System for sensing information fusion state of high-voltage wire inlet cavity of underground coal mining machine

Abstract

The invention discloses an information fusion state sensing system for a high-voltage wire inlet cavity of an underground coal mining machine. The information fusion state sensing system comprises an intelligent equipment sensing layer, an edge equipment and model layer, a model and calculation layer and an intelligent service layer. A reserved port of a coal mining machine network communication switch installed in a wall body adjacent to a wire inlet cavity is fully utilized, an anti-electromagnetic interference video and infrared image monitor is installed in the wire inlet cavity, three kinds of partial discharge signals of ultrahigh frequency, ultrasonic waves and transient ground potential are collected at the same time, and the partial discharge signals are acquired through space image and infrared temperature field mode recognition. Through information fusion processing of partial discharge signals and fault hidden danger early warning, state sensing and fault early warning in the harsh anti-explosion environment are achieved for the first time, and a foundation is laid for intelligent sensing and life cycle management of the coal mining machine.

Description

Technical Field

[0001] The present invention relates to the field of coal mine automation and intelligent technology, and more specifically, to an information fusion state perception system for a high-voltage inlet cavity of an underground coal mining machine. Background Art

[0002] Industrial monitoring systems are key technical support to ensure safe and efficient operation of industrial production processes. With the continuous improvement of industrial automation and intelligence, the demand for real-time monitoring of key equipment in the production process is also growing. Especially in the operation of heavy machinery such as coal mining machines, the high-voltage cavity of the incoming cable is a key part of power transmission, and its safety status is directly related to the stability of the entire production line. An effective monitoring system can monitor the status of equipment in real time, detect and warn of potential faults and abnormalities in a timely manner, thereby avoiding production interruptions, equipment damage and even safety accidents, and ensuring personnel safety and production efficiency. As the core equipment for coal mining, the safe and stable operation of the high-voltage cavity of the incoming cable of the coal mining machine is crucial.

[0003] Traditional monitoring methods have limitations and cannot obtain real-time and comprehensive status information inside the high-pressure chamber. Although existing industrial monitoring systems can monitor equipment status to a certain extent, most systems still have some limitations: (1) Data collection is not comprehensive. Many systems can only monitor some key parameters and cannot fully reflect the operating status of the equipment. (2) Data processing capabilities are limited. Existing systems often use traditional algorithms for data processing and analysis, which is difficult to cope with real-time processing and in-depth analysis of large-scale data. (3) The degree of visualization is not high, the data presentation method is single, and the equipment status cannot be intuitively displayed, making it difficult for operators to make quick and accurate judgments. (4) The early warning mechanism is not accurate enough. The early warning mechanism of some systems relies too much on threshold judgment and lacks intelligent analysis, which may cause false alarms or missed alarms. (5) Maintenance is inconvenient. System maintenance is complex and has high technical requirements for operators, which is not conducive to rapid response and problem handling.

[0004] Therefore, in order to improve the reliability and safety of coal mining machines and reduce downtime due to failures, it is necessary to develop an advanced information fusion status perception system for the high-voltage inlet cavity of underground coal mining machines to ensure the reliability of power supply and operation of the equipment. Summary of the invention

[0005] The purpose of the present invention is to propose an information fusion state perception system for the high-pressure inlet cavity of an underground coal mining machine to solve the above-mentioned technical problems.

[0006] To achieve the above purpose, the present invention provides an underground coal mining machine high-voltage inlet cavity information fusion state perception system, which aims to solve the limitations of traditional monitoring methods in coal mine operations and improve the safety and operating efficiency of coal mining machines. The system realizes comprehensive monitoring and intelligent management of the underground coal mining machine high-voltage inlet cavity through a multi-level structural design.

[0007] Intelligent device perception layer: This layer collects multi-dimensional data of the high-voltage cavity of the incoming cable of the coal mining machine in real time by integrating multiple sensors (temperature, humidity, partial discharge sensors, etc.) and cameras. The selection and arrangement of sensors ensures comprehensive monitoring of key parameters and provides a basis for subsequent data analysis.

[0008] Edge device and model layer: This layer includes edge devices, lightweight digital twin models, data fusion models, and computing models. Edge devices collect data and control feedback through the industrial bus protocol and the intelligent device perception layer to ensure the real-time and accuracy of the data. The model layer is responsible for the preliminary processing and fusion of the collected data to provide high-quality data input for the model and computing layer.

