Multi-dimensional perception monitoring system and method based on GIL

Abstract

The invention discloses a GIL-based multi-dimensional sensing monitoring system and method, and relates to the technical field of multi-dimensional monitoring, and the system comprises a multi-dimensional sensing module, an information transmission module, a data analysis module and a fault judgment early warning module. The multi-dimensional sensing module is used for collecting different types of to-be-monitored information through sensors with different functions; the information transmission module performs information transmission in a limited transmission mode and a wireless transmission mode; the data analysis module is used for analyzing information data acquired by the sensor; the fault judgment and early warning module is used for displaying a monitoring result and early warning information; according to the multi-dimensional monitoring system and method designed according to the characteristics of the GIL system, and a flexible information transmission layout method is introduced, so that the generalization ability of the monitoring system is improved; and the efficiency and the accuracy of the monitoring model are improved through single data monitoring, analysis and processing of fault risk values and introduction of data accuracy detection and data missing detection.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-dimensional monitoring, and specifically to a multi-dimensional perception monitoring system and method based on GIL. Background Art

[0002] Gas-insulated metal-enclosed transmission line (GIL) mainly relies on insulating gas to achieve electrical insulation, uses aluminum or copper materials with high electrical conductivity for conduction, and the most commonly used insulating gas is sulfur hexafluoride, which has excellent insulation performance; in recent years, GIL technology has been continuously innovated and developed, and GIL has developed towards higher voltage levels and larger transmission capacities; moreover, the third-generation GIL technology with compressed air as the insulating medium is also under research and exploration; GIL technology can be used in power stations and outgoing lines, urban power grids, and power transmission in special geographical locations, promoting infrastructure construction and industrial development; however, during the operation of GIL, it is necessary to ensure data such as internal temperature, humidity, and pressure, and to monitor in real time whether there is partial discharge in the humidity inside GIL, and to give early warnings and handle the detected abnormal situations in a timely manner. Because the number of indicators to be monitored is relatively large, a multi-dimensional perception monitoring technology is used to monitor the operating GIL.

[0003] The multi-dimensional perception monitoring system comprehensively uses a variety of sensors and advanced technologies to monitor and analyze information in multiple dimensions in real time and accurately; the functions of early monitoring systems were relatively single, mainly monitoring a specific physical quantity, such as temperature, pressure, etc., and the sensor technologies used were relatively simple, and the data processing and analysis capabilities were limited; with the continuous progress of sensor technology, multi-dimensional perception monitoring systems that can monitor multiple physical quantities simultaneously gradually emerged; different types of sensors were integrated together to achieve comprehensive monitoring of information in various aspects such as the environment and equipment; for example, in the field of meteorological monitoring, a variety of sensors such as temperature, humidity, wind speed, and air pressure began to be comprehensively used for comprehensive monitoring of meteorological elements; in recent years, with the rapid development of technologies such as the Internet of Things, big data, and artificial intelligence, the multi-dimensional perception monitoring system has entered the stage of intelligence and networking; the system can not only collect and transmit multi-dimensional data in real time, but also use advanced algorithms to deeply analyze and mine the data to achieve intelligent decision-making and early warning; however, the existing multi-dimensional monitoring systems have insufficient adaptability and pertinence and are too dependent on specific scenarios, so it is necessary to design a multi-dimensional monitoring system and method for data collection, analysis, and processing according to the operating characteristics of GIL. Summary of the Invention

[0004] The purpose of the present invention is to provide a multi-dimensional perception monitoring system and method based on GIL to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solutions: a multi-dimensional perception monitoring system based on GIL, the system includes a multi-dimensional sensing module, an information transmission module, a data analysis module, and a fault warning module;

[0006] The multi-dimensional sensing module collects five data information of temperature, humidity, pressure, vibration intensity, and partial discharge intensity inside the GIL through sensors;

[0007] The information transmission module uses two methods of wired transmission and wireless transmission for information transmission;

[0008] The data analysis module is used to process the data through preprocessing, data accuracy detection, and data missing detection, and analyze the data through a comparative analysis method to determine whether the GIL and the data acquisition device have failed;

[0009] The fault warning module is used to display the monitoring results and give an audible and visual warning when a fault is detected.

[0010] Furthermore, the multi-dimensional sensing module includes a temperature sensing unit, a pressure sensing unit, a vibration sensing unit, a humidity sensing unit, and a partial discharge sensing unit; the temperature sensing unit collects the internal temperature information of the GIL during operation in real time through a fiber Bragg grating temperature sensor; the pressure sensing unit collects the gas pressure information in real time by installing a pressure sensor in the GIL gas chamber; the vibration sensing unit monitors the vibration intensity generated inside the GIL by installing a vibration sensor; the humidity sensing unit monitors the gas humidity inside the GIL in real time through a humidity sensor; the partial discharge sensing unit is used to collect the position and intensity information of partial discharge inside the GIL;

[0011] Furthermore, the information transmission module includes a wired transmission unit and a wireless transmission unit; the wired transmission unit transmits the information collected by the multi-dimensional sensing module to the data analysis module through a fiber optic network. Fiber optic communication has the advantages of high transmission rate and strong anti-interference ability, and can ensure the accurate and fast transmission of data; the wireless transmission unit uses wireless sensor network technologies such as ZigBee and LoRa to transmit the data to the convergence node, and through the convergence node, the information is transmitted to the data analysis module using a fiber optic cable; the convergence node is a special node that plays a role in data aggregation and forwarding in the network. In a network containing many sensing nodes, such as sensor nodes, these sensing nodes are responsible for collecting various environmental information or monitoring data, while the convergence node receives data from multiple sensing nodes, performs preliminary processing on it, and then transmits the data to a higher-level data processing center or control center; the wired transmission unit and the wireless transmission unit can be used separately or set and used simultaneously according to the regional characteristics in different regions.

