Discharge detection method and system of power distribution cabinet based on artificial intelligence

Abstract

The invention relates to the field of discharge detection, is used for solving the problems that a discharge detection system cannot perform multivariate data acquisition on a discharge phenomenon to improve accuracy and lacks deep analysis on a partial discharge phenomenon, and particularly relates to a discharge detection method and system of a power distribution cabinet based on artificial intelligence. Comprising a basic acquisition unit, a waveform confirmation unit, a multidirectional data fusion unit, a multidirectional analysis unit, an evaluation generation unit and an acoustic emission rechecking unit, through multiple acquisition modes of pulse current and electromagnetic waves, the purpose of improving the accuracy degree of monitoring the partial discharge phenomenon is achieved, the phenomenon of misjudgment caused by background interference is avoided, meanwhile, fusion analysis of multiple data is carried out on the partial discharge phenomenon, and the accuracy of monitoring the partial discharge phenomenon is improved. And the occurrence frequency of the partial discharge phenomenon is determined and the damage degree of the partial discharge phenomenon to hardware facilities in the power distribution cabinet is predicted, so that the data monitoring result of the partial discharge phenomenon is deepened and the maintenance period of the power distribution cabinet is conveniently determined.

Description

Technical Field

[0001] The present invention relates to the field of discharge detection, and in particular to a discharge detection method and system for a power distribution cabinet based on artificial intelligence. Background Art

[0002] The main causes of partial discharge in distribution cabinets include uneven electric field distribution, aging of insulation materials, and poor electrode shape. Partial discharge refers to the discharge phenomenon that occurs in a local area of the insulator under the action of an electric field. Although the entire insulator does not undergo penetrating discharge, long-term accumulation will seriously affect the insulation performance.

[0003] The danger of partial discharge is that it will gradually erode the insulation material, leading to the overall failure of the insulation system. To detect and monitor partial discharge, a variety of methods can be used, including pulse current detection, ultra-high frequency current detection, and ultrasonic detection. The pulse current method obtains discharge information by measuring the pulse current caused by the discharge, while the ultra-high frequency current detection method uses the electromagnetic waves generated by the discharge for detection.

[0004] Currently, existing patent CN118731609A discloses a technical solution that achieves continuous online monitoring of the distribution cabinet by setting a detection position in the distribution cabinet and installing sensors for detection, avoiding the disadvantages of manual handheld device detection. However, this solution only detects partial discharge based on the partial discharge current and does not automatically analyze and preprocess the pre-detected data. As a result, the detection basis is relatively simple and prone to interference and misjudgment. At the same time, it cannot automatically generate frequency analysis results of the partial discharge phenomenon, nor can it estimate and verify the impact of the partial discharge phenomenon.

[0005] In response to the above technical problems, this application proposes a solution. Summary of the Invention

[0006] The present invention achieves the purpose of improving the accuracy of monitoring partial discharge phenomena through multiple collection methods of pulse current and electromagnetic waves, avoiding the phenomenon of misjudgment due to background interference. At the same time, the partial discharge phenomenon is subjected to a fusion analysis of multiple data to confirm the occurrence frequency of the partial discharge phenomenon and predict the degree of damage caused by the partial discharge phenomenon to the hardware facilities in the distribution cabinet, thereby deepening the data monitoring results of the partial discharge phenomenon, facilitating the determination of the maintenance cycle of the distribution cabinet, solving the problems that the discharge detection system cannot perform multivariate data collection on the discharge phenomenon to improve the accuracy and lacks in-depth analysis of the partial discharge phenomenon, and proposes a discharge detection method and system for the distribution cabinet based on artificial intelligence.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A method for detecting discharge of a power distribution cabinet based on artificial intelligence comprises the following steps:

[0009] Step 1: Monitor the pulse current and electromagnetic waves of partial discharge;

[0010] Step 2: Compare the waveforms of the acquired pulse current and electromagnetic wave to confirm the discharge data;

[0011] Step 3: Conduct multi-directional analysis of discharge data;

[0012] Step 4: Evaluate the analysis results;

[0013] Step 5: Review the acoustic emission based on the assessment results.

