Method for detecting partial discharge of direct-current gas insulated switchgear (GIS)

Abstract

The invention discloses a direct-current GIS (gas insulated switchgear) partial discharge detection method, which comprises the following steps: acquiring multi-modal sensing data in a direct-current GIS, the multi-modal sensing data comprising any two or more of current sensor data, optical sensor data, ultrasonic sensor data, infrared sensor data and ultrahigh-frequency sensor data; the collected multi-mode sensing data is preprocessed; performing feature extraction on the preprocessed multi-modal sensing data by using a deep learning model; and synthesizing the extracted features by using data fusion to obtain partial discharge information consistency explanation of the multi-modal sensing data. According to the method, multiple partial discharge detection methods can be comprehensively configured according to the requirements of the operation condition of the direct current GIS, so that the partial discharge condition of the direct current GIS can be more effectively evaluated.

Description

Technical Field

[0001] The present application relates to the technical field of electric power equipment, and in particular to a method for detecting partial discharge of a direct current gas insulated switch GIS. Background Art

[0002] DC gas insulated switchgear (GIS) is increasingly being used for long-distance and large-capacity power transmission due to its advantages such as small footprint and high insulation reliability. Partial discharge (PD) is a common electrical phenomenon in GIS equipment. It not only indicates the insulation status of GIS equipment, but is also often a precursor to insulation failure of GIS equipment. Therefore, accurate monitoring of partial discharge activities is essential to ensure the stable operation of power systems. Traditional partial discharge detection technologies, such as electrical measurements in the IEC60270 standard, are often difficult to adapt to the detection needs of DC GIS, because the lack of AC phase relationship under DC conditions makes it impossible to use phase-resolved PD detection mode.

[0003] During the partial discharge process, electric pulses, electromagnetic waves, ultrasound, light, local overheating and some new chemical products will be generated. Accordingly, electrical detection methods, acoustic detection methods, optical detection methods and chemical detection methods have emerged. Among them, the pulse current method mainly measures the magnitude of the current by building a circuit through the device when partial discharge occurs inside the GIS equipment, which is accompanied by a voltage increase. The range is within 1MHz, but it is easily mixed with other signals and is difficult to identify due to interference; the ultrasonic detection method mainly uses the non-electrical variable generated when partial discharge occurs inside the GIS, that is, the sound wave, and its frequency band is mainly 20KHz~100KHz. Ultrasonic sensors can be installed outside the equipment cavity for detection because it is a non-electrical detection method, which is not affected by electrical signals and has strong anti-electromagnetic interference performance; the optical detection method mainly detects optical radiation signals, and the optical sensor needs to be installed inside the equipment. The optical signal is also a non-electrical variable with good anti-electromagnetic interference ability, but when the inner wall of the equipment is smooth, the reflection effect or the existence of shielding will also cause interference, which can easily lead to large differences in measurement results. Since the above detection methods all have certain limitations and disadvantages, it is difficult to achieve accurate evaluation of the partial discharge situation of the DC GIS. Therefore, the present invention proposes a DC gas insulated switch GIS partial discharge detection method. Summary of the invention

[0004] The embodiment of the present application provides a DC gas insulated switch GIS partial discharge detection method, which enables comprehensive configuration of multiple partial discharge detection methods according to the requirements of the DC GIS operating conditions, thereby more effectively evaluating the partial discharge situation of the DC GIS.

[0005] The present application provides a method for detecting partial discharge in a DC gas insulated switch GIS, which specifically includes the following steps: acquiring multimodal sensor data in the DC GIS;

[0006] Preprocessing the collected multimodal sensor data;

[0007] Using a deep learning model to extract features from the preprocessed multimodal sensor data;

[0008] The extracted features are synthesized by using data fusion to obtain a consistent interpretation of the partial discharge information of the multimodal sensing data.

[0009] Optionally, the preprocessing includes performing signal filtering processing on the initial multimodal sensing data.

