

[0030] 2. In the present invention, a combined analysis is performed based on the partial discharge intensity and discharge position intensity change phenomena of the transformer obtained through monitoring, and a low-risk breakdown alarm or a high-risk breakdown alarm is obtained based on the combined analysis results, and corresponding early warnings and automatic control measures are generated based on the low-risk breakdown alarm or the high-risk breakdown alarm, thereby ensuring the safe operation of the transformer. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0032] Figure 1 is a system block diagram of the present invention;

[0033] Figure 2 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0034] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] Embodiment 1:

[0036] See also Figure 1 - Figure 2 As shown, an online monitoring system for partial discharge of transformers based on artificial intelligence includes an information acquisition unit, a processing and generating unit, a discharge evaluation unit, a discharge statistics unit and a monitoring and early warning unit, wherein the information acquisition unit collects information from the transformer through an acoustic emitter, and the acoustic emitter has multiple groups of collection ports of the acoustic emitter distributed at different positions on the transformer, thereby constructing an acoustic wave collection matrix on the transformer, and the information acquisition unit sends the collected monitoring information to the processing and generating unit, and the monitoring information includes the acoustic wave waveform, the collection time and the collection port position;

[0037] After receiving the monitoring information, the processing and generating unit performs noise filtering and threshold processing on the acoustic waveform in the monitoring information, deletes the noise interference waveform and the waveform that does not reach the threshold value in the acoustic waveform, and obtains the discharge action waveform. The processing and generating unit processes the monitoring information obtained by all the acquisition ports to obtain multiple groups of discharge action waveforms. The processing and generating unit performs spatial algorithm simulation on the multiple groups of discharge action waveforms and the time when the acoustic waveform is collected to obtain the generation position of the discharge action waveform in three-dimensional space, which is recorded as the local discharge position. The processing and generating unit performs algorithm simulation based on the discharge waveform to obtain the local discharge intensity when the discharge waveform is generated. The processing and generating unit sends the local discharge position and the local discharge intensity to the discharge evaluation unit;

[0038] After the discharge evaluation unit obtains the local discharge position and the local discharge intensity, it compares the local discharge intensity with the preset discharge intensity. If the local discharge intensity is greater than or equal to the preset discharge intensity, a discharge exceeding standard signal is generated. If the local discharge intensity is less than the preset discharge intensity, a discharge not exceeding standard signal is generated.

[0039] After the discharge evaluation unit generates the discharge exceeding standard signal, the local discharge intensity that generates the discharge exceeding standard signal is recorded. When the discharge evaluation unit obtains the local discharge intensity again subsequently, the discharge evaluation unit performs difference calculation on the local discharge intensities obtained multiple times, and calculates the result of the difference calculation by using the variance formula to obtain the discharge intensity fluctuation. The discharge evaluation unit compares the discharge intensity fluctuation with the preset fluctuation value. If the discharge intensity fluctuation is greater than the preset fluctuation threshold, a discharge random signal is generated. If the discharge intensity fluctuation is less than or equal to the preset fluctuation threshold, a discharge continuation signal is generated.

[0040] After generating a discharge not exceeding standard signal, the discharge evaluation unit records the local discharge intensity that generates the discharge not exceeding standard signal, and records the new local discharge intensity in chronological order each time after obtaining the discharge not exceeding standard signal subsequently. The discharge evaluation unit selects from the first group of recorded local discharge intensities, selects the first group of local discharge intensities to the nth group of local discharge intensities, and calculates the first discharge mean, and then selects the n+1th group of local discharge intensities to the 2nth group of local discharge intensities, and calculates the second discharge mean. The discharge evaluation unit calculates multiple groups of discharge means, and calculates the difference between two adjacent groups of discharge means, and records it as a discharge change rate. The discharge evaluation unit compares the discharge change rate with a preset threshold value, and generates a discharge aggravation signal if the discharge change rate is greater than the preset threshold value, and generates a discharge stable signal if the discharge change rate is less than or equal to the preset threshold value.

[0041] The discharge evaluation unit collects statistics on the partial discharge positions obtained each time, compares the statistically obtained partial discharge positions, and generates a fixed position signal if the distance dispersion degree between the partial discharge positions obtained multiple times is within a preset range; if the distance dispersion degree between the partial discharge positions obtained multiple times is outside the preset range, generates a random position signal;

[0042] The method for the discharge evaluation unit to obtain the degree of distance dispersion is:

[0043] Step 1: The discharge assessment unit marks all partial discharge locations in the 3D model;

[0044] Step 2: Select any partial discharge position as the starting point, and connect lines from the starting point to all other partial discharge positions to obtain the total distance of the discharge positions;

[0045] Step 3: Repeat step 2 multiple times, so that each local discharge position is used as a starting point to obtain a total distance of the discharge position, and select the smallest group of total distances of all discharge positions;

[0046] Step 4: Divide the total distance of the selected discharge positions by the total number of all partial discharge positions to obtain the distance dispersion degree.

