Converter station valve hall intelligent monitoring method and system based on data fusion

Abstract

The invention discloses a converter station valve hall intelligent monitoring method and system based on data fusion, and the method comprises the steps: collecting the operation parameters of a converter station valve hall in real time through a sensor group, and obtaining original multi-source data; the method comprises the following steps: preprocessing original multi-source data to obtain preprocessed operation parameters, and extracting corresponding key features; performing multi-modal data fusion on the key features to obtain a data feature set; analyzing the data feature set by using a machine learning algorithm to obtain a classification result of the equipment state; and if the equipment is in the fault state, an early warning mechanism is triggered, and a state report of the converter station valve hall equipment is generated, so that intelligent monitoring of the converter station valve hall is realized. According to the method, the multi-dimensional operation parameters of the equipment can be comprehensively obtained, operation and maintenance personnel can implement more accurate preventive maintenance according to the change of the state of the equipment through an advanced fault prediction and intelligent alarm mechanism, and the influence of sudden faults on production is reduced, so that the repair cost and the production loss of the equipment are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power system monitoring, and in particular to a method and system for intelligent monitoring of a converter station valve hall based on data fusion. Background Art

[0002] With the rapid development of power systems and the increasing demand for electricity, especially with the widespread adoption of high-voltage direct current (HVDC) transmission systems, converter stations, as a crucial component of power systems, are crucial for their stability and safety, ensuring reliable operation. The valve hall of a converter station is the core area of ​​the converter station, responsible for converting DC current into AC current. Due to the complexity and high workload of the equipment within the valve hall, equipment failure or abnormal operation can severely impact power system stability and may even lead to severe consequences such as large-scale power outages. Therefore, real-time monitoring and efficient management of the valve hall are particularly important.

[0003] Currently, traditional converter station valve hall monitoring methods mainly rely on manual inspections and single-sensor data collection. Although they can provide basic equipment status information, due to the single monitoring data and long acquisition cycle, it is difficult to achieve real-time, comprehensive fault warning and accurate analysis. In addition, with the gradual development of intelligent operation and maintenance technology, the existing monitoring system generally has the following problems:

[0004] 1. Inadequate data collection methods: Traditional monitoring systems mostly rely on local sensors for data collection, failing to comprehensively consider data of different types and dimensions. They lack comprehensive data fusion and intelligent analysis, making it difficult to monitor the operating status of the converter station valve hall from multiple angles and in all directions.

[0005] 2. Poor real-time performance: The existing monitoring system has a slow analysis and processing speed for equipment operating status, lacks real-time alarm and fault diagnosis capabilities, and is unable to detect potential risks in a timely manner, resulting in an inability to respond quickly when equipment failures occur, thus affecting the safety and stability of the power system.

[0006] 3. Low level of intelligence: Traditional monitoring systems are mostly single-function passive monitoring systems that are unable to conduct active analysis, early warning, and optimized scheduling based on the actual operating status of the equipment. This makes fault prevention and maintenance management of the converter station valve hall rely on manual judgment, reducing work efficiency and management accuracy.

[0007] 4. Lack of system integration: The operating status of the converter station valve hall involves multiple physical quantities (such as voltage, current, temperature, pressure, vibration, etc.), but existing monitoring systems are often unable to integrate these heterogeneous data for in-depth analysis, resulting in limited ability to comprehensively evaluate equipment and predict faults. Summary of the Invention

[0008] The purpose of the present invention is to provide a method and system for intelligent monitoring of the valve hall of a converter station based on data fusion. Through the collection, feature extraction and fusion of multi-dimensional sensor data, the key operating parameters of the valve hall of the converter station can be monitored in real time, and an alarm can be issued in time when an abnormality occurs or the warning threshold is triggered. The real-time and accuracy of the monitoring data are improved, and the ability to predict equipment failures is enhanced through intelligent analysis, thereby improving the operating efficiency and safety level of the converter station.

[0009] The present invention adopts the following technical solution: a method for intelligent monitoring of a converter station valve hall based on data fusion, comprising the following steps:

[0010] S1. Use the sensor group to collect the operating parameters of the converter station valve hall in real time and obtain original multi-source data.

[0011] S2. Preprocess the original multi-source data in step S1 to obtain preprocessed operating parameters and extract corresponding key features.

[0012] S3. Perform multimodal data fusion on the key features in step S2 to obtain a data feature set.

