Power plant remote monitoring and fault diagnosis system based on 5G network

Abstract

The invention discloses a power plant remote monitoring and fault diagnosis system based on a 5G network, and belongs to the technical field of remote monitoring, fault diagnosis and operation control. Comprising an on-site intelligent sensing and control terminal, a cloud intelligent operation and maintenance platform and a remote operation and display terminal. According to the invention, the characteristics of high bandwidth and ultra-low delay of the 5G network are fully utilized, and real-time acquisition, edge early warning analysis and high-speed transmission of multi-dimensional data of the power plant are realized; fusion of big data and artificial intelligence of a cloud intelligent operation and maintenance platform is provided, and accurate monitoring of the operation state of the power plant equipment, intelligent early warning of abnormity, rapid diagnosis of a fault source, real-time closed-loop execution of a remote control instruction and continuous optimization of a full-life-cycle predictive maintenance strategy can be realized. According to the invention, the safety, the stability and the operation and maintenance management efficiency of power plant operation are obviously improved, and the operation cost is effectively reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of remote monitoring, fault diagnosis and operation control, and particularly relates to a power plant remote monitoring and fault diagnosis system based on a 5G network. BACKGROUND

[0002] In modern industry, as a core component of national energy infrastructure, the safety and stability of the efficient operation of power plant equipment is of great importance. With the rapid development of industrial automation and information technology, the demand for real-time monitoring, state evaluation and fault diagnosis of power plant equipment is increasingly urgent. Traditional power plant equipment monitoring relies on on-site inspection or time sequence connection systems, which can provide production operation guarantee, but in the face of complex and dispersed equipment layout, as well as the demand for higher response speed and massive data transmission, the existing technology faces many challenges.

[0003] Currently, the widely used remote monitoring and diagnosis technology still has universal technical defects. The existing wireless communication technology often shows problems such as insufficient bandwidth and transmission delay when facing the high-speed and high-reliability transmission demand of power plant massive data, and it is difficult to meet the demand of real-time control and complex fault diagnosis. In addition, the existing diagnosis system still needs to be improved in terms of the accuracy of diagnosis, the timeliness of prediction and the refinement degree of remote control when dealing with dynamically changing equipment states and complex fault modes. SUMMARY

[0004] To solve the above problems, the present application provides a power plant remote monitoring and fault diagnosis system based on a 5G network, which adopts a connection mode of on-site intelligent sensing and control terminal, cloud intelligent operation and maintenance platform and remote operation and display terminal through a 5G network, and can realize real-time acquisition of power plant multi-dimensional data, edge early warning analysis and high-speed transmission, as well as accurate monitoring of power plant equipment operation state, intelligent abnormal early warning, rapid diagnosis of fault root cause, real-time closed-loop execution of remote control instruction, significantly improving the safety, stability and operation and maintenance efficiency of power plant operation, and effectively reducing the operation cost.

[0005] The above-mentioned object can be achieved by the following scheme: The utility model provides a kind of power plant remote monitoring and fault diagnosis system based on 5G network, including field intelligent perception and control terminal, cloud intelligent operation and maintenance platform, remote operation and display terminal, three are connected by 5G network;The field intelligent perception and control terminal is installed in the power plant, for the real-time acquisition of power plant field data and is transmitted to cloud intelligent operation and maintenance platform by 5G network at high speed, and executes remote control instruction;The cloud intelligent operation and maintenance platform is used to receive and process data from the field intelligent perception and control terminal, carries out intelligent fault diagnosis, prediction and early warning, generates diagnosis result and prediction and early warning information, while, according to diagnosis result and prediction and early warning information, generate accurate remote control instruction;The remote operation and display terminal is used to provide intuitive power plant operation state display, fault information report for power plant related personnel, authenticate and transmit the high-precision remote control instruction generated by cloud intelligent operation and maintenance platform, and synchronously present the feedback information of remote control instruction execution.

[0006] Optionally, the system further comprises: full life cycle predictive maintenance, using the diagnosis result and prediction and early warning information and feedback information, combined with the historical operation data and maintenance record of power plant equipment, intelligent assessment of the remaining life and fault risk of equipment, and automatically generate maintenance plan and optimization suggestion;5G network slicing technology, for power plant remote monitoring, fault diagnosis and remote control different service demand allocates independent logic network resource and customization network capability, makes remote control instruction ultra-low delay and high reliable transmission, real-time convergence of mass sensor data and stable return of high-definition video stream.

[0007] Optionally, the field intelligent perception and control terminal has data acquisition unit, built-in edge computing server and high-speed 5G communication interface, wherein: the data acquisition unit is configured with high-precision multi-source sensor array, for collecting multi-dimensional key data generated during the operation of power plant equipment to obtain local raw data;The built-in edge computing server can perform local preliminary analysis and rapid judgment on the local raw data, generate preliminary processed data, and also be used for high-precision execution of remote control instructions from the remote operation and display terminal;The high-speed 5G communication interface uses the high bandwidth and ultra-low delay characteristics of 5G network to efficiently transmit the preliminary processed data to the cloud intelligent operation and maintenance platform and synchronously receive remote control instructions from the remote operation and display terminal.

[0008] Optionally, the cloud intelligent operation and maintenance platform has a big data processing and analysis module, an intelligent fault diagnosis and prediction module, and a precise control instruction generation module, wherein: the big data processing and analysis module receives preliminary processing data from the field intelligent sensing and control terminal, and performs real-time aggregation, cleaning, fusion, and deep analysis of data to generate key operation characteristics and abnormal information of the power plant equipment; the intelligent fault diagnosis and prediction module evaluates the operation health status of the power plant equipment, predicts potential faults, and warns of abnormal situations based on the key operation characteristics and abnormal information of the power plant equipment, historical operation data, and artificial intelligence algorithms, to obtain diagnosis results and prediction warning information of the power plant equipment; and the precise control instruction generation module automatically generates high-precision remote control instructions according to the diagnosis results and prediction warning information of the power plant equipment and a preset control strategy.