[0014] As a preferred embodiment of the present invention, in step 3, when performing multidirectional analysis, the discharge data and background data are jointly analyzed;

[0015] The background data is collected at the same time as the pulse current or electromagnetic wave is monitored in step one.

[0016] An artificial intelligence-based discharge detection system for a power distribution cabinet includes a basic acquisition unit for continuously monitoring pulse current and electromagnetic waves of partial discharge in the power distribution cabinet;

[0017] a waveform confirmation unit, which compares the current waveform and electromagnetic wave waveform of the pulse current acquired by the basic acquisition unit with a set waveform diagram, and generates a partial discharge signal or an interference signal according to the comparison result;

[0018] a multi-directional data fusion unit, which obtains the monitored partial discharge signal through the waveform confirmation unit, immediately collects other data when the partial discharge signal is detected, and fuses the data into background data;

[0019] a multi-directional analysis unit, which performs multi-directional analysis on the background data and the partial discharge signal to confirm the occurrence frequency of the partial discharge signal and the expected damage accumulation amount;

[0020] an evaluation generating unit, the evaluation generating unit obtaining a partial discharge evaluation result based on the occurrence frequency of the partial discharge signal and the estimated damage accumulation amount;

[0021] An acoustic emission review unit is configured to obtain a partial discharge evaluation result, perform an acoustic wave review on the insulation material based on the partial discharge evaluation result, confirm the damage state of the insulation material, and generate a maintenance signal.

[0022] As a preferred embodiment of the present invention, the basic acquisition unit collects the pulse current of the partial discharge through a mutual inductance pulse current sensor arranged in the power distribution cabinet;

[0023] The basic collection unit collects the electromagnetic waves of partial discharge by using the ultra-high frequency detection principle.

[0024] As a preferred embodiment of the present invention, after obtaining the current waveform of the pulse current, the waveform confirmation unit compares the current waveform with a set current waveform database, obtains the similarity between the current waveform and the current sample waveform in the database, performs threshold analysis based on the similarity, and generates a current confirmation signal or a current rejection signal;

[0025] The waveform confirmation unit compares the electromagnetic wave waveform with the electromagnetic wave waveform in the database, obtains the similarity between the electromagnetic wave waveform and the electromagnetic wave sample waveform in the database, performs threshold analysis based on the similarity, and generates an electromagnetic confirmation signal or an electromagnetic rejection signal.

[0026] As a preferred embodiment of the present invention, the waveform confirmation unit generates a local discharge signal after simultaneously acquiring the current confirmation signal and the electromagnetic confirmation signal, generates a suspected discharge signal when only one of the current confirmation signal or the electromagnetic confirmation signal is acquired, and generates an interference signal when neither the current confirmation signal nor the electromagnetic confirmation signal is acquired.

[0027] As a preferred embodiment of the present invention, when the multivariate data fusion unit obtains the partial discharge signal, it immediately records the time information;

[0028] The multivariate data fusion unit continuously collects and records the distribution cabinet environmental information, wherein the distribution cabinet environmental information includes the humidity inside the cabinet and background discharge information. The multivariate data fusion unit obtains the environmental information at the same time from the recorded environmental information according to the time information, and records the environmental information and time information as background data.

[0029] As a preferred embodiment of the present invention, the multi-directional analysis unit calculates the frequency of the partial discharge information through time information;

[0030] The multi-directional analysis unit analyzes the environmental information and the partial discharge frequency through a formula to obtain an estimated cumulative amount of damage.

[0031] As a preferred embodiment of the present invention, the evaluation generation unit performs a threshold analysis on the estimated damage accumulation amount and obtains a material damage signal or a material normal signal according to the analysis result;

[0032] The evaluation and generation unit performs a threshold analysis on the occurrence frequency of the partial discharge signal and generates a high-frequency discharge signal or a low-frequency discharge signal according to the analysis result.