[0013] S4. Use a machine learning algorithm to analyze the data feature set in step S3 to obtain a classification result of the device status.

[0014] S5. Based on the result obtained in step S4, if the equipment is in a fault state, the early warning mechanism is triggered and a status report of the converter station valve hall equipment is generated to realize intelligent monitoring of the converter station valve hall.

[0015] Furthermore, in step S1 , the sensor group includes a voltage sensor, a current sensor, a temperature sensor, a pressure sensor, a vibration sensor, and a humidity sensor.

[0016] The voltage sensor is used to measure the voltage level of the equipment in the valve hall of the converter station; the current sensor is used to collect the current data in the valve hall of the converter station; the temperature sensor is used to collect the temperature data in the valve hall of the converter station; the pressure sensor is used to collect the working status of the hydraulic or pneumatic equipment in the valve hall of the converter station; the vibration sensor is used to collect the vibration data of the equipment or mechanical system in the valve hall of the converter station; the humidity sensor is used to collect the humidity data in the valve hall of the converter station.

[0017] Furthermore, in step S2, the original multi-source data is cleaned, normalized, and time-series aligned, and denoised using a digital filter to obtain pre-processed operating parameters.

[0018] The temperature rise rate and thermal stability are extracted from the preprocessed temperature and humidity data; the root mean square value of vibration acceleration, main frequency distribution, impact coefficient and kurtosis are extracted from the preprocessed vibration data; the current peak value, voltage fluctuation rate and overload duration are extracted from the preprocessed current and voltage data; the pressure fluctuation amplitude, pressure peak duration, pressure change rate and abnormal fluctuation frequency are extracted from the working status of the preprocessed hydraulic or pneumatic equipment.

[0019] Furthermore, in step S3, the data feature set obtained includes the following:

[0020] The key features are weighted and fused to obtain the data feature set; the formula for assigning weights is:

[0021]

[0022] Among them, ω i represents the weight of the i-th sensor data, α represents the weight attenuation coefficient, D i Represents the error index of the i-th sensor data.

[0023] The data feature set includes temperature fusion features, current fusion features, voltage fusion features, partial discharge fusion features, vibration fusion features and humidity / air pressure features.

[0024] Furthermore, in step S4, the classification result of the device status includes the following:

[0025] A fault diagnosis model is constructed using a machine learning algorithm. The training data includes historical sensor data under normal operating conditions and various fault conditions, and is labeled as the normal or fault state of the equipment. Based on the labeled training data, the fault diagnosis model is trained using a cross-validation method to obtain a trained fault diagnosis model, which includes an input layer, a hidden layer, and an output layer.

[0026] The data feature set is processed using time domain analysis and frequency domain analysis methods to extract preliminary features of the equipment status. Statistical methods are used to identify abnormal points in the preliminary features to obtain a preliminary feature set of the equipment status.

[0027] The preliminary feature set is input into the input layer of the trained fault diagnosis model. After nonlinear mapping and feature combination processing in the hidden layer, and through the operation of weight matrix and activation function, the deep features are extracted to obtain the classification result of the equipment status. The output layer performs probability scoring on the classification result and selects the category with the highest probability as the final classification result.

[0028] When the probability of the final classification result is greater than or equal to the fault threshold and at least two features in the data feature set are greater than the abnormality threshold, it indicates that the device is in a fault state.

[0029] The abnormal thresholds of the characteristics are as follows: the abnormal threshold of the temperature fusion characteristic is 75°C, the abnormal threshold of the current fusion characteristic is ±15% of the normal rated value, the abnormal threshold of the vibration fusion characteristic is 1.5g, and the abnormal threshold of the partial discharge fusion characteristic is 100pC; where g represents the acceleration of gravity.

[0030] Furthermore, the failure threshold is 0.85.

[0031] Furthermore, it also includes a visual monitoring platform, which displays the equipment's operating parameters, fault warning information, historical data and equipment health status scores in real time through a graphical interface.

[0032] Furthermore, the present invention also proposes a system for intelligent monitoring of a converter station valve hall based on data fusion, comprising:

[0033] The data acquisition module is used to use the sensor group to collect the operating parameters of the converter station valve hall in real time and obtain original multi-source data.

[0034] The data feature set acquisition module is used to preprocess the original multi-source data, obtain the preprocessed operating parameters, and extract the corresponding key features; the key features are subjected to multimodal data fusion to obtain the data feature set.