[0009] Optionally, the remote operation and display terminal has a multi-modal information presentation module, a visual interactive interface, and a safety instruction issuing mechanism, wherein: the multi-modal information presentation module is used to dynamically display the operation status, performance indicators, and abnormal alarm information of the power plant equipment in various forms such as charts, real-time video streams, and voice alarms; the visual interactive interface supports power plant personnel to review detailed fault diagnosis reports and prediction analysis results through intuitive graphical operations, and remotely operate the power plant equipment; and the safety instruction issuing mechanism is used to perform identity authentication, permission verification, and encrypted transmission on the high-precision remote control instructions generated by the cloud intelligent operation and maintenance platform, to ensure the safety and reliability of the instructions, and synchronously present instruction execution feedback information.

[0010] Optionally, the data acquisition unit has adaptive sampling capability, and can dynamically adjust the frequency, granularity, and data type of data acquisition according to the real-time operation status of the power plant equipment, preliminary processing data of the built-in edge computing server, or remote control instructions from the remote operation and display terminal, to quickly obtain high-density and high-timeliness diagnostic data in the presence of abnormalities or warnings.

[0011] Optionally, the full-life-cycle predictive maintenance function uses machine learning, optimization algorithms, and reinforcement learning techniques to intelligently schedule resources and optimize scheduling of generated maintenance plans and optimization suggestions, and adaptively adjusts and learns according to actual maintenance effects, so as to continuously optimize power plant equipment operation and maintenance strategies and maximize cost-effectiveness.

[0012] Optionally, the intelligent fault diagnosis and prediction module has model self-adaptive optimization capability, and can automatically iterate and optimize diagnosis models and prediction algorithms based on the latest data continuously received by the big data processing and analysis module and confirmed fault cases, to improve the accuracy of power plant equipment fault diagnosis and the timeliness of prediction, and adapt to dynamic changes in power plant equipment operation characteristics and fault modes.

[0013] Based on the same inventive concept, the application also provides a power plant remote monitoring and fault diagnosis method based on a 5G network, which comprises: collecting local raw data of power plant equipment operation in real time, and using a field intelligent sensing and control terminal to perform local pre-analysis on the local raw data to generate preliminary processed data; a cloud intelligent operation and maintenance platform performs deep analysis on the preliminary processed data, and based on historical operation data and artificial intelligence algorithms, evaluates the operation health status of the power plant equipment to obtain a diagnosis result and prediction and early warning information; according to the diagnosis result and prediction and early warning information, a remote control instruction is automatically generated, and the remote control instruction is authenticated and transmitted through a remote operation and display terminal; the field intelligent sensing and control terminal receives and executes the remote control instruction to perform real-time closed-loop remote adjustment on the power plant equipment; the remote operation and display terminal dynamically displays the operation status, performance indicators and abnormal alarm information of the power plant equipment, and supports the relevant personnel of the power plant to remotely operate the power plant equipment, while presenting feedback information of the remote control instruction execution; the diagnosis result and prediction and early warning information and the feedback information are used in combination with historical operation data and maintenance records of the power plant equipment to evaluate the remaining life and fault risk, and a maintenance plan and optimization suggestion are generated.

[0014] 1、The application makes full use of the high bandwidth and ultra-low delay characteristics of the 5G network, and constructs a high-speed and reliable communication link between the field intelligent sensing and control terminal, the cloud intelligent operation and maintenance platform and the remote operation and display terminal. This significantly solves the bandwidth and transmission delay problems existing in the massive data transmission of the existing wireless communication technology in the power plant, ensures the real-time collection of multi-dimensional data, edge early warning analysis and high-speed transmission, provides a solid foundation for real-time control and complex fault diagnosis of the power plant, and greatly improves the data transmission efficiency and system response speed.

[0015] 2、The application proposes a strategy of fusion of cloud intelligent operation and maintenance platform big data and artificial intelligence, realizes accurate monitoring of power plant equipment operation status, intelligent early warning of abnormalities and rapid diagnosis of fault sources. Through big data processing and analysis, intelligent fault diagnosis and prediction module, the system can evaluate the equipment health status, predict potential faults, and automatically generate accurate remote control instructions, effectively making up for the shortcomings of the existing diagnosis system in accuracy, timeliness and refinement, and significantly improving the intelligent level and decision efficiency of power plant operation and maintenance.

[0016] 3、The application realizes real-time closed-loop execution of remote control instructions and continuous optimization of full-life-cycle predictive maintenance strategy. Through the safety instruction issuing mechanism of remote operation and display terminal and the use of diagnostic results, prediction and early warning information and feedback information, the system can intelligently evaluate the remaining life of the equipment, generate a maintenance plan and perform resource scheduling and scheduling optimization. This not only changes traditional passive maintenance to active predictive maintenance, greatly reducing the equipment failure rate and operating cost, but also ensures continuous improvement of operation and maintenance strategies through model adaptive optimization capability, significantly improving the safety, stability and operation and maintenance management efficiency of power plant operation.

[0017] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art upon examination of the following or can be learned by practice of the application. The objects and other advantages of the application can be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0019] Figure 1 is a structural schematic diagram of a power plant remote monitoring and fault diagnosis system based on a 5G network according to an embodiment of the present application.

[0020] Figure 2 is a trend curve diagram of the health degree of power plant equipment changing with time according to an embodiment of the present application.

[0021] Figure 3 is a power plant sensor data correlation heat map according to an embodiment of the present application.

[0022] Figure 4 is a remote control instruction execution time delay and statistical analysis diagram according to an embodiment of the present application.