[0009] Model and calculation layer: This layer uses an improved weighted average formula to perform weighted fusion on the data from multiple sensors, aiming to improve the accuracy and reliability of data fusion by dynamically adjusting the weights. The formula is as follows:

[0010]

[0011] In the formula, is the output value after fusion, x i is the measurement value of the ith sensor, w i (t) is the dynamic weight at time t, n is the number of sensors;

[0012] Dynamic weight w i The calculation formula of (t) is:

[0013]

[0014] In the formula, is the variance of sensor i at time t, reflecting the reliability of sensor measurement; is the sum of the inverses of all sensor variances and is used to normalize the weights.

[0015] This layer also includes a fault prediction model, which performs fault pattern identification and risk assessment based on the fused data to ensure that the system can detect potential faults in a timely manner and issue early warnings.

[0016] Intelligent service layer: This layer provides user interface and data visualization services, analyzes and displays historical data, and realizes real-time monitoring and fault warning. Users can intuitively monitor the status of equipment through the interface of this layer, and dynamically set safety thresholds according to equipment characteristics and environmental conditions.

[0017] The technical innovation of the present invention is mainly reflected in several key aspects. First, through the data fusion of multiple sensors and cameras, multi-dimensional data collection of the high-voltage chamber of the incoming cable of the coal mining machine is realized, including key parameters such as temperature, humidity, and partial discharge, which overcomes the limitation of incomplete data collection of traditional monitoring methods and realizes comprehensive monitoring of the internal state of the high-voltage chamber. Partial discharge currently uses three methods: ultra-high frequency, ground wave, and ultrasonic wave to monitor the discharge amount and discharge frequency. After data collection and processing, it is transmitted to the system once every 5 minutes through Modbus-tcp. The data processing module adopts advanced data fusion algorithms, combined with time series analysis and machine learning technology, integrates data from different sources, forms a unified data set, and provides accurate analysis of the equipment status. Among them, an improved weighted average algorithm is used to dynamically adjust the weight of sensor data to improve the accuracy and reliability of data fusion, and effectively reduce the possibility of false alarms and missed alarms. In addition, the system has an intelligent early warning mechanism that can automatically adjust the early warning parameters according to the set rules and thresholds to ensure a rapid response when an abnormality is detected, thereby minimizing the impact of potential faults. Finally, the three-dimensional model of the high-pressure chamber constructed through modern WebGL technology realizes the visualization of data, allowing users to intuitively monitor the status of the equipment, thereby improving the decision-making efficiency of operators.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] The present invention realizes multi-dimensional data collection of the high-voltage cavity of the incoming cable of the coal mining machine by integrating multiple sensors and cameras, including key parameters such as temperature, humidity, and partial discharge. Data fusion and intelligent algorithms are used to improve the accuracy and real-time performance of data processing, and provide support for accurate judgment of the equipment status. It is helpful to improve the automation level and intelligence level of coal mine production and promote technological progress in the industry. It is also of great significance to protect the life safety and physical health of coal miners. Using three-dimensional visualization technology, the monitoring data is displayed in the form of an intuitive three-dimensional model to improve user experience and operational convenience. Combined with real-time data and historical data analysis, the warning threshold is intelligently set to reduce false alarms and missed alarms, improve the accuracy and reliability of the warning, provide a strong guarantee for the safe operation of the coal mining machine, and reduce the accident rate. The friendly user interface and automated maintenance strategy reduce the difficulty of system maintenance and improve the ease of use and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is the overall research architecture diagram of the system; DETAILED DESCRIPTION

[0021] The following is a clear and complete description of the technical solution in the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] This embodiment proposes an underground coal mining machine high-pressure inlet cavity information fusion state perception system as follows:

[0023] Hardware environment: The server is equipped with high-performance processors, large-capacity memory and high-speed storage devices to meet the real-time processing requirements of large amounts of data. The client is deployed on a mainstream industrial monitoring workstation with a stable Windows 10 operating system.

[0024] Network environment: Stable industrial Ethernet is used to ensure the real-time and reliability of data transmission. The network bandwidth meets the transmission requirements of high-definition video and large amounts of monitoring data.

[0025] Software environment: C# / .NET Core development environment and database management system (SQLite or MySQL) are installed on the server. The browser environment supported by TypeScript / React framework is installed on the client. SQLite or MySQL is selected as the system database to store monitoring data and history records.