[0033] As a preferred embodiment of the present invention, after the acoustic emission verification unit obtains the material damage signal and the high-frequency discharge signal, it performs flaw detection verification on the actual damage condition of the material through the acoustic emission technology.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. In the present invention, the partial discharge phenomenon in the distribution cabinet is confirmed through multiple collection methods of pulse current and electromagnetic waves, thereby improving the accuracy of partial discharge monitoring and avoiding the phenomenon of misjudgment due to background interference. At the same time, the partial discharge phenomenon is subjected to a fusion analysis of multiple data to confirm the frequency of partial discharge phenomenon and predict the degree of damage caused by partial discharge phenomenon to the hardware facilities in the distribution cabinet, thereby increasing the in-depth analysis of the partial discharge phenomenon.

[0036] 2. In the present invention, the insulation materials and cables in the distribution cabinet are monitored for damage through acoustic emission technology to confirm the damage degree of the insulation cables, and provide a basis for the maintenance and repair of the distribution cabinet according to the damage degree, thereby realizing the utilization of data after in-depth analysis of the partial discharge phenomenon and ensuring the detection effect of the overall discharge phenomenon of the distribution cabinet. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0038] Figure 1 is a system block diagram of the present invention;

[0039] Figure 2 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0040] The following is a clear and complete description of the technical solutions of the present invention in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0041] Example 1:

[0042] See also Figure 1 - Figure 2 As shown, a discharge detection method for a distribution cabinet based on artificial intelligence includes the following steps:

[0043] Step 1: Collect data on the pulse current of the partial discharge through a pulse current sensor, and collect data on the electromagnetic waves generated by the partial discharge phenomenon through an electromagnetic wave sensor to obtain waveform data of the pulse current and electromagnetic waves;

[0044] Step 2: Compare the acquired waveform data of the pulse current and electromagnetic wave with the corresponding sample waveforms in the database, confirm whether the pulse current and electromagnetic wave are caused by partial discharge based on the waveform comparison results, and generate a partial discharge signal or interference signal based on the confirmation results;

[0045] Step 3: When acquiring pulse current and electromagnetic waves, the time information is recorded and environmental information is acquired based on the time information. The environmental information and time information when the partial discharge signal is generated are analyzed to determine the frequency of partial discharge and the cumulative damage caused by the partial discharge phenomenon to the insulation material.

[0046] Step 4: Determine the normality of the circuit system in the distribution cabinet based on the frequency of partial discharge phenomena. Determine the working condition of the hardware facilities of the distribution cabinet based on the cumulative damage caused by the insulation materials to confirm whether the distribution cabinet can operate safely and stably.

[0047] Step 5: Based on the judgment results of the hardware facilities of the distribution cabinet, the hardware facilities of the distribution cabinet are re-checked through acoustic emission technology to confirm whether the hardware damage of the distribution cabinet is the same as the predicted results, and make a judgment on the maintenance and replacement of the hardware facilities in the distribution cabinet.

[0048] Example 2:

[0049] See also Figure 1 - Figure 2 As shown, a discharge detection system for a power distribution cabinet based on artificial intelligence includes a basic acquisition unit, a waveform confirmation unit, a multi-directional data fusion unit, a multi-directional analysis unit, an evaluation generation unit, and an acoustic emission review unit;

[0050] The basic acquisition unit is used to continuously monitor the pulse current and electromagnetic waves of partial discharge in the distribution cabinet. When collecting the pulse current of partial discharge, the basic acquisition unit collects the data through the mutual inductance pulse current sensor installed in the distribution cabinet.

[0051] The basic acquisition unit collects electromagnetic waves from partial discharge using the ultra-high frequency (UHF) detection principle. Partial discharge generates electromagnetic waves with frequencies up to several GHz. UHF sensors can receive these high-frequency electromagnetic wave signals, thereby detecting partial discharge phenomena through electromagnetic waves.

[0052] After obtaining the current waveform of the pulse current, the waveform confirmation unit compares the current waveform with a set current waveform database to obtain the similarity between the current waveform and the current sample waveform in the database. If the similarity is greater than a set threshold, a current confirmation signal is generated; if the similarity is not greater than the set threshold, a current rejection signal is generated.