[0012] Further, the data analysis module includes a data preprocessing unit, a data accuracy detection unit, a data analysis unit, a comparison and judgment unit, and a fault judgment unit; the data preprocessing unit is used to preprocess the information collected by the multi-dimensional sensing module; the data accuracy detection unit is used to perform timed accuracy detection on the five collected data respectively, and judge whether the data acquisition device fails according to the detection results; the data missing detection unit is used to detect whether there is data missing in the collected data; the data analysis unit includes five data processing centers, which respectively perform feature extraction and analysis calculations on the five collected data, set comparison thresholds for the five data, and analyze whether there are abnormalities in the collected data; the comparison and analysis unit is used to set a fault risk threshold, perform weighted fusion on the five processed data to calculate a fault risk value, and compare the fault risk value with the set fault risk threshold; the fault judgment unit judges whether the GIL fails according to the comparison and analysis results of the fault risk value and the fault risk threshold.

[0013] Further, the fault warning module includes a visualization unit and a warning unit; the visualization unit is used to visually present the detection data; the warning unit is used to issue an audible and visual warning when a fault is diagnosed.

[0014] The multi-dimensional perception monitoring method based on GIL includes the following steps:

[0015] S1. Collect five data information of the temperature, humidity, pressure, vibration intensity, and partial discharge intensity inside the GIL.

[0016] S2. Select and set wired transmission and wireless transmission methods for information transmission according to the environmental conditions.

[0017] S3. Preprocess the collected data, perform data accuracy detection and data missing detection on the data after preprocessing, and perform comparative analysis on the detected data to judge whether the GIL and the data acquisition device fail.

[0018] S4. Issue an audible and visual warning for the faults occurring in the GIL and the data acquisition device.

[0019] Further, in step S1: Use a temperature sensor to collect the internal temperature value T of the GIL during operation. Since heat is generated due to the current heating effect and other factors during the operation of the GIL, excessive temperature may affect performance or even cause failures. Use a humidity sensor to collect the humidity value S of the gas inside the GIL to keep the humidity of the gas inside the GIL normal and prevent the insulation performance from decreasing due to excessive humidity. Use a pressure sensor to collect the pressure value P of the gas inside the GIL, because the gas pressure inside the GIL is a key parameter to ensure insulation performance and operation safety. Use a vibration sensor to collect the vibration intensity K inside the GIL. Since the GIL may vibrate due to factors such as electromagnetic force and mechanical stress during operation, abnormal vibration may indicate problems such as loose internal structure. By installing a vibration sensor and monitoring parameters such as the frequency and amplitude of the vibration, potential failures can be detected in a timely manner. Partial discharge is an important precursor to GIL insulation failure. Use an ultra-high frequency sensor and an ultrasonic sensor to collect the position and intensity value C of the partial discharge. The ultra-high frequency sensor can capture the ultra-high frequency electromagnetic waves generated by the partial discharge inside the GIL, and the ultrasonic sensor can detect the ultrasonic signals generated by the partial discharge, so as to realize the positioning and intensity monitoring of the partial discharge; and classify and store the five collected data.

[0020] Further, in step S2: The methods of information transmission include two information transmission methods: wired transmission and wireless transmission. Set the two information transmission methods according to the actual situation inside the GIL to be monitored. The specific settings are as follows:

[0021] a. For areas in the internal space of the GIL where the openness is greater than p, use the wired transmission method of laying a fiber optic network for information transmission;

[0022] b. For areas in the internal space of the GIL where the openness is less than p, use wireless transmission by placing wireless sensors.

[0023] The two transmission methods can be used separately or simultaneously, and a transmission method can be used as the main transmission means and the other as the auxiliary transmission means; the main transmission means is defined as the first option when selecting an information transmission method, and the auxiliary transmission means is the second option. The second option is a supplementary transmission means for the first option, and information transmission is completed by setting the second option when the first option cannot be used due to the objective environment. When the two information transmission methods are used simultaneously, first divide the internal space of the GIL into n parts, monitor the emptiness of each space area, use the wired transmission method for the areas with an emptiness greater than p, and use the wireless transmission method for the areas with an emptiness less than p. For the n divided areas, the emptiness is monitored uniformly. If the number of areas with an emptiness greater than p among these n areas is greater than n / 2, then select the wired transmission method as the main transmission means and the wireless transmission method as the auxiliary transmission means; otherwise, use the wireless transmission as the main transmission means and the wired transmission as the auxiliary transmission means. Finally, summarize the information transmitted by the two transmission methods for data analysis; the area emptiness is the ratio of the volume of the unoccupied space in each area to the total volume of this area space. The flexible selection of the two information transmission methods improves the adaptability of the monitoring system.

[0024] Further, in step S3: preprocess the five collected data, and the preprocessing includes denoising, filtering, and normalizing the collected raw data;

[0025] And design data accuracy detection, and perform timed data accuracy detection on the five collected data respectively; the data accuracy detection method performs a comprehensive comparison and analysis of data accuracy every fixed time interval Δt, and records the moment t when entering the data accuracy detection stage a , and collect a set of data within the time period [t a , t b and combine it with m sets of anomaly-free data collected when the sensor was initially enabled for data accuracy detection. The time interval Δt0 between the moment t a and the moment t b is Δt0 = t b - t a ; after entering the accuracy detection, a set of data collected within the time period [t a , t b is recorded as data set Q0, and each set of data contains x values; the m sets of anomaly-free data collected when the sensor was initially enabled are recorded as data sets {Q1, Q2,..., Q m}, each data set contains x data, and the collection time interval for each data set is also Δt0; then there are a total of m + 1 sets of data used for data calibration, and these m + 1 sets of data are {Q0, Q1, Q2,..., Q m}; After entering the data accuracy detection stage, calculate the mean of the x data in each data group, and perform a squared difference calculation on the m + 1 means to obtain the data accuracy detection value D; and set the data accuracy threshold D0, compare the detection value with the threshold for analysis. If D < D0, it is judged that the data accuracy is high and the data acquisition device is normal; if D > D0, it is judged that the data accuracy is low, the data acquisition device is abnormal, and an abnormal warning for the data acquisition device is issued; The sensors used by the system to collect information may experience problems such as aging and inaccurate calibration after long-term use. These problems will cause deviations between the collected data and the actual situation; for example, the humidity sensor after being used for a period of time may have a situation where the measured value is too high or too low; if the data collected by the sensor is too high or too low due to long-term use, the squared difference calculated using the data collected in different time periods will increase due to the deviation of the collected data; Therefore, the squared difference calculated using m + 1 means can be used as the data accuracy for comparison with the accuracy threshold, so as to detect the accuracy of the data collected by the sensor, timely discover the aging problem of the sensor, and improve the accuracy of the monitoring system;