[0023] Figure 5 is a flowchart of a power plant remote monitoring and fault diagnosis method based on a 5G network according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0025] With reference to Figure 1 One embodiment of the present application proposes a power plant remote monitoring and fault diagnosis system based on a 5G network. The system adopts a connection mode of a field intelligent sensing and control terminal, a cloud intelligent operation and maintenance platform and a remote operation and display terminal through a 5G network, can realize real-time collection of multi-dimensional data of a power plant, edge early warning analysis and high-speed transmission, and can realize accurate monitoring of the running state of power plant equipment, intelligent early warning of abnormalities, rapid diagnosis of fault sources and real-time closed-loop execution of remote control instructions, significantly improving the safety, stability and operation and maintenance efficiency of the power plant operation and effectively reducing the operation cost.

[0026] The system of the embodiment specifically comprises: a field intelligent sensing and control terminal, a cloud intelligent operation and maintenance platform and a remote operation and display terminal, which are connected through a 5G network, The field intelligent sensing and control terminal is installed in the power plant, is used for real-time collection of field data of the power plant and high-speed transmission of the field data to the cloud intelligent operation and maintenance platform through the 5G network, and executes remote control instructions; Specifically, the field intelligent sensing and control terminal is deployed in the core equipment area of the power plant, for example, near key equipment such as a generator set, a transformer, a boiler, a steam turbine and an auxiliary system. It obtains multi-dimensional data generated in the running process of the power plant equipment in real time by integrating various high-precision sensors such as temperature sensors, pressure sensors, vibration sensors, current sensors, voltage sensors and flow sensors. After preliminary processing, the data are transmitted to the remote cloud intelligent operation and maintenance platform efficiently and stably by using the high bandwidth and ultra-low delay characteristics of the 5G network through the built-in high-speed 5G communication interface. At the same time, the terminal can also receive and accurately execute control instructions from the remote operation and display terminal, realizing real-time adjustment and fault processing of the power plant equipment.

[0027] The cloud intelligent operation and maintenance platform is used for receiving and processing data from the field intelligent sensing and control terminal, performing intelligent fault diagnosis and prediction and early warning, generating diagnosis results and prediction and early warning information, and generating accurate remote control instructions according to the diagnosis results and the prediction and early warning information; Specifically, the cloud-based intelligent operation and maintenance platform receives and processes power plant data transmitted from the on-site intelligent sensing and control terminals. The platform intelligently analyzes these data to identify the operating status, potential problems, and abnormal conditions of the power plant equipment. Based on these analysis results, the platform can perform fault diagnosis and predictive warning, generating detailed diagnostic results and corresponding warning information. Meanwhile, according to these diagnostic results and warning information, the platform will automatically generate precise remote control instructions for remotely adjusting and intervening in the power plant equipment.

[0028] The remote operation and display terminal is used to provide power plant personnel with intuitive power plant operating status display and fault information reporting, authenticate and transmit high-precision remote control instructions generated by the cloud-based intelligent operation and maintenance platform, and synchronously present feedback information of remote control instruction execution.

[0029] Specifically, the remote operation and display terminal, as the core interface for human-computer interaction, can display information to power plant personnel in various forms. It provides intuitive power plant operating status display, including real-time performance indicators of equipment, key parameter trend charts, and overall operation overview. When the system detects abnormalities or faults, the terminal generates detailed fault information reports, including fault types, occurrence times, possible causes, and recommended handling measures. In addition, the terminal is responsible for authenticating and transmitting high-precision remote control instructions generated by the cloud-based intelligent operation and maintenance platform. This includes identity authentication and permission verification to ensure that only authorized personnel can approve and issue control instructions, and encrypted transmission to ensure the security of the instructions. After the instructions are issued, the terminal synchronously presents the execution feedback information of the remote control instructions, such as whether the instructions are successfully executed and the impact of the execution results on the equipment status, thereby forming a complete closed-loop control chain. Power plant personnel can review detailed fault diagnosis reports, predictive analysis results, and remotely operate the power plant equipment through the visual interactive interface of the terminal.

[0030] Optionally, the system further comprises: Full-life-cycle predictive maintenance uses the diagnostic results, predictive warning information, and feedback information, combined with historical operating data and maintenance records of the power plant equipment, to intelligently assess the remaining life and fault risk of the equipment and automatically generate maintenance plans and optimization suggestions.

[0031] Specifically, the full life cycle predictive maintenance function in the embodiment receives the diagnosis results, prediction and early warning information, and execution feedback information of remote control instructions generated by the cloud intelligent operation and maintenance platform, and performs deep fusion analysis with the historical operation data and past maintenance records of the power plant equipment. The core of this function is to intelligently evaluate the remaining life and failure risk of the equipment. The system will quantify the failure risk level of the equipment according to the evaluated remaining life and potential failure mode. On this basis, the predictive maintenance function will automatically generate detailed maintenance plans and optimization suggestions. These plans include the time point of preventive replacement of components, priority ranking of maintenance tasks, spare parts demand estimation, and maintenance personnel scheduling suggestions, such as Figure 2 As shown in the figure, the trend curve of the health degree of the power plant equipment over time in the embodiment of the application is shown. The curve dynamically reflects the continuous evaluation and optimization of the equipment operating state by the predictive maintenance function.

[0032] Exemplarily, by analyzing the real-time operation data such as vibration spectrum, temperature curve, current and voltage fluctuation of the power plant equipment, combining with the historical failure mode and maintenance cycle of the equipment, this module can predict the wear degree and fatigue accumulation of key components (such as bearings, blades, transformer windings, etc.), and thus accurately calculate the remaining available time under the current working condition. When the system predicts that the bearing of a certain generator set will reach the wear limit in the next three months, it will immediately generate a maintenance plan containing the recommended replacement time, required spare part model and inventory status, and suggest the best downtime maintenance window to minimize the impact on the power plant production.