[0053] The waveform confirmation unit compares the electromagnetic wave waveform with the electromagnetic wave waveform in the database to obtain the similarity between the electromagnetic wave waveform and the electromagnetic wave sample waveform in the database. If the similarity is greater than a set threshold, an electromagnetic confirmation signal is generated; if the similarity is not greater than the set threshold, an electromagnetic rejection signal is generated.

[0054] The waveform confirmation unit generates a partial discharge signal after simultaneously acquiring the current confirmation signal and the electromagnetic confirmation signal. When only one of the current confirmation signal or the electromagnetic confirmation signal is acquired, a suspected discharge signal is generated. When neither the current confirmation signal nor the electromagnetic confirmation signal is acquired, an interference signal is generated.

[0055] Example 3:

[0056] See also Figure 1 - Figure 2 As shown, when the multivariate data fusion unit obtains the partial discharge signal, it immediately records the time information;

[0057] The multivariate data fusion unit continuously collects and records the distribution cabinet environmental information, including the humidity inside the cabinet and background discharge information. The multivariate data fusion unit obtains the environmental information at the same time from the recorded environmental information based on the time information, and records the environmental information and time information as background data.

[0058] The multi-directional analysis unit calculates the frequency of the partial discharge information through the time information to obtain the partial discharge frequency;

[0059] The multi-directional analysis unit analyzes the environmental information and the partial discharge frequency through a formula to obtain the estimated cumulative damage QF. Where f is the partial discharge frequency, RH is the humidity inside the cabinet, QB is the background discharge intensity, θ and β are the weighting coefficients of the partial discharge frequency and background discharge intensity, respectively. The humidity inside the cabinet has a positive correlation with the damage to the insulation material. The background discharge intensity interferes with partial discharge, slightly increasing the actual detected partial discharge phenomenon, which will cause the estimated damage amount to be biased. Therefore, the background discharge intensity is inversely proportional to the estimated cumulative damage amount. θ and β are used to weight the background discharge intensity and the humidity inside the cabinet to balance the numerical magnitudes of the humidity inside the cabinet and the background discharge intensity. t is the continuous operation time of the distribution cabinet after maintenance. The above formulas are all dimensionless calculations, and only the numerical part is used for reference.

[0060] The evaluation generation unit performs a threshold analysis on the expected damage accumulation. If the expected damage accumulation is greater than the set threshold, a material damage signal is generated. If the expected damage accumulation is not greater than the set threshold, a material normal signal is generated.

[0061] The evaluation and generation unit performs a threshold analysis on the occurrence frequency of the partial discharge signal. If the occurrence frequency of the partial discharge signal is greater than the set threshold, a high-frequency discharge signal is generated. If the frequency of the partial discharge signal is not greater than the set threshold, a low-frequency discharge signal is generated.

[0062] After the acoustic emission review unit obtains the material damage signal and high-frequency discharge signal, it uses acoustic emission technology to detect the actual damage of the material. Based on the detection results, it determines whether the material is damaged to confirm the prediction results of the evaluation generation unit. When it is confirmed that the material is damaged, an alarm reminder is generated to facilitate staff to maintain and repair the distribution cabinet.

[0063] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for detecting discharge of a power distribution cabinet based on artificial intelligence, characterized in that: The following steps are involved: Step 1: Monitor the pulse current and electromagnetic waves of partial discharge; Step 2: Compare the waveforms of the acquired pulse current and electromagnetic wave to confirm the discharge data; Step 3: Conduct multi-directional analysis of discharge data; Step 4: Evaluate the analysis results; Step 5: Conduct acoustic emission review based on the assessment results.

2. The method for detecting discharge of a power distribution cabinet based on artificial intelligence according to claim 1, characterized in that: In the step 3, when performing multidirectional analysis, the discharge data and the background data are jointly analyzed; The background data is collected at the same time as the pulse current or electromagnetic wave is monitored in step one.