[0035] The monitoring module is used to analyze the data feature set using a machine learning algorithm to obtain the classification results of the equipment status. If the equipment is in a fault state, the early warning mechanism is triggered and a status report of the converter station valve hall equipment is generated, realizing intelligent monitoring of the converter station valve hall.

[0036] Furthermore, the present invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the steps of the intelligent monitoring method of the converter station valve hall based on data fusion are implemented.

[0037] Furthermore, the present invention also proposes a computer-readable storage medium, which stores a computer program. When the computer program is run by a processor, it executes the intelligent monitoring method for the converter station valve hall based on data fusion.

[0038] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0039] 1. The present invention integrates multiple types of sensors (such as voltage, current, temperature, vibration, etc.) to collect data in real time, overcoming the defect of the limited monitoring range of existing single sensors. It can comprehensively obtain the multi-dimensional operating parameters of the equipment, achieve a more comprehensive and accurate assessment of the equipment status, provide more accurate monitoring information for operation and maintenance personnel, and avoid the misjudgment problem that may be caused by a single data source.

[0040] 2. By adopting efficient sensor data transmission and real-time data processing technology, the present invention can complete the entire process of data collection, processing, analysis and early warning within milliseconds, significantly improving the response speed.

[0041] 3. By adopting data fusion and machine learning algorithms, the present invention can monitor the current status of the equipment and predict the future operating trends and potential failures of the equipment based on the historical operating data and real-time data of the equipment, greatly improving the level of intelligence, and can actively identify hidden dangers in the operation of the equipment, provide early warning and intelligent diagnosis.

[0042] 4. The present invention greatly improves the efficiency and safety of equipment operation and maintenance through functions such as real-time data collection, fault prediction, and automatic alarm.

[0043] 5. This invention utilizes an integrated intelligent monitoring platform that enables data collection, processing, and integration from multiple sensors, and seamlessly integrates with other systems (such as operations and maintenance management systems and remote monitoring platforms). Furthermore, this invention is highly scalable, allowing for the flexible integration of new sensors and algorithm models based on future upgrades to converter station valve hall equipment and new technological developments.

[0044] 6. The present invention can greatly reduce downtime and maintenance costs caused by equipment failure. The early fault prediction and intelligent alarm mechanism enable operation and maintenance personnel to implement more accurate preventive maintenance according to changes in equipment status, reducing the impact of sudden failures on production, thereby reducing equipment repair costs and production losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is an overall implementation flow chart of the present invention. DETAILED DESCRIPTION

[0046] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0047] To achieve the above objectives, the present invention proposes an intelligent monitoring method for converter station valve hall based on data fusion, such as Figure 1 The specific steps are as follows:

[0048] S1. Use the sensor group to collect the operating parameters of the converter station valve hall in real time to obtain original multi-source data; specifically:

[0049] The sensor group includes voltage sensor, current sensor, temperature sensor, pressure sensor, vibration sensor and humidity sensor.

[0050] The voltage sensor is used to measure the voltage level of the equipment in the valve hall of the converter station; the current sensor is used to collect the current data in the valve hall of the converter station; the temperature sensor is used to collect the temperature data in the valve hall of the converter station; the pressure sensor is used to collect the working status of the hydraulic or pneumatic equipment in the valve hall of the converter station; the vibration sensor is used to collect the vibration data of the equipment or mechanical system in the valve hall of the converter station; the humidity sensor is used to collect the humidity data in the valve hall of the converter station.

[0051] S2. The original multi-source data is usually high-dimensional, multi-modal, and may contain noise and redundant information. Therefore, the original multi-source data in step S1 is preprocessed to obtain the preprocessed operating parameters and extract the corresponding key features; specifically:

[0052] The original multi-source data is cleaned, normalized, and time-series aligned to ensure that the units and dimensions of different data sources are consistent; digital filters (such as Kalman filters or median filters) are used for denoising to eliminate errors caused by environmental interference or sensor accuracy limitations, and obtain preprocessed operating parameters.

[0053] Common noises include abnormal fluctuations caused by sensor failures and electromagnetic interference.