[0026] In terms of sensor selection, this system can use Hikvision binocular cameras, flat rail housing 485 temperature and humidity sensors, and switch cabinet intelligent three-in-one sensors. Hikvision binocular cameras have the characteristics of high resolution and high sensitivity, and can adapt to various lighting conditions in industrial environments. Each binocular camera provides normal images and infrared imaging, providing two different data perspectives for monitoring, which helps to understand the state of the high-voltage cavity more comprehensively. The camera communicates with the system through the Modbus-TCP protocol, and transmits the average temperature, minimum temperature and maximum temperature data in real time to ensure the real-time and reliability of data transmission. The flat rail housing 485 temperature and humidity sensor can be built-in, compact, and has a standard DIN35 rail mounting card, which can be directly used for rail installation. 485 communication interface, standard ModBus-RTU communication protocol, communication address and baud rate can be set. With pluggable terminals, installation and maintenance are simple and convenient. 5-30V DC wide voltage power supply. Realize real-time monitoring of temperature and humidity parameters inside the high-voltage cavity. The switch cabinet intelligent three-in-one sensor is particularly suitable for status monitoring of electrical equipment such as switch cabinets. The sensor can collect ultra-high frequency signals, ultrasonic signals and geomagnetic wave signals at the same time. It has high sensitivity and versatility. Comprehensive monitoring of partial discharge provides strong support for the safe operation and fault warning of power equipment. By integrating multiple sensors and cameras, multi-dimensional data collection of the high-voltage cavity of the incoming cable of the coal mining machine is realized, including key parameters such as temperature, humidity, and partial discharge. Combined with real-time data and historical data analysis, the warning threshold is intelligently set to reduce false alarms and missed alarms, and improve the accuracy and reliability of warnings.

[0027] The data flow starts from the sensor and camera and is transmitted to the server through the Modbus-TCP protocol. The data processing module on the server analyzes and fuses the data to generate status reports and warning information. This information is then sent to the client, which displays the data to the user in a three-dimensional visual form and provides a user interaction interface to configure system parameters and query historical records. As a data acquisition protocol, Modbus-TCP is widely used in the field of industrial automation to ensure the stability and compatibility of data acquisition. By centrally managing and processing data, multiple types of information are integrated and transmitted, reducing the frequency of data transmission and improving the efficiency of information transmission.

[0028] Partial discharge currently uses three methods: ultra-high frequency, ground wave, and ultrasonic wave to monitor the discharge amount and discharge frequency. After data collection and processing, it is transmitted to the system once every 5 minutes through Modbus-tcp. The data processing module uses advanced data fusion algorithms, combined with time series analysis and machine learning technology, to integrate data from different sources to form a unified data set and provide accurate analysis of equipment status. An improved weighted average algorithm is used to dynamically adjust the weight of sensor data to improve the accuracy and reliability of data fusion and reduce the possibility of false alarms and missed alarms.

[0029] In terms of real-time performance, users can dynamically set safety thresholds for temperature, humidity, and partial discharge based on equipment characteristics and environmental conditions. The system monitors various indicators in real time and immediately triggers an alarm mechanism once data exceeds the set threshold. In addition, the system design ensures a fast response time from detecting an anomaly to issuing an alarm, thereby minimizing the impact of potential failures.

[0030] In the intelligent monitoring system, the pattern recognition technology combining spatial video images and infrared temperature field images can realize accurate monitoring of the environment and equipment status. This technology not only improves the visualization of data, but also enhances the ability to identify abnormal situations. In most traditional systems, the data presentation method is single, and the equipment status cannot be intuitively displayed, which makes it difficult for operators to make quick and accurate judgments. Using three-dimensional visualization technology, the monitoring data is displayed in the form of an intuitive three-dimensional model to improve user experience and operational convenience. The three-dimensional model of the high-voltage chamber is constructed using modern WebGL technologies such as Three.js to realize the visualization of data and high-precision rendering and interaction of the model. The visualization display effect includes three-dimensional mapping of real-time temperature, humidity and partial discharge data, as well as trend analysis and curve display of historical data. The system realizes multi-dimensional view integration through data fusion technology, including information integration, integrating data collected by different sensors and cameras into a unified data set for easy analysis and understanding; state evaluation, integrating information such as temperature, humidity and partial discharge, and comprehensively evaluating the equipment status, thereby improving the accuracy of fault warning; and three-dimensional model mapping, mapping the fused data to the three-dimensional model of the high-voltage chamber, so that users can intuitively understand the relationship between equipment status and environmental parameters.