3. A discharge detection system for a power distribution cabinet based on artificial intelligence, applicable to the discharge detection method for a power distribution cabinet based on artificial intelligence according to claim 1, characterized in that: It includes a basic acquisition unit, which is used to continuously monitor the pulse current and electromagnetic waves of partial discharge in the distribution cabinet; a waveform confirmation unit, which compares the current waveform and electromagnetic wave waveform of the pulse current acquired by the basic acquisition unit with a set waveform diagram, and generates a partial discharge signal or an interference signal according to the comparison result; a multi-directional data fusion unit, which obtains the monitored partial discharge signal through the waveform confirmation unit, immediately collects other data when the partial discharge signal is detected, and fuses the data into background data; a multi-directional analysis unit, which performs multi-directional analysis on the background data and the partial discharge signal to confirm the occurrence frequency of the partial discharge signal and the expected damage accumulation amount; an evaluation generating unit, the evaluation generating unit obtaining a partial discharge evaluation result based on the occurrence frequency of the partial discharge signal and the estimated damage accumulation amount; An acoustic emission review unit is configured to obtain a partial discharge evaluation result, perform an acoustic wave review on the insulation material based on the partial discharge evaluation result, confirm the damage state of the insulation material, and generate a maintenance signal.

4. The artificial intelligence-based discharge detection system for a power distribution cabinet according to claim 3, characterized in that: When the basic acquisition unit acquires the pulse current of the partial discharge, it acquires the pulse current through the mutual inductance pulse current sensor arranged in the power distribution cabinet; The basic collection unit collects the electromagnetic waves of partial discharge by using the ultra-high frequency detection principle.

5. The artificial intelligence-based discharge detection system for a power distribution cabinet according to claim 3, characterized in that: After obtaining the current waveform of the pulse current, the waveform confirmation unit compares the current waveform with a set current waveform database, obtains the similarity between the current waveform and the current sample waveform in the database, performs threshold analysis based on the similarity, and generates a current confirmation signal or a current rejection signal; The waveform confirmation unit compares the electromagnetic wave waveform with the electromagnetic wave waveform in the database, obtains the similarity between the electromagnetic wave waveform and the electromagnetic wave sample waveform in the database, performs threshold analysis based on the similarity, and generates an electromagnetic confirmation signal or an electromagnetic rejection signal.

6. The artificial intelligence-based discharge detection system for a power distribution cabinet according to claim 5, characterized in that: The waveform confirmation unit generates a partial discharge signal after simultaneously acquiring the current confirmation signal and the electromagnetic confirmation signal, generates a suspected discharge signal when only one of the current confirmation signal or the electromagnetic confirmation signal is acquired, and generates an interference signal when neither the current confirmation signal nor the electromagnetic confirmation signal is acquired.

7. The artificial intelligence-based discharge detection system for a power distribution cabinet according to claim 3, characterized in that: When the multivariate data fusion unit obtains the partial discharge signal, the time information is immediately recorded; The multivariate data fusion unit continuously collects and records the distribution cabinet environmental information, wherein the distribution cabinet environmental information includes the humidity inside the cabinet and background discharge information. The multivariate data fusion unit obtains the environmental information at the same time from the recorded environmental information according to the time information, and records the environmental information and time information as background data.

8. The artificial intelligence-based discharge detection system for a power distribution cabinet according to claim 7, characterized in that: The multi-directional analysis unit calculates the frequency of the partial discharge information through time information; The multi-directional analysis unit analyzes the environmental information and the partial discharge frequency through a formula to obtain an estimated cumulative amount of damage.

9. The artificial intelligence-based discharge detection system for a power distribution cabinet according to claim 7, characterized in that: The evaluation generation unit performs a threshold analysis on the estimated damage accumulation amount and obtains a material damage signal or a material normal signal according to the analysis result; The evaluation and generation unit performs a threshold analysis on the occurrence frequency of the partial discharge signal and generates a high-frequency discharge signal or a low-frequency discharge signal according to the analysis result.

10. The artificial intelligence-based discharge detection system for a power distribution cabinet according to claim 7, characterized in that: After the acoustic emission verification unit obtains the material damage signal and the high-frequency discharge signal, it performs flaw detection verification on the actual damage condition of the material through the acoustic emission technology.

Images