[0033] 5G network slicing technology allocates independent logical network resources and customized network capabilities for different service needs of power plant remote monitoring, fault diagnosis and remote control, enabling ultra-low delay and high reliable transmission of remote control instructions, real-time aggregation of massive sensor data, and stable return of high-definition video stream.

[0034] Specifically, the 5G network slicing technology in this embodiment creates multiple independent logical network slices on the basis of the physical 5G network infrastructure to meet the differentiated needs of different business scenarios in the power plant for network performance. For the transmission of remote control instructions, the system allocates a network slice with ultra-low latency and high reliability. This slice has priority transmission bandwidth and extremely low end-to-end latency, ensuring that the power plant remote control instructions can respond within milliseconds and be delivered to the field equipment stably. For the real-time aggregation of massive sensor data, a network slice with large bandwidth and high throughput is allocated, which is specifically used to carry the continuous and massive data stream generated by various sensors in the power plant, ensuring that data can be quickly and completely aggregated from the field to the cloud intelligent operation and maintenance platform for analysis. In addition, for the stable return of high-definition video streams, the system allocates a network slice that optimizes the uplink bandwidth and transmission stability, ensuring that the operation and maintenance personnel can clearly and smoothly watch the real-time high-definition video pictures of the power plant site. Each network slice is logically isolated from each other and does not interfere with each other, thereby ensuring the quality of service and security of different business traffic and significantly improving the overall network support capability of remote monitoring, fault diagnosis and remote control of the power plant.

[0035] For example, emergency shutdown instructions or valve opening adjustment instructions are transmitted through an ultra-low latency slice. A large amount of real-time data stream from temperature, pressure, vibration and other sensors is aggregated to a large bandwidth slice. The operation and maintenance personnel monitor the site through high-definition video, and the video stream is stably returned through a slice that optimizes the uplink bandwidth.

[0036] Optionally, the field intelligent sensing and control terminal has a data acquisition unit, a built-in edge computing server and a high-speed 5G communication interface. The data acquisition unit is configured with a high-precision multi-source sensor array for acquiring multi-dimensional key data generated during the operation of the power plant equipment to obtain local raw data. Specifically, the data acquisition unit integrates multiple types of high-precision sensors to form a multi-source sensor array. These sensors are strategically deployed on key equipment in the power plant, such as generators, transformers, boilers, steam turbines and auxiliary systems, etc. The sensor array can comprehensively and real-time monitor the equipment operating state, including but not limited to temperature, pressure, vibration, current, voltage, flow, liquid level, speed, flue gas composition and acoustic signals, etc. These sensors have high sensitivity and fast response characteristics, and can capture subtle changes and abnormal fluctuations generated during the operation of the equipment. The data acquisition unit continuously acquires these multi-dimensional key data and aggregates the raw, unprocessed measurement data to form local raw data. These local raw data form the basis for subsequent edge computing and cloud intelligent analysis.

[0037] The built-in edge computing server can perform local preliminary analysis and rapid judgment on the local raw data to generate preliminary processed data, and is also used for high-precision execution of remote control instructions from a remote operation and display terminal. Specifically, the built-in edge computing server is deployed inside the on-site intelligent sensing and control terminal, close to the data acquisition unit. It has strong local computing capability and performs preliminary analysis and rapid judgment on the local raw data obtained by the data acquisition unit. This local processing includes data cleaning, denoising, format conversion, feature extraction, and anomaly detection based on preset rules or lightweight models. After preliminary analysis and judgment, the server generates preliminary processed data, which is more refined than the raw data, reduces redundancy, and helps to reduce the transmission burden of the subsequent 5G network and the processing pressure of the cloud intelligent operation and maintenance platform. At the same time, the built-in edge computing server also undertakes the important task of high-precision execution of remote control instructions from a remote operation and display terminal. It directly receives and analyzes authenticated control instructions and converts them into specific operations on power plant equipment, ensuring that the instructions can accurately and timely act on on-site equipment.

[0038] Illustratively, the server can monitor whether the threshold value of sensor data is out of limit or the data change rate is abnormal in real time, thereby rapidly identifying potential equipment failure or operation deviation. It can convert remote control instructions into adjustments of valve opening, changes of motor speed, or start / stop operations of specific equipment units.

[0039] The high-speed 5G communication interface utilizes the high bandwidth and ultra-low delay characteristics of 5G network to efficiently transmit the preliminary processed data to the cloud intelligent operation and maintenance platform and synchronously receive remote control instructions from the remote operation and display terminal.

[0040] Specifically, the high-speed 5G communication interface is a key component for data exchange between the on-site intelligent sensing and control terminal and external networks. It fully utilizes the high bandwidth and ultra-low delay characteristics of 5G network to ensure that preliminary processed data can be efficiently and rapidly transmitted from the on-site to the remote cloud intelligent operation and maintenance platform. This efficient transmission capability means that even in the face of a large amount of real-time monitoring data generated by power plant equipment, the smoothness and integrity of data flow can be guaranteed, avoiding transmission bottlenecks. At the same time, the communication interface also has the ability to synchronously receive remote control instructions from the remote operation and display terminal. Due to the extremely low communication delay provided by 5G network, remote operation instructions can almost instantaneously reach the on-site intelligent sensing and control terminal, thereby ensuring the real-time nature and response speed of remote control. This is crucial for power plant operation regulation and fault handling scenarios that require immediate response.