[0047] Embodiment 2:

[0048] See also Figure 1 - Figure 2 As shown, the discharge statistics unit does not respond after obtaining the discharge exceeding standard signal through the discharge evaluation unit. After the discharge statistics unit obtains the discharge not exceeding standard signal through the discharge evaluation unit:

[0049] If the discharge statistics unit obtains the position fixed signal and the discharge aggravation signal at the same time, a high-risk breakdown alarm is generated, and the high-risk breakdown alarm is sent to the monitoring and early warning unit. After receiving the high-risk breakdown alarm, the monitoring and early warning unit generates a high-risk breakdown reminder, emits an audible and visual alarm, and displays the transformer number and position corresponding to the high-risk breakdown alarm on the display device, and stops the operation of the transformer at the same time;

[0050] If the discharge statistics unit obtains the position fixed signal and the discharge stable signal at the same time, a breakdown low risk alarm is generated; if the discharge statistics unit obtains the position random signal and the discharge aggravation signal at the same time, a breakdown low risk alarm is generated. The discharge statistics unit sends the breakdown low risk alarm to the monitoring and early warning unit. After receiving the breakdown low risk alarm, the monitoring and early warning unit generates a breakdown low risk reminder and displays the transformer number and transformer position corresponding to the breakdown low risk reminder on the display screen;

[0051] If the discharge statistics unit obtains both the position random signal and the discharge stable signal at the same time, it will not respond;

[0052] The discharge assessment unit sends the discharge exceeding standard signal to the monitoring and early warning unit. After receiving the discharge exceeding standard signal and the discharge continuous signal, the monitoring and early warning unit generates a partial discharge early warning. After obtaining the discharge exceeding standard signal and the discharge random signal, the discharge assessment unit generates an abnormal discharge early warning, and sends the partial discharge early warning and the abnormal discharge early warning to the display device through the network, and issues an audible and visual alarm reminder at the same time.

[0053] The manager obtains the transformer number and location of the transformer that generated the alarm through the display device, making it easier for the manager to go for maintenance and repair.

[0054] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. The transformer partial discharge online monitoring system based on artificial intelligence is characterized by: It includes an information collection unit, and the information collection unit is used to collect monitoring information of the transformer; A processing and generating unit, wherein the processing and generating unit can analyze the monitoring information acquired by the information acquisition unit, and generate a partial discharge position and a partial discharge intensity according to the monitoring information; A discharge evaluation unit, wherein the discharge evaluation unit can perform a comprehensive analysis based on the local discharge position and the local discharge intensity, generate a discharge exceeding standard signal or a discharge not exceeding standard signal according to the local discharge intensity judgment result, analyze according to the local discharge intensity change factor, generate a discharge continuation signal, a discharge random signal, a discharge aggravation signal or a discharge stable signal, and analyze according to the local discharge position to generate a position fixed signal or a position random signal; a discharge statistics unit, wherein the discharge statistics unit obtains a discharge evaluation signal through the discharge evaluation unit, and performs combined analysis according to the discharge evaluation signal to generate different warning signals; The monitoring and early warning unit obtains the early warning signal and the discharge evaluation signal through the discharge statistics unit, and generates a corresponding alarm reminder.

2. The transformer partial discharge online monitoring system based on artificial intelligence according to claim 1 is characterized in that: When monitoring the transformer, the information collection unit arranges multiple groups of collection ports of acoustic emitters at different positions on the transformer, constructs an acoustic wave collection matrix on the transformer, and uses the collection matrix of the multiple groups of acoustic emitters to collect monitoring information of the transformer, wherein the monitoring information includes acoustic wave waveform, collection time and collection port position.

3. The transformer partial discharge online monitoring system based on artificial intelligence according to claim 1 is characterized in that: After receiving the monitoring information, the processing and generating unit deletes the noise interference waveform and the waveform that does not reach the threshold value in the sound wave waveform to obtain the discharge action waveform. The processing and generating unit processes the monitoring information obtained by all the acquisition ports to obtain multiple groups of discharge action waveforms. The processing and generating unit performs spatial algorithm simulation on the multiple groups of discharge action waveforms and the time when the sound wave waveform is collected to obtain the generation position of the discharge action waveform in three-dimensional space, which is recorded as the local discharge position. The processing and generating unit performs algorithm simulation based on the discharge waveform to obtain the local discharge intensity when the discharge waveform is generated. The processing and generating unit sends the local discharge position and the local discharge intensity to the discharge evaluation unit.