[0026] Analyze and process the preprocessed data. The analysis and processing process includes two steps:

[0027] S3-1: Set the safety thresholds for the five collected data respectively, compare the collected data with the safety thresholds for analysis, and use the safety thresholds to analyze in real time whether the five monitoring data inside the GIL are abnormal;

[0028] S3-2: Perform weighted fusion on the five collected data, calculate the real-time fault risk value, and set the fault risk threshold for analyzing whether the GIL has a fault;

[0029] In step S3-1: Set the safety thresholds corresponding to the five collected data. The set thresholds include the temperature safety threshold T max , the pressure safety threshold P max , the humidity safety threshold S max , the vibration intensity safety threshold K max , the partial discharge intensity safety threshold C max ; The values of the five data inside the GIL in the normal operating state are within the set safety thresholds, and a fluctuation image of the single data value changing with time is jointly established for the real-time collected single data and the collection time and visualized; for the partial discharge phenomenon, when the partial discharge situation is monitored, the intensity of the partial discharge and the location where the partial discharge occurs should be presented at the same time, which is convenient for the staff to check and handle the faults of the GIL;

[0030] During the data acquisition process, problems such as sensor failures, communication interruptions, and data loss during data transmission may lead to partial data loss. Therefore, before comparing and analyzing individual data with safety thresholds, data loss detection is performed on the three data items of temperature, pressure, and humidity in the collected individual data; if a certain data item fails to be collected among the three data items of temperature, pressure, and humidity in the collected data, the data value of this collection is recorded as 0, and when the data value is recorded as 0, it is determined that data acquisition is missing; if a certain data item in the collected data has a data loss situation during data loss detection, this group of data is marked as a special state, and a group of data marked as a special state does not participate in the comparison analysis of safety thresholds and the calculation of fault risk values; after data loss detection, data anomalies caused by problems in the data collection or information transmission process can be detected before data analysis, and no redundant abnormal data analysis process is entered; the efficiency and accuracy of data monitoring are improved;

[0031] In step S3-2: After weighted fusion of the collected data, the fault risk value M of the GIL is calculated, and the calculation formula is as follows:

[0032]

[0033] Among them, j1, j2, j3, j4, and j5 respectively represent the weights of temperature, pressure, humidity, vibration intensity, and partial discharge intensity in the calculation of the fault risk value;

[0034] Set the fault risk threshold M0 to compare and analyze with the calculated fault risk value for monitoring the fault risk value;

[0035] During the process of detecting the internal data of the GIL, if it is detected that there is an individual data exceeding the set safety threshold, enter the risk assessment mode, extract the fault risk value calculated in real time, and analyze whether the fault risk value exceeds the fault risk threshold. The analysis results include two situations:

[0036] c. If the fault risk value does not exceed the fault risk threshold, and only a single data item exceeds the safety threshold, then record the time t1 when the data anomaly occurs, and maintain the risk assessment mode during the time period [t1, t2]. During the time period [t1, t2], focus on monitoring the data that exceeds the safety threshold and the fault risk value. The monitoring and analysis results include three situations: i. If the fault risk values do not exceed the fault risk threshold during the time period [t1, t2], and the single data item that originally exceeded the safety threshold has returned to normal after fluctuations, then the risk assessment mode is lifted and normal monitoring is resumed. ii. If the fault risk value exceeds the fault risk threshold during the time period [t1, t2], and the data that exceeded the safety threshold has not returned to normal, a fault warning is triggered. iii. If the fault risk values do not exceed the fault risk threshold during the time period [t1, t2], but the single data item that originally exceeded the safety threshold has not returned to normal, then it is judged to be a single data anomaly, and a single data risk warning is issued.

[0037] d. If a single item of data is detected to exceed the safety threshold and the fault risk value is detected to exceed the fault risk threshold at the same time, a fault risk warning will be issued immediately.

[0038] Furthermore, in step S4: the fluctuation image drawn by jointly drawing the numerical value of the single data collected in real time and its collection time is classified and presented on the monitoring result display screen, and the fault risk value calculated in real time is visualized and presented on the display screen; when the data acquisition equipment is abnormal, the orange warning light is turned on to warn, and a prompt is given in the visualization device; when the GIL is operating normally, the warning light is green, and when an abnormal value is monitored, the color of the warning light changes with the situation; when entering the risk assessment mode, the warning light turns blue; when it is determined that a single data is abnormal, the warning light turns yellow, and a risk prompt for the abnormal data is given on the display screen; when a fault warning is triggered, the warning light turns red and a buzzer alarm is issued; for situation i, the warning light changes from blue to green; for situation ii, the warning light changes from blue to red, and a buzzer alarm is issued; for situation iii, the warning light changes from blue to yellow, and a risk prompt for the abnormal data is given on the display screen.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention designs a multi-dimensional monitoring system for the operating characteristics of GIL. For the data that is prone to fluctuations during the operation of GIL and thus affects the normal operation of GIL, it can timely give early warnings about the faults generated during the operation of GIL, facilitating the staff to troubleshoot and handle the faults. Moreover, the information transmission method designed by the present invention can be flexibly arranged according to the characteristics of different GIL systems, improving the adaptability and generalization ability of the multi-dimensional monitoring system in the present invention. In terms of data analysis and early warning, the present invention judges the failure rate by adding two aspects of data, namely single-item data and the failure risk value of comprehensive calculation. And when the single-item data is abnormal, a risk assessment mode is introduced, improving the accuracy of the monitoring system. In addition, data accuracy detection and data missing detection methods are added, which can timely detect data anomalies caused by abnormal data acquisition devices before data analysis, and stop further data analysis on the data groups with missing data when data missing is detected, improving the data monitoring efficiency of the monitoring system and the accuracy of data monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic structural diagram of the multi-dimensional perception monitoring system based on GIL of the present invention;