[0054] The temperature rise rate and thermal stability are extracted from the preprocessed temperature and humidity data; the root mean square value of vibration acceleration, main frequency distribution, impact coefficient and kurtosis are extracted from the preprocessed vibration data; the current peak value, voltage fluctuation rate and overload duration are extracted from the preprocessed current and voltage data; the pressure fluctuation amplitude, pressure peak duration, pressure change rate and abnormal fluctuation frequency are extracted from the working status of the preprocessed hydraulic or pneumatic equipment to reflect the operating stability and potential fault hazards of the hydraulic or pneumatic system.

[0055] S3. Perform multimodal data fusion on the key features in step S2 to obtain a data feature set; specifically:

[0056] According to the reliability and accuracy of each sensor, the key features are weighted and fused to obtain the data feature set; the formula for assigning weights is:

[0057]

[0058] Among them, ω i represents the weight of the i-th sensor data, α represents the weight attenuation coefficient, Di Represents the error index of the i-th sensor data.

[0059] When D i The smaller it is, the more reliable the data is. i The closer it is to 1. When the error is large, the weight gradually decreases, reflecting a decrease in data credibility.

[0060] The data feature set includes temperature fusion features, current fusion features, voltage fusion features, partial discharge fusion features, vibration fusion features and humidity / air pressure features.

[0061] S4. Analyze the data feature set in step S3 using a machine learning algorithm to obtain a classification result of the device status; specifically:

[0062] A fault diagnosis model is constructed using machine learning algorithms (such as multi-layer perceptron, support vector machine, random forest, etc.); the training data includes historical sensor data under normal operating conditions and various fault conditions, and is labeled as the normal or fault state of the equipment; based on the labeled training data, the fault diagnosis model is trained using the cross-validation method to obtain a trained fault diagnosis model, which includes an input layer, a hidden layer, and an output layer.

[0063] Time domain analysis methods (such as statistical indicators such as mean, variance, and kurtosis) and frequency domain analysis methods (such as Fourier transform and power spectral density analysis) are used to deeply explore the data feature set and extract preliminary features that can reflect the equipment status. Statistical methods (such as box plot outlier detection and Z-score standardization) are used to identify abnormal points in the preliminary features, assist in screening effective features, and obtain a preliminary feature set of the equipment status.

[0064] The preliminary feature set is input into the input layer of the trained fault diagnosis model. After nonlinear mapping and feature combination processing in the hidden layer, and through the operation of weight matrix and activation function, deep features are extracted to obtain the classification results of the equipment status (such as "normal", "fault type A", "fault type B", etc.). The output layer performs probability scoring on the classification results and selects the category with the highest probability as the final classification result.

[0065] When the probability of the final classification result is greater than or equal to 0.85 and at least two features in the data feature set are greater than the abnormal threshold, it indicates that the device is in a fault state.

[0066] The specific abnormality thresholds for the characteristics are as follows: For temperature fusion results, which generally range from 40 to 80°C, the abnormality threshold for the temperature fusion characteristic is 75°C; for current fusion results, which range from 500 to 1500A, the abnormality threshold for the current fusion characteristic is ±15% of the normal rated value; for vibration acceleration fusion values, which range from 0.2 to 2.0g, the abnormality threshold for vibration fusion characteristics is 1.5g; for partial discharge fusion values, which range from 10 to 500pC, the abnormality threshold for partial discharge fusion characteristics is 100pC. g represents the acceleration of gravity.

[0067] S5. Based on the result obtained in step S4, if the equipment is in a fault state, the early warning mechanism is triggered and a status report of the converter station valve hall equipment is generated. Before the equipment fails, potential problems can be identified through the prediction model and an alarm can be issued in advance to remind the operation and maintenance personnel to check and maintain the equipment in time, thereby realizing intelligent monitoring of the converter station valve hall.

[0068] The status report of the converter station valve hall equipment includes the current values, trend analysis and historical data of key parameters such as equipment temperature, voltage, current and vibration.

[0069] To enhance O&M personnel's real-time understanding of equipment status, a visual monitoring platform is also provided. This graphical interface displays equipment operating parameters, fault warnings, and historical data in real time. The platform supports real-time data refresh, trend chart display, and equipment health status scoring, allowing O&M personnel to intuitively understand the overall operation of the converter station valve hall through a unified interface.

[0070] Example:

[0071] Data Collection: A sensor array deployed in the valve hall of a converter station collects operating parameters such as temperature, current, vibration, and humidity in real time at a sampling frequency of 1 Hz for 30 days. The raw data collected contains high-dimensional, multimodal information, and some data contains noise and missing information. The key features obtained are:

[0072] Temperature rise rate characteristics: under normal conditions, the temperature rise rate is ≤2℃ / minute, and under abnormal conditions, the temperature rise rate is >5℃ / minute; vibration amplitude characteristics: under normal operation, the vibration amplitude is less than 0.3g, and in the event of mechanical failure, the vibration amplitude exceeds 0.7g.