4. The transformer partial discharge online monitoring system based on artificial intelligence according to claim 1 is characterized in that: After obtaining the local discharge position and the local discharge intensity, the discharge evaluation unit compares the local discharge intensity with the preset discharge intensity. If the local discharge intensity is greater than or equal to the preset discharge intensity, a discharge exceeding standard signal is generated; if the local discharge intensity is less than the preset discharge intensity, a discharge not exceeding standard signal is generated.

5. The transformer partial discharge online monitoring system based on artificial intelligence according to claim 4 is characterized in that: After the discharge evaluation unit generates the discharge exceeding standard signal, the local discharge intensity that generates the discharge exceeding standard signal is recorded. When the discharge evaluation unit obtains the local discharge intensity again subsequently, the discharge evaluation unit performs difference calculation on the local discharge intensities obtained multiple times, and calculates the result of the difference calculation by using a variance formula to obtain the discharge intensity fluctuation. The discharge evaluation unit compares the discharge intensity fluctuation with a preset fluctuation value. If the discharge intensity fluctuation is greater than a preset fluctuation threshold, a discharge random signal is generated. If the discharge intensity fluctuation is less than or equal to the preset fluctuation threshold, a discharge continuation signal is generated. After generating the discharge not exceeding standard signal, the discharge evaluation unit records the local discharge intensity that generates the discharge not exceeding standard signal, and records the new local discharge intensity in chronological order each time after obtaining the discharge not exceeding standard signal subsequently. The discharge evaluation unit selects the n+1th group of local discharge intensity to the 2nth group of local discharge intensity, calculates the discharge mean, obtains multiple groups of discharge mean values, and calculates the difference between two adjacent groups of discharge mean values, and records it as the discharge change rate. The discharge evaluation unit compares the discharge change rate with a preset threshold value, generates a discharge aggravation signal if the discharge change rate is greater than the preset threshold value, and generates a discharge stable signal if the discharge change rate is less than or equal to the preset threshold value.

6. The transformer partial discharge online monitoring system based on artificial intelligence according to claim 1 is characterized in that: The discharge evaluation unit counts the partial discharge positions obtained each time, compares the partial discharge positions obtained by statistics, generates a fixed position signal if the distance dispersion degree between the partial discharge positions obtained multiple times is within a preset range, and generates a random position signal if the distance dispersion degree between the partial discharge positions obtained multiple times is outside the preset range.

7. The transformer partial discharge online monitoring system based on artificial intelligence according to claim 6 is characterized in that: The method for the discharge evaluation unit to obtain the distance dispersion degree is: Step 1: The discharge assessment unit marks all partial discharge locations in the 3D model; Step 2: Select any partial discharge position as the starting point, and connect lines from the starting point to all other partial discharge positions to obtain the total distance of the discharge positions; Step 3: Repeat step 2 multiple times, so that each local discharge position is used as a starting point to obtain a total distance of the discharge position, and select the smallest group of total distances of all discharge positions; Step 4: Divide the total distance of the selected discharge positions by the total number of all partial discharge positions to obtain the distance dispersion degree.

8. The transformer partial discharge online monitoring system based on artificial intelligence according to claim 1 is characterized in that: The discharge statistics unit does not respond after obtaining the discharge exceeding standard signal through the discharge evaluation unit. After the discharge statistics unit obtains the discharge not exceeding standard signal through the discharge evaluation unit: If the discharge statistics unit obtains the position fixed signal and the discharge aggravation signal at the same time, a breakdown high risk alarm is generated and sent to the monitoring and early warning unit; If the discharge statistics unit simultaneously obtains the position fixed signal and the discharge stable signal, a breakdown low risk alarm is generated; if the discharge statistics unit simultaneously obtains the position random signal and the discharge aggravation signal, a breakdown low risk alarm is generated. The discharge statistics unit sends the breakdown low risk alarm to the monitoring and early warning unit. After receiving the breakdown low risk alarm, the monitoring and early warning unit generates a breakdown low risk reminder. If the discharge statistics unit obtains both the position random signal and the discharge stable signal at the same time, it will not respond; The discharge assessment unit sends the discharge exceeding standard signal to the monitoring and early warning unit. After receiving the discharge exceeding standard signal and the discharge continuous signal, the monitoring and early warning unit generates a partial discharge early warning. After acquiring the discharge exceeding standard signal and the discharge random signal, the discharge assessment unit generates an abnormal discharge early warning.

Images