[0042] Figure 2 It is a schematic flow diagram of the multi-dimensional perception monitoring method based on GIL of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] As Figure 1 - Figure 2 shown, the present invention provides a technical solution, a multi-dimensional perception monitoring system based on GIL, and the system includes a multi-dimensional sensing module, an information transmission module, a data analysis module, and a fault warning module;

[0045] The multi-dimensional sensing module collects five pieces of data information, namely temperature, humidity, pressure, vibration intensity, and partial discharge intensity inside GIL through sensors;

[0046] The information transmission module uses two methods, wired transmission and wireless transmission, for information transmission;

[0047] The data analysis module is used to process the data through preprocessing, data accuracy detection, and data missing detection, and analyze the data through a comparative analysis method to judge whether GIL and the data acquisition device have faults;

[0048] The fault warning module is used to display the monitoring results and give an audible and visual warning when a fault is detected.

[0049] The multi-dimensional sensing module includes a temperature sensing unit, a pressure sensing unit, a vibration sensing unit, a humidity sensing unit, and a partial discharge sensing unit; the temperature sensing unit uses a fiber Bragg grating temperature sensor to collect the internal temperature information of the GIL during operation in real time; the pressure sensing unit uses a pressure sensor installed in the GIL gas chamber to collect the gas pressure information in real time; the vibration sensing unit monitors the vibration intensity generated inside the GIL by installing a vibration sensor; the humidity sensing unit monitors the gas humidity inside the GIL in real time through a humidity sensor; the partial discharge sensing unit is used to collect the position and intensity information of partial discharges inside the GIL.

[0050] The information transmission module includes a wired transmission unit and a wireless transmission unit; the wired transmission unit transmits the information collected by the multi-dimensional sensing module to the data analysis module through an optical fiber network; the wireless transmission unit uses wireless sensor network technology to transmit the data to the aggregation node, and through the aggregation node, the information is transmitted to the data analysis module using an optical fiber; the wired transmission unit and the wireless transmission unit can be used separately or set and used simultaneously according to the regional characteristics in different regions.

[0051] The data analysis module includes a data preprocessing unit, a data accuracy detection unit, a data analysis unit, a comparison and judgment unit, and a fault judgment unit; the data preprocessing unit is used to preprocess the information collected by the multi-dimensional sensing module; the data accuracy detection unit is used to perform timed accuracy detection on the five collected data respectively, and judge whether there is a fault in the data acquisition device according to the detection results; the data missing detection unit is used to detect whether there is a situation of missing data in the collected data; the data analysis unit includes five data processing centers, which respectively perform feature extraction and analysis calculations on the five collected data, set comparison thresholds for the five data, and analyze whether there are abnormalities in the collected data; the comparison and analysis unit is used to set a fault risk threshold, perform weighted fusion calculation on the five processed data to calculate the fault risk value, and compare the fault risk value with the set fault risk threshold; the fault judgment unit judges whether the GIL has a fault according to the comparison and analysis results of the fault risk value and the fault risk threshold.

[0052] The fault warning module includes a visualization unit and a warning unit; the visualization unit is used to visually present the detection data; the warning unit is used to give an audible and visual warning when a fault is diagnosed.

[0053] A multi-dimensional perception monitoring method based on GIL includes the following steps:

[0054] S1. Collect five pieces of data information including the temperature, humidity, pressure, vibration intensity, and partial discharge intensity inside the GIL.

[0055] S2. Select and set wired and wireless transmission methods for information transmission according to the environmental conditions.

[0056] S3. Preprocess the collected data, and after preprocessing, perform data accuracy detection and data missing detection on the data, and conduct comparative analysis on the detected data to determine whether the GIL and the data acquisition device are faulty.

[0057] S4. Give an audible and visual warning for the faults that occur in the GIL and the data acquisition device.

[0058] In step S1: Use a temperature sensor to collect the internal temperature value T of the GIL during operation; use a humidity sensor to collect the internal gas humidity value S of the GIL; use a pressure sensor to collect the internal gas pressure value P of the GIL; use a vibration sensor to collect the internal vibration intensity K of the GIL; use an ultra-high frequency sensor and an ultrasonic sensor to collect the position and intensity value C of partial discharge; and classify and store the five pieces of collected data.

[0059] In step S2: The methods of information transmission include two information transmission methods, wired transmission and wireless transmission. Set the two information transmission methods according to the actual situation inside the GIL to be monitored. The specific settings are as follows:

[0060] a. For areas in the internal space of the GIL where the openness degree of the area is greater than p, use the wired transmission method of laying a fiber optic network for information transmission.

[0061] b. For areas in the internal space of the GIL where the openness degree of the area is less than p, place wireless sensors for wireless transmission.

[0062] The two transmission methods can be used separately or simultaneously. A transmission method can be used as the primary means while the other as the secondary means. The primary transmission means is defined as the first option when selecting an information transmission method, and the secondary transmission means is the second option. The second option is a supplementary transmission means for the first option. When the first option cannot be used due to objective circumstances, the second option is set to complete information transmission. When the two information transmission methods are used simultaneously, first divide the internal space of GIL into n parts, monitor the emptiness degree of each space area. For areas with an emptiness degree greater than p, use the wired transmission method, and for areas with an emptiness degree less than p, use the wireless transmission method. For the n divided areas, uniformly monitor the emptiness degree. If the number of areas with an emptiness degree greater than p among these n areas is greater than n / 2, then select the wired transmission method as the primary means and the wireless transmission method as the secondary means; otherwise, select the wireless transmission as the primary means and the wired transmission as the secondary means. Finally, summarize the information transmitted by the two transmission methods for data analysis. The area emptiness degree is the ratio of the volume of the unoccupied space in each area to the total volume of this area space.

[0063] In step S3: Preprocess the five collected data. The preprocessing includes denoising, filtering, and normalizing the collected raw data.