[0073] The weights are determined based on the historical accuracy of the sensors. Specifically, the weight of the temperature sensor is 0.4, the weight of the vibration sensor is 0.35, and the weight of the current sensor is 0.25.

[0074] The fault diagnosis model has been trained to achieve an accuracy of 92%. Testing has shown that when the probability of a fault exceeds 80%, an alert is automatically triggered, and the following response measures are taken:

[0075] 1. Fault Warning: The warning module issues an alarm to inform maintenance personnel of potential equipment problems. Warning information includes equipment type, fault type, warning time, and recommended treatment measures.

[0076] 2. Intelligent diagnosis: Further analyze the possible causes of the fault and intelligently diagnose possible causes of the fault, such as looseness, wear, aging, etc. based on historical data and current equipment status.

[0077] 3. Response measures: After receiving early warning and diagnostic information, corresponding maintenance suggestions will be given, such as "it is recommended to check the vibration source" or "it is recommended to replace worn parts", so that operation and maintenance personnel can respond quickly.

[0078] 4. Maintenance records and analysis: All fault information and maintenance records are automatically saved and a health report of the equipment operation is generated. Operation and maintenance personnel can use these records to carry out subsequent equipment maintenance and optimization.

[0079] During testing, the model successfully predicted three equipment anomalies, providing an hour's advance warning. Operations and maintenance personnel promptly addressed these issues and prevented equipment damage. The system's overall monitoring accuracy reached 90%, with a false alarm rate of less than 5%.

[0080] This invention displays and updates real-time operating parameters within the converter station valve hall. Based on analysis of real-time data and predictive models, it automatically identifies and diagnoses equipment faults and generates fault warnings. Status reports from converter station valve hall equipment help maintenance personnel understand the long-term operating trends of equipment. By integrating multi-dimensional data acquisition, data preprocessing, intelligent analysis, and data fusion, it enables real-time monitoring of equipment operating status, accurately predicts potential faults, and significantly improves the stability and safety of the power system.

[0081] The present invention also proposes a data fusion-based intelligent monitoring system for converter station valve halls, comprising a data acquisition module, a data feature set acquisition module, a monitoring module, and a computer program executable on a processor. It should be noted that each module in the aforementioned system corresponds to a specific step of the method provided in the present invention, and possesses the corresponding functional modules and beneficial effects of the method. For technical details not fully described in this embodiment, please refer to the method provided in the present invention.

[0082] An embodiment of the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable by the processor. It should be noted that when the processor executes the computer program, it corresponds to the specific steps of the method provided in the embodiment of the present invention and has the corresponding functional modules and beneficial effects of the method. For technical details not fully described in this embodiment, please refer to the method provided in the embodiment of the present invention.

[0083] The present invention also provides a computer-readable storage medium storing a computer program. It should be noted that when executed by a processor, the computer program corresponds to the specific steps of the method provided in the present invention and has the corresponding functional modules and beneficial effects. For technical details not fully described in this embodiment, please refer to the method provided in the present invention.

[0084] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A converter station valve hall intelligent monitoring method based on data fusion, characterized in that: include: S1. Use the sensor group to collect the operating parameters of the converter station valve hall in real time and obtain original multi-source data; S2. Preprocessing the original multi-source data in step S1 to obtain preprocessed operating parameters and extract corresponding key features; S3, performing multimodal data fusion on the key features in step S2 to obtain a data feature set; S4. Analyze the data feature set in step S3 using a machine learning algorithm to obtain a classification result of the device status; S5. Based on the result obtained in step S4, if the equipment is in a fault state, the early warning mechanism is triggered and a status report of the converter station valve hall equipment is generated to realize intelligent monitoring of the converter station valve hall.

2. The intelligent monitoring method for converter station valve hall based on data fusion according to claim 1 is characterized in that: In step S1, the sensor group includes a voltage sensor, a current sensor, a temperature sensor, a pressure sensor, a vibration sensor, and a humidity sensor; The voltage sensor is used to measure the voltage level of the equipment in the valve hall of the converter station; the current sensor is used to collect the current data in the valve hall of the converter station; the temperature sensor is used to collect the temperature data in the valve hall of the converter station; the pressure sensor is used to collect the working status of the hydraulic or pneumatic equipment in the valve hall of the converter station; the vibration sensor is used to collect the vibration data of the equipment or mechanical system in the valve hall of the converter station; the humidity sensor is used to collect the humidity data in the valve hall of the converter station.