[0041] Optionally, the cloud intelligent operation and maintenance platform has a big data processing and analysis module, an intelligent fault diagnosis and prediction module, and a precise control instruction generation module, wherein: The big data processing and analysis module receives preliminary processing data from the field intelligent sensing and control terminal, and performs real-time aggregation, cleaning, fusion, and deep analysis of data to generate key operation characteristics and abnormal information of power plant equipment. Specifically, the big data processing and analysis module is the data processing core of the cloud intelligent operation and maintenance platform. It receives and processes preliminary processing data transmitted from the field intelligent sensing and control terminal in real time. After receiving the data, the module performs a series of data processing operations, including real-time aggregation, cleaning, fusion, and deep analysis of data. Through these processes, the module can identify and extract key operation characteristics of power plant equipment from massive data, and discover and generate abnormal information, providing a basis for subsequent intelligent fault diagnosis and prediction, such as Figure 3 As shown, the correlation heat map between multiple sensors of the power plant in the embodiment of the application is shown. The atlas intuitively shows the complex interrelation strength between the sensor data through the depth of color, which helps the big data processing and analysis module to identify key features and perform efficient data fusion.

[0042] For example, the module will collect relevant data from different devices, different sensors, and different time points, remove noise, redundancy, or error information, and integrate data from different sources and different types. By using complex data mining and pattern recognition algorithms, device performance curves, health indicators, or load response characteristics can be identified, as well as small drifts, trend changes, or sudden abnormal events in device parameters.

[0043] The intelligent fault diagnosis and prediction module assesses the operation health status of the power plant equipment, predicts potential faults, and warns of abnormal situations based on the key operation characteristics and abnormal information of the power plant equipment, historical operation data, and artificial intelligence algorithms, to obtain diagnosis results and prediction warning information of the power plant equipment. Specifically, the intelligent fault diagnosis and prediction module is the core intelligent analysis part of the cloud intelligent operation and maintenance platform. It deeply integrates the key operation characteristics and abnormal information of the power plant equipment generated by the big data processing and analysis module, and combines rich historical operation data and advanced artificial intelligence algorithms to comprehensively and real-time assess the operation health status of the power plant equipment. Through continuous learning and pattern recognition, the module can identify subtle deviations and early signs of failure in device operation. One of the core functions of the module is to predict potential faults. By analyzing the historical fault patterns and current operation trends of the device, the module can predict the type and time of future possible faults. For example, the module can use a time series prediction model to predict the remaining life of key components of the device, such as bearings, gears, and transformer windings. represents the remaining life of the equipment component after the usage time t, which can be represented by the following general model: wherein, represents the prediction function, represents the equipment key operating characteristics and abnormal information extracted at time t, represents the equipment historical operating data up to time t, and represents the parameters of the artificial intelligence algorithm model adopted. At the same time, this module is also responsible for early warning of abnormal situations. When the equipment operating state deviates from the normal range, or when a failure is predicted to occur soon, the module will immediately trigger the early warning mechanism. This early warning not only includes simple threshold alarms, but also covers abnormal trend warnings based on complex pattern recognition. For example, the module can calculate the equipment health index in real time , and determine whether early warning is needed by setting a dynamic threshold . When , early warning is triggered.

[0044] wherein, is the number of key operating parameters, is the weight of each parameter, is the normalized value of the i-th parameter. Finally, through the evaluation of the equipment operating health status, the prediction of potential failures and the early warning of abnormal situations, this module outputs detailed power plant equipment diagnosis results and prediction and early warning information. These information includes fault type, occurrence probability, suggested points of attention and warning level, providing timely and accurate basis for power plant operation and maintenance personnel to make decisions.

[0045] The precise control instruction generation module automatically generates high-precision remote control instructions according to the diagnosis results and prediction and early warning information of the power plant equipment and the preset control strategy.

[0046] Specifically, the precise control instruction generation module is the key link for the cloud-based intelligent operation and maintenance platform to realize remote control and intelligent intervention. It takes the diagnosis results and prediction and early warning information of power plant equipment output by the intelligent fault diagnosis and prediction module as the main input basis. When the module receives the evaluation of the device health status, the prediction of potential faults or the early warning of abnormal conditions, it will automatically generate a series of high-precision remote control instructions in combination with the control strategy library pre-stored in the system. These control strategies are pre-set according to the operation specifications of the power plant, the characteristics of the equipment, the safety regulations and the optimization objectives. When generating instructions, the module will consider the current equipment state, the expected fault mode and the optimal intervention path to ensure that the generated instructions can effectively solve the problem without negatively affecting the stable operation of the power plant. The generated remote control instructions can be directly issued to the remote operation and display terminal for authentication and transmission, and finally executed by the on-site intelligent sensing and control terminal.

[0047] Optionally, the remote operation and display terminal has a multi-modal information presentation module, a visual interactive interface and a safety instruction issuing mechanism, wherein: The multi-modal information presentation module is used to dynamically display the running state, performance indicators and abnormal alarm information of the power plant equipment through charts, real-time video streams and voice alarms. Specifically, the multi-modal information presentation module aims to provide comprehensive and intuitive equipment information for power plant personnel. It realizes dynamic and real-time display of the running state of power plant equipment by integrating multiple data presentation methods. The system can display various performance indicators and key parameters of power plant equipment in the form of charts. Secondly, through real-time video streams, operation and maintenance personnel can remotely watch real-time high-definition pictures of the power plant site and visually confirm the appearance of the equipment and the running environment. In addition, the system also has a voice alarm function. When serious abnormalities or emergencies are detected, in addition to screen display, the system will also issue an alarm to the operator in the form of voice broadcast in a timely manner to ensure that critical information is received in the first time. This combination of multiple modes significantly improves the efficiency of information transmission and the response speed in emergency situations, enabling power plant personnel to more quickly and accurately grasp the running state of the equipment and abnormal alarm information.

[0048] Illustratively, the chart can display the power generation, efficiency curve or energy consumption data of the generator set. Real-time video stream can be used to observe whether there are physical abnormalities such as oil leakage, smoke or loose parts in the equipment. Voice alarm can timely remind through voice broadcast when personnel do not continuously focus on the screen.