[0064] And design data accuracy detection, and perform timing data accuracy detection on the five collected data respectively. The data accuracy detection method conducts a comprehensive comparative analysis of data accuracy every fixed time interval Δt, and records the moment t when entering the data accuracy detection stage. a , and collect a set of data within the time period [t a , t b and combine it with the m sets of anomaly-free data collected when the sensor was initially enabled for data accuracy detection. The time interval Δt0 between the moment t a and the moment t b is Δt0 = t b - t a . After entering the accuracy detection, a set of data collected within the time period [t a , t b is denoted as data group Q0, and each set of data contains x values. Denote the m sets of anomaly-free data collected when the sensor was initially enabled as data groups {Q1, Q2,..., Q m}, each data group contains x data, and the collection time interval of each data group is also Δt0. Then, a total of m + 1 sets of data are used for data calibration, and these m + 1 data groups are {Q0, Q1, Q2,..., Q m}; After entering the data accuracy detection stage, calculate the mean of the x data in each data group, and calculate the squared difference of the m + 1 means to obtain the data accuracy detection value D; and set the data accuracy threshold D0, and compare and analyze the detection value with the threshold. If D < D0, it is judged that the data accuracy is high and the data acquisition device is normal; if D > D0, it is judged that the data accuracy is low, the data acquisition device is abnormal, and an abnormal warning of the data acquisition device is issued.

[0065] Analyze and process the preprocessed data. The analysis and processing process includes two steps:

[0066] S3-1: Set the safety thresholds for the five collected data respectively, compare and analyze the collected data with the safety thresholds, and use the safety thresholds to analyze in real time whether the five monitored data inside the GIL are abnormal;

[0067] S3-2: Perform weighted fusion on the five collected data, calculate the real-time fault risk value, and set the fault risk threshold to analyze whether the GIL has a fault;

[0068] In step S3-1: Set the safety thresholds corresponding to the five collected data. The set thresholds include the temperature safety threshold T max , the pressure safety threshold P max , the humidity safety threshold S max , the vibration intensity safety threshold K max , the partial discharge intensity safety threshold C max ; Set the single data comparison and analysis rule as that the values of the five data inside the GIL under normal operating conditions are within the set safety thresholds, and jointly establish a fluctuation image of the single data value changing with time with the real-time collected single data and the collection time and visualize it; for the partial discharge phenomenon, when the partial discharge situation is monitored, the intensity of the partial discharge and the position where the partial discharge occurs should be presented at the same time.

[0069] During the data collection process, problems such as sensor failures, communication interruptions, and data loss during data transmission may lead to partial data loss. Therefore, before comparing and analyzing individual data with safety thresholds, data loss detection is performed on the three data items of temperature, pressure, and humidity in the collected individual data. If a certain data item fails to be collected among the three data items of temperature, pressure, and humidity in the collected data, the data value of this collection is recorded as 0. When the data value is recorded as 0, it is determined that data collection is missing. If a certain data item in the collected data has a data loss situation during data loss detection, this group of data is marked as a special state. A group of data marked as a special state does not participate in the comparison analysis of safety thresholds and the calculation of fault risk values. After performing data loss detection, data anomalies caused by problems with the data collection device or information transmission can be detected before data analysis, and thus do not enter the redundant abnormal data analysis process. This improves the efficiency and accuracy of data monitoring;

[0070] In step S3-2: After weighted fusion of the collected data, the fault risk value M of the GIL is calculated, and the calculation formula is as follows:

[0071]

[0072] Among them, j1, j2, j3, j4, and j5 respectively represent the weights of temperature, pressure, humidity, vibration intensity, and partial discharge intensity in the calculation of the fault risk value;

[0073] A fault risk threshold M0 is set and compared with the calculated fault risk value for monitoring the fault risk value;

[0074] During the process of detecting the internal data of the GIL, if it is detected that there is an individual data exceeding the set safety threshold, the risk assessment mode is entered, and the fault risk value calculated in real time is extracted to analyze whether the fault risk value exceeds the fault risk threshold. The analysis results include two situations:

[0075] c. If the fault risk value does not exceed the fault risk threshold, and only a single data item exceeds the safety threshold, then record the time t1 when the power generation data is abnormal, and maintain the risk assessment mode in the time period [t1, t2]. In the time period [t1, t2], focus on monitoring the data that exceeds the safety threshold and the fault risk value; the monitoring and analysis results include three situations: i. If the fault risk values in the time period [t1, t2] do not exceed the fault risk threshold, and the single data item that originally exceeded the safety threshold has returned to normal after fluctuations, then the risk assessment mode is released and normal monitoring is resumed; ii. If the fault risk value in the time period [t1, t2] exceeds the fault risk threshold, and the data that exceeded the safety threshold has not returned to normal, a fault warning is triggered; iii. If the fault risk values in the time period [t1, t2] do not exceed the fault risk threshold, but the single data item that originally exceeded the safety threshold has not returned to normal, it is judged that the single data item is abnormal, and a single data risk warning is issued;

[0076] d. If a single item of data is detected to exceed the safety threshold and the fault risk value is detected to exceed the fault risk threshold at the same time, a fault risk warning will be issued immediately.

[0077] In step S4: the fluctuation image drawn by combining the numerical value of the single data collected in real time with its collection time is classified and presented on the monitoring result display screen, and the fault risk value calculated in real time is visualized and presented on the display screen; when the GIL is operating normally, the warning light is green, and when an abnormal value is detected, the color of the warning light changes with the situation; when entering the risk assessment mode, the warning light turns blue; when it is determined that a single data is abnormal, the warning light turns yellow, and a risk warning of the abnormal data is given on the display screen; when a fault warning is triggered, the warning light turns red and a buzzer alarm is issued; for situation i, the warning light changes from blue to green; for situation ii, the warning light changes from blue to red, and a buzzer alarm is issued; for situation iii, the warning light changes from blue to yellow, and a risk warning of the abnormal data is given on the display screen.

[0078] Example 1:

[0079] In step S1: the temperature value T = 320K inside the GIL is collected by the temperature sensor; the humidity value S = 280μL / L of the gas inside the GIL is collected by the humidity sensor; the pressure value P = 0.65MPa of the gas inside the GIL is collected by the pressure sensor; the vibration intensity K = 0 inside the GIL is collected by the vibration sensor; the location of the partial discharge is collected by the ultra-high frequency sensor and the ultrasonic sensor. and intensity value C=0; and the five collected data are classified and stored.