3. The intelligent monitoring method for converter station valve hall based on data fusion according to claim 2 is characterized in that: In step S2, the original multi-source data is cleaned, normalized, and time-series aligned, and denoised using a digital filter to obtain pre-processed operating parameters; The temperature rise rate and thermal stability are extracted from the preprocessed temperature and humidity data; the root mean square value of vibration acceleration, main frequency distribution, impact coefficient and kurtosis are extracted from the preprocessed vibration data; the current peak value, voltage fluctuation rate and overload duration are extracted from the preprocessed current and voltage data; the pressure fluctuation amplitude, pressure peak duration, pressure change rate and abnormal fluctuation frequency are extracted from the working status of the preprocessed hydraulic or pneumatic equipment.

4. The intelligent monitoring method for converter station valve hall based on data fusion according to claim 1 is characterized in that: In step S3, the data feature set obtained includes the following: The key features are weighted and fused to obtain the data feature set; the formula for assigning weights is: Among them, ω i represents the weight of the i-th sensor data, α represents the weight attenuation coefficient, D i represents the error index of the i-th sensor data; The data feature set includes temperature fusion features, current fusion features, voltage fusion features, partial discharge fusion features, vibration fusion features and humidity / air pressure features.

5. The intelligent monitoring method for converter station valve hall based on data fusion according to claim 4 is characterized in that: In step S4, the classification result of the device status includes the following: Use machine learning algorithms to build fault diagnosis models; The training data includes historical sensor data under normal operating conditions and various fault conditions, and is labeled as normal or faulty equipment status. Based on the labeled training data, a fault diagnosis model is trained using a cross-validation method to obtain a trained fault diagnosis model, which includes an input layer, a hidden layer, and an output layer. The data feature set is processed using time domain analysis and frequency domain analysis methods to extract preliminary features of the equipment status. Statistical methods are used to identify abnormal points in the preliminary features to obtain a preliminary feature set of the equipment status. The preliminary feature set is input into the input layer of the trained fault diagnosis model. After nonlinear mapping and feature combination processing in the hidden layer, deep features are extracted through the operation of weight matrix and activation function to obtain the classification result of the equipment status. The output layer performs probability scoring on the classification result and selects the category with the highest probability as the final classification result. When the probability of the final classification result is greater than or equal to the fault threshold and at least two features in the data feature set are greater than the abnormality threshold, it indicates that the device is in a fault state; The abnormal thresholds of the characteristics are as follows: the abnormal threshold of the temperature fusion characteristic is 75°C, the abnormal threshold of the current fusion characteristic is ±15% of the normal rated value, the abnormal threshold of the vibration fusion characteristic is 1.5g, and the abnormal threshold of the partial discharge fusion characteristic is 100pC; where g represents the acceleration of gravity.

6. The intelligent monitoring method for converter station valve hall based on data fusion according to claim 5 is characterized in that: The failure threshold is 0.

85.

7. The intelligent monitoring method for converter station valve hall based on data fusion according to claim 1 is characterized in that: Also includes: The graphical interface of the visual monitoring platform displays the equipment's operating parameters, fault warning information, historical data and equipment health status scores in real time.

8. A system for the intelligent monitoring method for converter station valve hall based on data fusion according to claim 1, characterized in that: include: The data acquisition module is used to collect the operating parameters of the converter station valve hall in real time using a sensor group to obtain original multi-source data; The data feature set acquisition module is used to preprocess the original multi-source data, obtain the preprocessed operating parameters, and extract the corresponding key features; perform multimodal data fusion on the key features to obtain the data feature set; The monitoring module is used to analyze the data feature set using machine learning algorithms to obtain classification results of the equipment status; If the equipment is in a fault state, the early warning mechanism will be triggered and a status report of the converter station valve hall equipment will be generated, realizing intelligent monitoring of the converter station valve hall.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the converter station valve hall intelligent monitoring method based on data fusion according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the converter station valve hall intelligent monitoring method based on data fusion according to any one of claims 1 to 7 is executed.

Images