[0049] The visual interactive interface supports power plant personnel to review detailed fault diagnosis reports, prediction analysis results and remotely operate power plant equipment through intuitive graphical operations. Specifically, the visual interactive interface greatly simplifies the access and control of complex system information for power plant related personnel through intuitive graphical operation design. Users can easily browse and review detailed fault diagnosis reports generated by the cloud intelligent operation and maintenance platform. These reports show the type, location, cause analysis and recommended solutions of the fault in clear diagrams and written instructions. At the same time, the interface also supports viewing predictive analysis results, such as the remaining life curve of the equipment, potential fault trends and warning levels, helping operation and maintenance personnel to make early judgments and develop maintenance plans. More importantly, the interface supports remote operation of power plant equipment by power plant related personnel. Through simple clicking, dragging or other graphical instructions, authorized users can remotely start, stop, parameter adjustment or mode switching of power plant equipment without being on site. This intuitive operation method significantly reduces the complexity of operation, improves operation efficiency and response speed, and ensures the accuracy and safety of operation.

[0050] The security instruction issuing mechanism is used for identity authentication, permission verification and encrypted transmission of high-precision remote control instructions generated by the cloud intelligent operation and maintenance platform, to ensure the security and reliability of the instructions, and to synchronously present instruction execution feedback information.

[0051] Specifically, the security instruction issuing mechanism ensures the security and reliability of high-precision remote control instructions generated from the cloud intelligent operation and maintenance platform during transmission and execution. When the cloud intelligent operation and maintenance platform generates remote control instructions and sends them to the remote operation and display terminal, the mechanism first performs strict identity authentication to verify the identity of the operator, ensuring that only authorized personnel can access and handle the instructions. Permission verification is performed to determine whether the current operator has the right to approve or issue the specific instruction according to the pre-set user roles and permission levels, preventing unauthorized operation. After authentication and verification, the instructions are transmitted in encrypted form, using industry standard encryption algorithms to encode the instruction data, preventing illegal interception, tampering or eavesdropping during 5G network transmission, thus ensuring the confidentiality and integrity of the instructions. After the instructions are successfully issued and executed on site, the mechanism also synchronously presents instruction execution feedback information, including whether the instructions have arrived successfully, whether they have been successfully executed, and the device state changes after execution, forming a complete operation loop, so that the operator can keep abreast of the execution of the instructions, such as Figure 4 As shown, a time delay distribution histogram of remote control instructions from issuance to execution completion in the embodiment of the application and its statistical analysis are shown. The graph reflects the ultra-low delay characteristics and high reliability of instruction transmission supported by 5G network through detailed delay distribution and key statistical quantities.

[0052] Optionally, the data acquisition unit has adaptive sampling capability, which can dynamically adjust the frequency, granularity and data type of data acquisition according to the real-time operation state of power plant equipment, preliminary processing data of the built-in edge computing server or remote control instructions from the remote operation and display terminal, and quickly obtain high-density and high-time-efficiency diagnostic data in the presence of abnormalities or warnings.

[0053] Specifically, the adaptive sampling capability of the data acquisition unit can flexibly optimize the data acquisition process according to the actual situation of power plant operation. This capability is reflected in the ability to dynamically adjust the sampling strategy according to various real-time inputs. It can be adjusted according to the real-time operation state of the equipment. When the equipment is in normal and stable operation mode, the data acquisition unit will use a lower sampling frequency and granularity to reduce data volume and resource consumption. However, as soon as a significant change in equipment load, start / stop process or entry into a critical operation phase is monitored, the data acquisition frequency and granularity will be increased accordingly to capture more detailed state data. The data acquisition unit will adjust according to the preliminary processing data of the built-in edge computing server. When the edge computing server finds potential abnormalities or trends in local preliminary analysis, it will issue instructions to the data acquisition unit to increase the sampling frequency or collect more relevant data types to obtain more comprehensive diagnostic information. Remote control instructions from the remote operation and display terminal can also trigger sampling adjustment. When the maintenance personnel issue specific control instructions through the remote operation and display terminal, the data acquisition unit will immediately respond and switch to a high-density and high-time-efficiency sampling mode to ensure timely and detailed data feedback during control execution. This dynamic adjustment strategy allows the system to save resources during normal operation, and quickly obtain high-density and high-time-efficiency diagnostic data in the presence of abnormalities or warnings, providing more sufficient and accurate data support for fault diagnosis.

[0054] For example, the data acquisition frequency can be dynamically adjusted according to the equipment operation health indicators and warning states : wherein, is the basic sampling frequency, is the frequency increment calculated according to the health indicators and warning states. When decreases or becomes a warning state, significantly increases. At the same time, the data granularity can also be adjusted to obtain more detailed data, such as: wherein, is the default data granularity, The granularity amplification factor is calculated according to the health indicators and the early warning state. These adaptive adjustment mechanisms ensure that the system can operate with the optimal data acquisition strategy under different working conditions, improving the timeliness and accuracy of diagnosis.

[0055] Optionally, the full life cycle predictive maintenance function uses machine learning, optimization algorithms and reinforcement learning techniques to intelligently schedule resources and optimize scheduling for the generated maintenance plans and optimization suggestions, and to adaptively adjust and learn according to the actual maintenance effect, so as to continuously optimize the equipment operation and maintenance strategy of the power plant and maximize the cost-effectiveness.