[0080] In step S2: Design a dual-line information transmission method. The dual-line transmission method includes two information transmission methods: wired transmission and wireless transmission. Arrange the two information transmission methods according to the actual situation inside the GIL that needs to be monitored. The specific arrangement is as follows:

[0081] a. For areas in the GIL internal space with an open space ratio greater than 40%, use a wired transmission method of laying out a fiber optic network for information transmission;

[0082] b. For areas in the GIL internal space with an open space ratio less than 40%, place wireless sensors for wireless transmission;

[0083] The two transmission methods can be used separately or simultaneously, and can be used in a transmission mode where one is the main transmission means and the other is the auxiliary transmission means. The definition of the main transmission means is the first option when selecting the information transmission method, and the auxiliary transmission means is the second option. The second option is a supplementary transmission means for the first option. When the first option cannot be used due to the objective environment, the second option is used to complete information transmission. When the two information transmission methods are used simultaneously, first divide the GIL internal space into 100 parts, monitor the open space ratio of each space area. For areas with an open space ratio greater than 40%, use the wired transmission method, and for areas with an open space ratio less than 40%, use the wireless transmission method. For the 100 divided areas, uniformly monitor the open space ratio. If the number of areas with an open space ratio greater than 40% among these 100 areas is greater than 50, then select the wired transmission method as the main transmission means and the wireless transmission method as the auxiliary transmission means; otherwise, use the wireless transmission as the main transmission means and the wired transmission as the auxiliary transmission means. Finally, summarize the information transmitted by the two transmission methods for data analysis. The open space ratio of the area is the ratio of the volume of the unoccupied space in each area to the total volume of this area space.

[0084] In step S3: Preprocess the five collected data. The preprocessing includes denoising, filtering, and normalizing the collected raw data;

[0085] And design data accuracy detection, and perform timed data accuracy detection on the five collected data respectively. The data accuracy detection method conducts a comprehensive comparative analysis of data accuracy every fixed time interval Δt = 300h, and records the moment t when entering the data accuracy detection stage a , and collect a set of data within the time period [t a , t b and combine it with 20 sets of anomaly-free data collected when the sensor was initially enabled for data accuracy detection. The time interval Δt0 between the moment t a and the moment t b is Δt0 = tb -t a ; After entering the accuracy detection, a set of data collected within the time period [t a , t b is recorded as data set Q0, and each set of data contains 10 values; m sets of anomaly-free data collected when the sensor is initially enabled are recorded as data sets {Q1, Q2,..., Q 20}, each data set contains 10 data, and the acquisition time interval for each data set is also Δt0; then a total of 21 sets of data are used for data calibration, and these 21 sets of data are {Q0, Q1, Q2,..., Q 20}; After entering the data accuracy detection stage, calculate the mean value of each data set, and perform a variance calculation on the 21 mean values to obtain the data accuracy detection value D; and set the data accuracy threshold D0 = 0.05, compare the detection value with the threshold for analysis. If D < D0, it is judged that the data accuracy is high and the data acquisition device is normal; if D > D0, it is judged that the data accuracy is low and the data acquisition device has an anomaly, and an anomaly warning for the data acquisition device is issued;

[0086] Analyze and process the preprocessed data, and the analysis and processing process includes two steps:

[0087] S3-1: Set the safety thresholds for the five collected data respectively, compare the collected data with the safety thresholds for analysis, and use the safety thresholds to analyze in real time whether the five monitored data inside the GIL are abnormal;

[0088] S3-2: Perform weighted fusion on the five collected data, calculate the real-time fault risk value, and set a fault risk threshold for analyzing whether the GIL has a fault;

[0089] In step S3-1: Set the safety thresholds corresponding to the five collected data. The set thresholds include the temperature safety threshold T max = 330K, the pressure safety threshold P max = 0.7MPa, the humidity safety threshold S max = 300 μL / L, the vibration intensity safety threshold K max = 150 Hz, and the partial discharge intensity safety threshold C max = 0.5 J / s; Set the single-data comparison and analysis rule as follows: The values of the five data inside the GIL under normal operating conditions are within the set safety thresholds, and a fluctuation image of the single-data value changing with time is jointly established for the real-time collected single data and the acquisition time and visualized; for the partial discharge phenomenon, when the partial discharge situation is monitored, the intensity of the partial discharge and the location where the partial discharge occurs should be presented simultaneously;

[0090] During the data acquisition process, due to problems such as sensor failures, communication interruptions, and data loss during data transmission, some data may be missing. Therefore, before comparing individual data with safety thresholds for analysis, data missing detection is performed on the temperature, pressure, and humidity data among the collected individual data; if a certain data acquisition fails among the temperature, pressure, and humidity data in the collected data, the data value of this acquisition is recorded as 0, and when the data value is recorded as 0, it is determined that data acquisition is missing; if there is data missing in a certain data collected during data missing detection, this group of data is marked as a special status, and a group of data marked as a special status does not participate in the comparison analysis of safety thresholds and the calculation of fault risk values; after data missing detection, data anomalies caused by equipment or information transmission problems in the collected data can be detected before data analysis, and thus do not enter the redundant abnormal data analysis process; this improves the efficiency and accuracy of data monitoring.

[0091] In step S3-2: After weighted fusion of the collected data, the fault risk value M of the GIL is calculated, and the calculation formula is as follows:

[0092]

[0093] Among them, j1, j2, j3, j4, and j5 respectively represent the weights of temperature, pressure, humidity, vibration intensity, and partial discharge intensity in the calculation of the fault risk value, and their values are j1 = 0.1; j2 = 0.1; j3 = 0.1; j4 = 0.3; j5 = 0.4; the calculated M = -0.717;

[0094] Set the fault risk threshold M0 = 0.1 and compare it with the calculated fault risk value for monitoring the fault risk value; according to the monitoring results, the temperature, humidity, and pressure of the gas inside the GIL are all less than the safety thresholds, and no vibration and partial discharge are detected, and the calculated fault risk value is less than the fault risk threshold, so it is determined that the GIL is operating normally and the warning light is green.