[0010] Optionally, the deep learning model is constructed based on a convolutional neural network or a recurrent neural network, and is used to extract local discharge information based on the preprocessed multimodal sensing data to preliminarily determine the type and location of local discharge.

[0011] Optionally, the fusion rule of the data fusion adopts a maximum fusion method, an average fusion method and a Bayesian fusion method.

[0012] Optionally, the multimodal sensing data includes any two or more of current sensor data, optical sensor data, ultrasonic sensor data, infrared sensor data and ultra-high frequency sensor data.

[0013] Optionally, the local discharge information includes discharge amount, discharge position, discharge number and discharge time interval.

[0014] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages: the present application obtains multimodal sensor data in DC GIS and pre-processes the collected multimodal sensor data; uses a deep learning model to extract features from the pre-processed multimodal sensor data; uses data fusion to synthesize the extracted features to obtain a consistent interpretation of the partial discharge information of the multimodal sensor data. It can comprehensively configure a variety of partial discharge detection methods according to the requirements of the DC GIS operating conditions, and thus can more effectively evaluate the partial discharge situation of the DC GIS, greatly improving the accuracy of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a method for detecting partial discharge of a DC gas insulated switch GIS in an embodiment of the present application;

[0016] Figure 2 This is a flow chart of the data fusion method in the embodiment of the present application;

[0017] Figure 3 It is a schematic diagram of a partial discharge detection device for a DC gas insulated switch GIS in an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0019] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0020] Unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0021] This application provides an embodiment of a method for detecting partial discharge of a DC gas insulated switch GIS. Figure 1 .

[0022] The DC gas insulated switch GIS partial discharge detection method in this embodiment includes: acquiring multimodal sensor data in the DC GIS and preprocessing the collected multimodal sensor data; using a deep learning model to extract features from the preprocessed multimodal sensor data; using data fusion to synthesize the extracted features to obtain a consistent interpretation of the partial discharge information of the multimodal sensor data.

[0023] It should be noted that: the present application can comprehensively configure a variety of partial discharge detection methods according to the requirements of the DC GIS operating conditions, thereby more effectively evaluating the partial discharge situation of the DC GIS and greatly improving the accuracy of detection.

[0024] The above is a first embodiment of a method for detecting partial discharge of a DC gas insulated switch GIS provided in an embodiment of the present application. The following is a second embodiment of a method for detecting partial discharge of a DC gas insulated switch GIS provided in an embodiment of the present application. For details, please refer to Figure 1 and Figure 2 .

[0025] The DC gas insulated switch GIS partial discharge detection method in this embodiment includes the following steps: acquiring multimodal sensor data in the DC GIS and preprocessing the collected multimodal sensor data; using a deep learning model to extract features from the preprocessed multimodal sensor data; using data fusion to synthesize the extracted features to obtain a consistent interpretation of the partial discharge information of the multimodal sensor data.

[0026] The multimodal sensing data includes any two or more of current sensor data, optical sensor data, ultrasonic sensor data, infrared sensor data, and ultra-high frequency sensor data.

[0027] The preprocessing includes signal filtering of the initial multimodal sensing data to remove interference outside the local discharge frequency band.

[0028] The deep learning model is built based on a convolutional neural network or a recurrent neural network (which is a proprietary deep learning model for different types of sensor data). It is used to extract partial discharge information based on the preprocessed multimodal sensor data to preliminarily determine the type and location of partial discharge.

[0029] Data fusion fuses the output results of different types of proprietary deep learning models in the late decision-making layer. Its fusion rules use maximum fusion method, average fusion method and Bayesian fusion method to accurately identify the type of local discharge and locate the local discharge position by obtaining the local discharge time difference based on sensors at different positions.

[0030] like Figure 2 As shown in the figure, the specific method of data fusion is: first, based on historical data, determine the prior probability of specific faults occurring in specific components of the DC GIS and the reliability of different sensors; based on the reliability data of the sensors, calculate the likelihood probability of specific sensors giving specific detection results in different states and partial discharge modes of the DC GIS; calculate the marginal probability of at least one sensor reporting a certain type of partial discharge; calculate the posterior probability of the DC GIS experiencing a certain type of partial discharge when the sensor reports the occurrence of a certain type of partial discharge; perform the above operations on different types of sensors respectively, and make reasonable estimates based on the calculation results.