[0056] Specifically, after generating the preliminary maintenance plans and optimization suggestions, the full life cycle predictive maintenance function further uses advanced machine learning, optimization algorithms and reinforcement learning techniques to intelligently schedule resources and optimize scheduling for these plans. This includes fine allocation and arrangement of human resources, spare parts inventory, tools and equipment, and downtime window. For example, the module can dynamically generate the optimal scheduling scheme according to the urgency, required skills, duration and available resources of different maintenance tasks. In order to realize intelligent scheduling and scheduling optimization of resources, the system will build an optimization model, the objective of which is to minimize the total maintenance cost or maximize the equipment availability. Assuming that we want to optimize the scheduling of multiple maintenance tasks, the objective function (total cost) can be expressed as: where, is the total number of maintenance tasks, is the labor cost of task j, is the spare parts cost of task j, is the downtime loss cost caused by task j. The system will use optimization algorithms such as linear programming or integer programming to solve this problem and find the optimal solution under various constraints. This function also has the ability to adaptively adjust and learn according to the actual maintenance effect. After the execution of maintenance tasks, the system will collect relevant feedback data, including maintenance duration, spare parts consumption, actual cost, and performance after the equipment resumes operation. These feedback data will be input into the machine learning model for continuous optimization of the predictive maintenance model and scheduling algorithm. In particular, through reinforcement learning techniques, the system can learn and improve its maintenance strategy like an intelligent agent in continuous trial and evaluation. The reinforcement learning model adjusts its decision logic by receiving "reward" signals. This adaptive learning mechanism enables the equipment operation and maintenance strategy of the power plant to continuously adapt to new operating conditions and failure modes, thereby continuously optimizing the operation and maintenance strategy and maximizing cost-effectiveness.

[0057] Optionally, the intelligent fault diagnosis and prediction module has a model self-adaptive optimization capability, which can automatically iterate and optimize the diagnosis model and prediction algorithm based on the latest data continuously received by the big data processing and analysis module and the confirmed fault cases, thereby improving the accuracy of power plant equipment fault diagnosis, the timeliness of prediction, and the dynamic changes of power plant equipment operation characteristics and fault modes.

[0058] Specifically, the core of the intelligent fault diagnosis and prediction module lies in its model self-adaptive optimization capability. The module continuously receives the latest data and confirmed real fault cases provided by the big data processing and analysis module as input. Based on this information, the module can automatically iterate and optimize the diagnosis model and prediction algorithm, such as through online learning or incremental learning. For a neural network-based diagnosis model, the module will adjust the model's weights and biases through the backpropagation algorithm according to new data and fault labels to minimize prediction error. The iterative update process of the model can be represented as: wherein, and represent the updated and pre-updated model weights, respectively, is the learning rate, is the gradient of the loss function L with respect to the weight W, is the latest data continuously received, is the set of confirmed fault cases. This continuous optimization mechanism significantly improves the accuracy of power plant equipment fault diagnosis and the timeliness of prediction. More importantly, it enables the system to adapt to the dynamic changes of power plant equipment operation characteristics and fault modes, ensuring that the diagnosis model always maintains accurate judgment of the equipment health status.

[0059] Based on the same inventive concept, as shown in Figure 5 the present application also provides a power plant remote monitoring and fault diagnosis method based on a 5G network, which comprises: real-time acquisition of local raw data of power plant equipment operation, and local pre-analysis of the local raw data by the on-site intelligent sensing and control terminal to generate preliminary processed data; the cloud intelligent operation and maintenance platform performs in-depth analysis on the preliminary processed data, and based on historical operation data and artificial intelligence algorithms, evaluates the operation health status of the power plant equipment to obtain diagnosis results and prediction warning information; according to the diagnosis results and prediction warning information, automatically generating remote control instructions, and authenticating and transmitting the remote control instructions through the remote operation and display terminal; the on-site intelligent sensing and control terminal receives and executes the remote control instructions to real-time closed-loop remote adjust the power plant equipment; The remote operation and display terminal dynamically displays the running state, performance index and abnormal alarm information of the power plant equipment, and supports the remote operation of the power plant equipment by the relevant personnel of the power plant, and presents the feedback information of the remote control instruction execution; By using the diagnosis result, prediction and early warning information and feedback information, combined with the historical running data and maintenance record of the power plant equipment, the remaining life and fault risk are evaluated, and a maintenance plan and optimization suggestion are generated.

[0060] It should be noted that the electrical connection between the above-mentioned various units does not necessarily mean the direct connection of the line, and the indirect connection mode can be applied to the embodiments of the present application as long as the purpose of the present application is achieved. The above-mentioned is only an exemplary embodiment of the present application, and cannot limit the scope of the present application.

[0061] That is, any equivalent changes and modifications made according to the teachings of the present application are still within the scope of the present application. Other embodiments of the present application will be readily apparent to those skilled in the art upon considering the description and practice of the true disclosure. This application is intended to cover any variations, uses or adaptive changes of the present application that follow the general principles of the present application and include common knowledge or conventional technical means in the art not disclosed by the present application.

Claims

1. A power plant remote monitoring and fault diagnosis system based on a 5G network, characterized in that, include: The on-site intelligent sensing and control terminal, the cloud-based intelligent operation and maintenance platform, and the remote operation and display terminal are connected via a 5G network; The on-site intelligent sensing and control terminal is installed in the power plant and is used to collect on-site data in real time and transmit it to the cloud intelligent operation and maintenance platform at high speed through the 5G network, and execute remote control commands. The cloud-based intelligent operation and maintenance platform is used to receive and process data from on-site intelligent sensing and control terminals, perform intelligent fault diagnosis and prediction and early warning, generate diagnostic results and prediction and early warning information, and generate precise remote control commands based on the diagnostic results and prediction and early warning information. The remote operation and display terminal is used to provide power plant personnel with intuitive displays of power plant operating status and fault information reports, to authenticate and transmit high-precision remote control commands generated by the cloud-based intelligent operation and maintenance platform, and to simultaneously present feedback information on the execution of remote control commands.

2. The power plant remote monitoring and fault diagnosis system based on a 5G network according to claim 1, characterized in that, The system also includes: Predictive maintenance throughout the entire lifecycle utilizes the diagnostic results, predictive warning information, and feedback information, combined with historical operating data and maintenance records of power plant equipment, to intelligently assess the remaining lifespan and failure risk of the equipment, and automatically generate maintenance plans and optimization suggestions.