[0095] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, in any regard, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. The multi-dimensional perception monitoring system based on GIL is characterized in that: The system includes a multi-dimensional sensing module, an information transmission module, a data analysis module, and a fault warning module; The multi-dimensional sensing module collects five pieces of data information, namely temperature, humidity, pressure, vibration intensity, and partial discharge intensity inside the GIL through sensors; The information transmission module uses two methods, wired transmission and wireless transmission, for information transmission; The data analysis module is used to process the data through preprocessing, data accuracy detection, and data missing detection, and analyze the data through a comparative analysis method to determine whether the GIL and the data acquisition device have failures; The fault warning module is used to display the monitoring results and give an audible and visual warning when a fault is detected.

2. The multi-dimensional perception monitoring system based on GIL according to claim 1, wherein: The multi-dimensional sensing module includes a temperature sensing unit, a pressure sensing unit, a vibration sensing unit, a humidity sensing unit, and a partial discharge sensing unit; the temperature sensing unit collects the internal temperature information of the GIL during operation in real time through a fiber Bragg grating temperature sensor; the pressure sensing unit collects the gas pressure information in real time by installing a pressure sensor in the GIL gas chamber; the vibration sensing unit monitors the vibration intensity generated inside the GIL by installing a vibration sensor; the humidity sensing unit monitors the gas humidity inside the GIL in real time through a humidity sensor; the partial discharge sensing unit is used to collect the position and intensity information of partial discharges inside the GIL.

3. The multi-dimensional perception monitoring system based on GIL according to claim 1, characterized in that: The information transmission module includes a wired transmission unit and a wireless transmission unit; the wired transmission unit transmits the collected information to the data analysis module through an optical fiber network; the wireless transmission unit uses wireless sensor network technology to transmit the data to the aggregation node, and through the aggregation node, transmits the information to the data analysis module using an optical fiber; the wired transmission unit and the wireless transmission unit can be used separately or set and used simultaneously according to the regional characteristics in different regions.

4. The multi-dimensional perception monitoring system based on GIL according to claim 1, wherein: The data analysis module includes a data preprocessing unit, a data accuracy detection unit, a data missing detection unit, a data analysis unit, a comparative judgment unit, and a fault judgment unit; the data preprocessing unit is used to preprocess the information collected by the multi-dimensional sensing module; the data accuracy detection unit is used to separately detect the accuracy of the five pieces of collected data and judge whether the data acquisition device has a failure according to the detection results; The data missing detection unit is used to detect whether there is a situation of missing data in the collected data; the data analysis unit includes five data processing centers, which respectively perform feature extraction and analysis calculations on the five pieces of collected data, set the comparison thresholds for the five pieces of data, and analyze whether there are abnormalities in the collected data; The comparative analysis unit is used to set the fault risk threshold, perform weighted fusion on the processed five pieces of data to calculate the fault risk value, and perform comparative analysis on the fault risk value and the set fault risk threshold; The fault judgment unit judges whether the GIL has a failure according to the comparative analysis result of the fault risk value and the fault risk threshold.

5. The multi-dimensional perception monitoring system based on GIL according to claim 1, wherein: The fault warning module includes a visualization unit and a warning unit; the visualization unit is used to visually present the detection data; the warning unit is used to give an audible and visual warning when a fault is diagnosed.

6. The multi-dimensional perception monitoring method based on GIL is characterized in that: It includes the following steps: S1. Collect five data information items, namely the temperature, humidity, pressure, vibration intensity, and partial discharge intensity inside the GIL; S2. Select and set the wired transmission and wireless transmission methods for information transmission according to the environmental conditions; S3. Preprocess the collected data, and after preprocessing, conduct data accuracy detection and data missing detection on the data, and conduct comparative analysis on the detected data to determine whether the GIL and the data acquisition device have failures; S4. Give an audible and visual warning for the failures of the GIL and the data acquisition device.

7. The multi-dimensional perception monitoring method based on GIL according to claim 6, characterized in that: In step S1: Use a temperature sensor to collect the internal temperature value T of the GIL during operation; use a humidity sensor to collect the internal gas humidity value S of the GIL; use a pressure sensor to collect the internal gas pressure value P of the GIL; use a vibration sensor to collect the internal vibration intensity K of the GIL; use an ultra-high frequency sensor and an ultrasonic sensor to collect the position and intensity value C of partial discharge; and classify and store the five collected data items.

8. The multi-dimensional perception monitoring method based on GIL according to claim 6, wherein: In step S2: The methods of information transmission include two information transmission methods, namely wired transmission and wireless transmission. Set the two information transmission methods according to the actual situation inside the GIL to be monitored as needed. The specific settings are as follows: a. For areas in the internal space of the GIL where the degree of openness of the area is greater than p, use the wired transmission method of laying out a fiber optic network for information transmission; b. For areas in the internal space of the GIL where the degree of openness of the area is less than p, use wireless sensors for wireless transmission; The two transmission methods can be used separately or simultaneously, and can be used in a transmission mode where one is used as the main transmission means and the other is used as the auxiliary transmission means; the definition of the main transmission means is the first option when selecting the information transmission method, and the auxiliary transmission means is the second option for selecting the information transmission method. The second option is a supplementary transmission means for the first option. Information transmission is completed by setting the second option at positions where the first option cannot be used due to objective circumstances; when the two information transmission methods are used simultaneously, first divide the internal space of the GIL into n parts, monitor the degree of openness of each part of the space area, use the wired transmission method for areas where the degree of openness of the area is greater than p, and use the wireless transmission method for areas where the degree of openness of the area is less than p; For the n divided areas, uniformly conduct the degree of openness detection. If the number of areas with a degree of openness greater than p among these n areas is greater than n / 2, then select the wired transmission method as the main transmission means and the wireless transmission method as the auxiliary transmission means; On the contrary, select the wireless transmission as the main transmission means and the wired transmission as the auxiliary transmission means; finally, summarize the information transmitted by the two transmission methods for data analysis; the degree of openness of the area is the ratio of the volume of the unoccupied space in each area to the total volume of the space in this area.