[0031] After data collection is completed, it is necessary to effectively manage and store the data. It is crucial to classify and store the data collected underground to ensure the durability and reliability of the data. In addition, regular data backup is also a necessary measure to protect the security of important data and prevent data loss caused by system failure. On this basis, it is necessary to implement permission management to ensure that only authorized personnel can access and operate key data, thereby maintaining data security. At the same time, by establishing a regular data synchronization and update mechanism, ensure that the data collected underground can be uploaded to the ground database in real time. Adopting a differentiated update strategy to only process new or modified data can not only reduce the burden of data transmission, but also improve the efficiency of the entire system.

[0032] The system has excellent data analysis capabilities for historical record query. The system's historical record query function records the information of each alarm in detail, including time, over-limit parameters and related images, which is convenient for subsequent analysis. At the same time, the system also provides trend analysis tools, such as curve charts, to help users observe the trend of equipment parameters over time. By mining historical data, the system can predict possible problems of the equipment and provide an important basis for preventive maintenance.

[0033] The above content is merely an illustration of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific structure described. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. The underground coal mining machine high-pressure inlet cavity information fusion state perception system is characterized by: include: The intelligent device perception layer integrates multiple sensors and cameras to collect multi-dimensional data of the high-voltage cavity of the coal mining machine's incoming cable, including key parameters such as temperature, humidity, and partial discharge; The edge device and model layer includes edge devices, lightweight digital twin models, data fusion models, and computing models. This layer collects data and provides control feedback with the intelligent device perception layer through industrial bus protocols and interfaces, and interacts with the model and computing layer through diversified interfaces to perform data and model interaction. The model and computing layer includes a digital twin model and a power-based line monitoring and fault warning model. This layer performs model parameter adjustment with the intelligent service layer through the model parameter adjustment interface, and exchanges data with edge devices and the model layer through the data interface. The intelligent service layer provides user interface and data visualization services, analyzes and displays historical data, realizes real-time monitoring and fault warning, and provides decision support for users.

2. According to the system of claim 1, an improved weighted average formula is used in the model and calculation layer to perform weighted fusion on the data from multiple sensors, aiming to improve the accuracy and reliability of multi-sensor data fusion by dynamically adjusting the weights. The formula is as follows: In the formula, is the output value after fusion, x i is the measurement value of the ith sensor, w i (t) is the dynamic weight at time t, n is the number of sensors; Dynamic weight w i The calculation formula of (t) is: In the formula, is the variance of sensor i at time t, reflecting the reliability of sensor measurement; is the sum of the inverses of all sensor variances and is used to normalize the weights.

3. According to the system of claim 1, the fault warning module can automatically adjust the warning parameters according to the set rules and thresholds to reduce false alarms and missed alarms, and the user can dynamically set the safety thresholds of temperature, humidity and partial discharge according to the equipment characteristics and environmental conditions; the system monitors various indicators in real time, and once it detects that the data exceeds the threshold, the alarm mechanism is immediately triggered; the system design ensures a fast response time from detecting an abnormality to issuing an alarm, minimizing the impact of potential faults.

4. The system according to claim 1, wherein the three-dimensional visualization unit is used to display the processed data in the form of a three-dimensional model, so that users can intuitively monitor the status of the equipment, and use modern WebGL technologies such as Three.js to build a three-dimensional model of the high-voltage chamber to achieve visual display of data and high-precision rendering and interaction of the model; the visual display effect includes three-dimensional mapping of real-time temperature, humidity and partial discharge data, as well as trend analysis and curve display of historical data.

5. The system according to claim 1, characterized in that The model training process of the system includes: sorting the data set according to specific standards to improve the efficiency of data loading and processing; using the optimized parameter adjustment strategy to dynamically adjust the learning rate and batch size in each competitive cycle to achieve faster convergence; applying the improved batch loss calculation method to ensure the stability and robustness of the training process through accumulation and averaging operations; periodically calling the validation set during the training process to monitor the model performance in real time to ensure the generalization ability of the model on different data sets; thereby generating an improved training model file, so that the system can operate effectively in the complex environment of coal mine underground.

Images