[0031] Partial discharge information includes discharge amount, discharge location, discharge number and discharge time interval.

[0032] This application also provides a DC gas insulated switch GIS partial discharge detection device. Figure 3 , which includes a multimodal sensing module for collecting multimodal sensing data at different locations of a DC GIS; a data processing unit for receiving multimodal sensing data, filtering data and performing feature extraction and analysis; a diagnostic unit for outputting diagnostic information based on the features of the multimodal sensing data; a display module for displaying diagnostic information and warnings of partial discharge events; the data processing unit includes a filtering module and a deep learning module, wherein the filtering module is used to preprocess the collected information and filter out interference outside the partial discharge frequency band; the deep learning module uses a deep learning model to extract features from the collected multimodal sensing data; the diagnostic unit includes a data fusion module and a decision module, wherein the data fusion module is used to synthesize the extracted features using data fusion technology; the decision module obtains a consistent interpretation of the partial discharge information of the multimodal sensing data and outputs diagnostic information; the multimodal sensing module is communicatively connected to the filtering module, the filtering module is electrically connected to the deep learning module, the deep learning module is electrically connected to the data fusion module, the data fusion module is electrically connected to the decision module, and the decision module is electrically connected to the display module.

[0033] The multimodal sensing module includes a current sensor, an optical sensor, an ultrasonic sensor, an infrared sensor and an ultra-high frequency sensor to capture electric pulses, electromagnetic waves, ultrasound, light and local overheating generated by local discharge; the current sensor, optical sensor, ultrasonic sensor, infrared sensor and ultra-high frequency sensor are respectively arranged at different positions of the DC GIS, and the current sensor, optical sensor, ultrasonic sensor, infrared sensor and ultra-high frequency sensor are all communicatively connected with the filtering module.

[0034] It should be noted that current sensors, optical sensors, ultrasonic sensors, infrared sensors and UHF sensors detect DC GIS partial discharge information in real time, and transmit the information to the data processing unit for processing, extract partial discharge characteristics, and preliminarily determine the type of partial discharge by the deep learning model. The partial discharge information determined by different types of sensor data is summarized in the data fusion module for late fusion in the decision layer. The fused data is finally processed by the decision module to determine the type of partial discharge, realize partial discharge positioning, and finally output the judgment result.

[0035] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting partial discharge of a DC gas insulated switch GIS, characterized in that: The following steps are involved: Acquire multimodal sensor data in DC GIS; Preprocessing the collected multimodal sensor data; Using a deep learning model to extract features from the preprocessed multimodal sensor data; The extracted features are synthesized by using data fusion to obtain a consistent interpretation of the partial discharge information of the multimodal sensing data.

2. The method for detecting partial discharge of a DC gas insulated switch GIS according to claim 1, characterized in that: The preprocessing includes performing signal filtering processing on the initial multimodal sensing data.

3. The method for detecting partial discharge of a DC gas insulated switch GIS according to claim 1, characterized in that: The deep learning model is constructed based on a convolutional neural network or a recurrent neural network, and is used to extract local discharge information based on the preprocessed multimodal sensing data to preliminarily determine the type and location of local discharge.

4. The method for detecting partial discharge of a DC gas insulated switch GIS according to claim 1, characterized in that: The fusion rules of the data fusion adopt the maximum fusion method, the average fusion method and the Bayesian fusion method.

5. The method for detecting partial discharge of a DC gas insulated switch GIS according to claim 1, characterized in that: The multimodal sensing data includes any two or more of current sensor data, optical sensor data, ultrasonic sensor data, infrared sensor data and ultra-high frequency sensor data.

6. The method for detecting partial discharge of a DC gas insulated switch GIS according to claim 1, characterized in that: The partial discharge information includes discharge amount, discharge position, discharge times and discharge time interval.

Images