9. The multi-dimensional perception monitoring method based on GIL according to claim 6, characterized in that: In step S3: Preprocess the five collected data items. The preprocessing includes denoising, filtering, and normalization of the collected original data; And design data accuracy detection, and perform timed data accuracy detection on the five collected data respectively; the data accuracy detection method conducts a comprehensive comparative analysis of data accuracy every fixed time interval Δt, and records the moment t when entering the data accuracy detection stage a , and collect a set of data within the time period [t a , t b and combine it with the m groups of anomaly-free data collected when the sensor was initially enabled for data accuracy detection. The time interval Δt0 between the moment t a and the moment t b is Δt0 = t b - t a ; after entering the accuracy detection stage, a set of data collected within the time period [t a , t b is recorded as data group Q0, and each group of data contains x values; the m groups of anomaly-free data collected when the sensor was initially enabled are recorded as data groups {Q1, Q2,..., Q m}, each data group contains x data, and the collection time interval of each data group is also Δt0; then there are a total of m + 1 groups of data used for data calibration; after entering the data accuracy detection stage, calculate the mean of the x data contained in each data group, and perform a squared difference calculation on the m + 1 means to obtain the data accuracy detection value D; and set the data accuracy threshold D0, and compare the detection value with the threshold for analysis. If D < D0, it is judged that the data accuracy is high and the data acquisition device is normal; if D > D0, it is judged that the data accuracy is low, the data acquisition device is abnormal, and an abnormal warning for the data acquisition device is issued; Conduct analysis and processing on the preprocessed data. The analysis and processing process includes two steps: S3-1: Respectively set the safety thresholds for the five collected data items, compare the collected data with the safety thresholds for comparative analysis, and use the safety thresholds in real time to analyze whether the five monitored data items inside the GIL are abnormal; S3-2: Conduct weighted fusion on the five collected data items, calculate the real-time failure risk value, and set a failure risk threshold for analyzing whether the GIL has failed; In step S3-1: Set the safety thresholds corresponding to the five collected data. The set thresholds include the temperature safety threshold T max , the pressure safety threshold P max , the humidity safety threshold S max , the vibration intensity safety threshold K max , and the partial discharge intensity safety threshold C max ; The values of the five data inside the GIL under normal operating conditions are within the set safety thresholds, and a fluctuation image of the single data value changing with time is jointly established with the real-time collected single data and the collection time and visualized; for the partial discharge phenomenon, when the partial discharge situation is monitored, the intensity of the partial discharge and the location where the partial discharge occurs should be presented simultaneously; Before comparing and analyzing individual data items with the safety threshold, a data missing check is performed on the temperature, pressure, and humidity data items collected. If any of the temperature, pressure, and humidity data items fails to be collected, the data value collected is recorded as 0. If a data value is recorded as 0, it is determined that data collection is missing. If a certain data item collected during data missing detection is missing, this group of data will be marked as a special state. The group of data marked as a special state will not be included in the comparative analysis of safety thresholds and the calculation of fault risk values. In step S3-2: After weighted fusion of the collected data, the fault risk value M of the GIL is calculated. The weighted fusion process is as follows: the weights of the five data items in the calculation of the fault risk value are set as j1, j2, j3, j4, and j5, which represent the weights of temperature, pressure, humidity, vibration intensity, and partial discharge intensity in the calculation of the fault risk value respectively; each monitored value is subtracted from its corresponding safety threshold, and the difference is divided by the safety threshold. The obtained ratio is then multiplied by the weight of the data item to obtain the contribution value of each data item to the fault risk value. The five calculated data items are then added together to obtain the fault risk value M. And set the fault risk threshold M0 and compare and analyze the calculated fault risk value to monitor the fault risk value; During the monitoring of GIL internal data, if a single item of data is detected to exceed the set safety threshold, the system enters risk assessment mode and extracts the fault risk value calculated in real time to analyze whether the fault risk value exceeds the fault risk threshold. The analysis results include two situations: c. If the fault risk value does not exceed the fault risk threshold, and only a single data item exceeds the safety threshold, then record the time t1 when the data anomaly occurred and maintain the risk assessment mode for the time period [t1, t2]. During the time period [t1, t2], focus on monitoring the data that exceeds the safety threshold and the fault risk value. The monitoring and analysis results include three scenarios: i. If the fault risk values do not exceed the fault risk threshold during the time period [t1, t2], and the single data item that originally exceeded the safety threshold returns to normal after fluctuations, then release the risk assessment mode and resume normal monitoring. ii. If the fault risk value exceeds the fault risk threshold during the time period [t1, t2], and the data exceeding the safety threshold has not returned to normal, a fault warning is triggered; iii. If the fault risk values within the time period [t1, t2] do not exceed the fault risk threshold, but the individual data item that originally exceeded the safety threshold has not returned to normal, the individual data item is considered abnormal and a risk warning for the individual data item is issued; d. If a single item of data is detected to exceed the safety threshold and the fault risk value is detected to exceed the fault risk threshold at the same time, a fault risk warning will be issued immediately.

10. The multi-dimensional perception monitoring method based on GIL according to claim 6, characterized in that: In step S4: the fluctuation image obtained by combining the value of the real-time collected single data with its collection time is classified and presented on the monitoring result display screen, and the real-time calculated fault risk value is visualized and presented on the display screen; When the data acquisition device is abnormal, the orange warning light lights up for warning and a prompt is given in the visualization device; When the GIL is operating normally, the warning light is green. When abnormal values are detected, the color of the warning light changes according to the situation; When entering the risk assessment mode, the warning light turns blue; When it is determined that a single item of data is abnormal, the warning light turns yellow and a risk prompt for the abnormal data is given on the display screen; When a fault warning is triggered, the warning light turns red and a buzzer alarm is issued; for situation i, the warning light changes from blue to green; for situation ii, the warning light changes from blue to red and a buzzer alarm is issued; for situation iii, the warning light changes from blue to yellow and a risk prompt for the abnormal data is given on the display screen.

Images