3. The power plant remote monitoring and fault diagnosis system based on a 5G network according to claim 1, characterized in that, The system also includes: 5G network slicing technology allocates independent logical network resources and customized network capabilities to meet the different service needs of power plant remote monitoring, fault diagnosis and remote control, enabling ultra-low latency and high reliability transmission of remote control commands, real-time aggregation of massive sensor data and stable backhaul of high-definition video streams.

4. The power plant remote monitoring and fault diagnosis system based on a 5G network according to claim 1, characterized in that, The on-site intelligent sensing and control terminal has a data acquisition unit, a built-in edge computing server, and a high-speed 5G communication interface, wherein: The data acquisition unit is equipped with a high-precision multi-source sensor array, which is used to collect multi-dimensional key data generated during the operation of power plant equipment to obtain local raw data. The built-in edge computing server can perform local preliminary analysis and quick judgment on the local raw data, generate preliminary processed data, and also execute remote control commands from remote operation and display terminals with high precision. The high-speed 5G communication interface utilizes the high bandwidth and ultra-low latency characteristics of the 5G network to efficiently transmit the preliminary processed data to the cloud-based intelligent operation and maintenance platform, and simultaneously receive remote control commands from the remote operation and display terminal.

5. The power plant remote monitoring and fault diagnosis system based on a 5G network according to claim 1, characterized in that, The cloud-based intelligent operation and maintenance platform includes a big data processing and analysis module, an intelligent fault diagnosis and prediction module, and a precise control command generation module, among which: The big data processing and analysis module receives preliminary processed data from the on-site intelligent sensing and control terminal, and performs real-time data aggregation, cleaning, fusion and in-depth analysis to generate key operating characteristics and abnormal information of power plant equipment. The intelligent fault diagnosis and prediction module assesses the operational health status of the power plant equipment, predicts potential faults, and provides early warnings of abnormal situations based on the key operating characteristics and abnormal information of the power plant equipment, historical operating data, and artificial intelligence algorithms, thereby obtaining the diagnostic results and predictive warning information of the power plant equipment. The precision control command generation module automatically generates high-precision remote control commands based on the diagnostic results and predictive warning information of the power plant equipment and the preset control strategy.

6. The power plant remote monitoring and fault diagnosis system based on a 5G network according to claim 1, characterized in that, The remote operation and display terminal has a multimodal information presentation module, a visual interactive interface, and a secure command issuance mechanism, wherein: The multimodal information presentation module is used to dynamically display the operating status, performance indicators and abnormal alarm information of power plant equipment through various forms such as charts, real-time video streams and voice alarms; The visual interactive interface allows power plant personnel to view detailed fault diagnosis reports and predictive analysis results through intuitive graphical operations, and to remotely operate power plant equipment. The security command issuance mechanism is used to perform identity authentication, permission verification, and encrypted transmission of high-precision remote control commands generated by the cloud-based intelligent operation and maintenance platform, ensuring the security and reliability of the commands, and simultaneously presenting command execution feedback information.

7. The power plant remote monitoring and fault diagnosis system based on a 5G network according to claim 4, characterized in that, The data acquisition unit has adaptive sampling capability, which can dynamically adjust the frequency, granularity and data type of data acquisition according to the real-time operating status of power plant equipment, the preliminary processed data of the built-in edge computing server or the remote control commands from the remote operation and display terminal. When an anomaly or warning occurs, it can quickly acquire high-density and high-timeliness diagnostic data.

8. The power plant remote monitoring and fault diagnosis system based on a 5G network according to claim 2, characterized in that, The aforementioned predictive maintenance throughout the entire lifecycle utilizes machine learning, optimization algorithms, and reinforcement learning techniques to intelligently schedule and optimize resources for the generated maintenance plans and optimization suggestions. It also adaptively adjusts and learns based on actual maintenance results, thereby continuously optimizing the power plant equipment operation and maintenance strategy and maximizing cost-effectiveness.

9. The power plant remote monitoring and fault diagnosis system based on a 5G network according to claim 5, characterized in that, The intelligent fault diagnosis and prediction module has the ability to adaptively optimize the model. Based on the latest data continuously received by the big data processing and analysis module and confirmed fault cases, it can automatically iterate and optimize the diagnostic model and prediction algorithm, improve the accuracy of power plant equipment fault diagnosis and the timeliness of prediction, and adapt to the dynamic changes in the operating characteristics and fault modes of power plant equipment.

10. A method for remote monitoring and fault diagnosis of power plants based on 5G networks, characterized in that, The method includes: Real-time acquisition of local raw data on the operation of power plant equipment, and local pre-analysis of the local raw data using on-site intelligent sensing and control terminals to generate preliminary processed data; The cloud-based intelligent operation and maintenance platform performs in-depth analysis on the preliminary processed data and, based on historical operating data and artificial intelligence algorithms, assesses the operational health status of power plant equipment to obtain diagnostic results and predictive warning information. Based on the diagnostic results and predictive warning information, remote control commands are automatically generated, and the remote control commands are authenticated and transmitted through a remote operation and display terminal. The on-site intelligent sensing and control terminal receives and executes remote control commands to remotely adjust the power plant equipment in real time in a closed loop. The remote operation and display terminal dynamically displays the operating status, performance indicators and abnormal alarm information of the power plant equipment, supports remote operation of the power plant equipment by relevant personnel, and presents feedback information on the execution of remote control commands. Using the diagnostic results, predictive warning information, and feedback information, combined with the historical operating data and maintenance records of the power plant equipment, the remaining lifespan and failure risk are assessed, and maintenance plans and optimization suggestions are generated.

Images