Intelligent monitoring system for operation state of thermal power plant

Abstract

The invention discloses an intelligent monitoring system for the operation state of a thermal power plant, relates to the technical field of thermal power plant monitoring, and aims to solve the problems that a balance point between over-maintenance and under-maintenance of equipment cannot be found well and the benefit maximization of unit operation cannot be guaranteed. A data acquisition sub-module, a data preprocessing sub-module, a data analysis sub-module and other sub-modules of the monitoring equipment module operate in sequence, equipment operation states and fault probabilities are transmitted to the fault prediction and early warning module, the fault prediction and early warning module predicts trends, time points and threshold exceeding through an ARIMA model after processing, early warning is sent to the maintenance decision module, and the maintenance decision module makes a plan by combining multiple factors and displays and transmits the plan. The maintenance process is tracked and recorded, the data division module firstly performs preliminary division and then performs secondary division according to fault judgment, fault details are determined through multi-stage analysis, the maintenance direction is determined through overall assistance, maintenance is accurate, and unit benefits are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal power plant monitoring, and particularly to an intelligent monitoring system for the operating state of a thermal power plant. Background Art

[0002] With the strong promotion of the national "dual carbon" strategic work, more energy and power enterprises have joined the practice of smart power plant transformation, especially traditional thermal power plants. As an important pillar of China's electric power energy, thermal power enterprises shoulder the important responsibilities and missions of carbon peak and carbon neutrality. Facing the transformation of China's thermal power industry from the "dual control" assessment of energy consumption to the "dual control" of total carbon emissions and carbon emission intensity, higher requirements are put forward for thermal power plants in aspects such as deep peak shaving, ultra-low emissions, and flexible operation. Low-carbon transformation has become the only way to promote the high-quality development of the thermal power industry.

[0003] The control and management platform of the intelligent monitoring system for thermal power plants is an advanced system based on intelligent technologies such as 5G, Internet +, Internet of Things, big data analysis, and cloud computing. It upgrades the traditional power plant operation monitoring systemically by connecting the production data, operation data, and external data of power generation enterprises and deeply integrating the DCS system data.

[0004] There are a large number of equipment in thermal power generating units, with a high degree of structural complexity and technical content, involving multiple majors such as machinery, electricity, and instrumentation. The stability of the operation of the entire unit depends to a large extent on the stable operation of individual equipment or important auxiliary equipment. Once a single piece of equipment fails, it is very likely to trigger protection actions and even cause the unit to crash. Therefore, the data analysis and maintenance of equipment are also very important work contents in the daily operation and maintenance of thermal power generating units.

[0005] The current equipment maintenance methods in the power industry mainly include the following four methods: passive maintenance after equipment failure, preventive maintenance based on cycle maintenance, maintenance based on equipment parameter status, and predictive maintenance.

[0006] For example, Duke Energy in the United States has more than 90 factories in 7 states across the country, with a total power generation capacity of 58,000 MW. Its power generation facilities include thermal power, simple cycle gas turbines, combined cycles, and integrated gasification combined cycle power plants, as well as large-scale wind power and solar power generation. Duke Energy established a Monitoring and Diagnostic Center (M&D Center) as early as 2004 to carry out smart power generation projects, design predictive diagnostic models for key equipment facilities (including boiler auxiliary equipment), and avoid unplanned shutdown accidents of key unit facilities. So far, 11,000 prediction models have been designed, covering 237 power generation facilities and 87% of the power generation capacity.

[0007] For predictive maintenance models, currently, the foreign manufacturers with advanced technologies in equipment maintenance diagnosis mainly include GE, Aspen, Honeywell, Baker Hughes, Microsoft & Amazon.

[0008] However, their equipment maintenance methods still mainly rely on the first three maintenance methods, mainly relying on preventive maintenance modules in the production information management system (MIS), etc. Although it can solve the problem of equipment failure to a certain extent, the biggest problem is that it is impossible to find a good balance between over-maintenance and under-maintenance of equipment, and it is impossible to ensure the maximum benefit of unit operation.

[0009] Therefore, an intelligent monitoring system for the operating status of thermal power plants is needed. Summary of the Invention

[0010] To solve all or part of the above problems, the purpose of the present invention is to provide an intelligent monitoring system for the operating status of thermal power plants, so as to solve the problem that it is impossible to find a good balance between over-maintenance and under-maintenance of equipment and it is impossible to ensure the maximum benefit of unit operation.

[0011] To achieve the above object, the present invention provides the following technical solution: An intelligent monitoring system for the operating status of thermal power plants, including:

[0012] Equipment operation unit: used to comprehensively collect numerous operation parameters of thermal power plant equipment, establish a large equipment data model, and judge whether the operation condition of the equipment is normal by real-time monitoring of equipment parameters and comparing and analyzing them with the large model. When it is found that the parameters deviate from the set threshold, give an early warning and repair in time;

[0013] Environmental protection monitoring unit: Based on the equipment operation unit, it is responsible for separately collecting and monitoring the environmental protection-related data of thermal power plants, real-time monitoring the concentration and total amount of pollutants emitted by thermal power plants into the atmosphere, as well as wastewater pollutant indicators, solid waste generation and disposal information, and plant boundary noise values. Through data feedback, it can promptly detect changes in the operation efficiency of environmental protection facilities. When the emissions exceed the standard, it will promptly remind to adjust. At the same time, it also provides a basis for enterprises to formulate environmental protection strategies, optimize environmental protection processes, and improve the efficiency of resource recovery and utilization;

[0014] Production management unit: Collect data around multiple management dimensions of thermal power plant production and operation, covering the intake, storage, consumption and coal quality of fuel, unit start-stop and operation time, production plan and scheduling arrangements, and personnel management information. Coordinate from multiple aspects to ensure the orderly and efficient development of production;

[0015] System and equipment status unit: responsible for monitoring the operation status between each independent system and equipment of the thermal power plant, ensuring the accurate and stable transmission of control instructions, promptly discovering and handling control anomalies, and enabling the normal automatic operation control of the unit.

[0016] Furthermore, the device operation unit includes:

[0017] Big data modeling module: Collect real-time data on historical device self-fault factors, device operation environment fault factors, and device historical operation fault factors, classify and organize the data and extract features, and build a model using neural network algorithms based on the extracted features;

[0018] Monitoring device module: Real-time collect data on the device's own factors, device operation environment factors, and device historical operation factors and input them into the big data model, and perform real-time analysis on the data through model algorithms. Once the device shows signs of failure, the system can quickly capture the abnormal changes in the data.

[0019] Furthermore, the monitoring device module includes:

[0020] Data acquisition module: Responsible for collecting sensor data on the thermal power plant equipment for monitoring the device's own factors and device operation environment factors, and at the same time receiving the device historical operation factor data recorded in the device, and transmitting the collected data to the data preprocessing module;

[0021] Data preprocessing module: Clean the collected raw data, remove noise data, outliers, and duplicate data, and at the same time perform standardization processing on the data to make data of different types and magnitudes comparable, classify and organize the processed data according to the device's own factors, operation environment factors, and historical operation factors, and store it in the data buffer for the data analysis module to call;

[0022] Data analysis module: Read the preprocessed data from the data buffer and input it into the pre-built big data model, perform real-time analysis on the data using model algorithms, comprehensively judge the operating state of the device, identify whether the device shows signs of failure or potential risks, calculate the failure probability, and transmit the analysis results to the fault prediction and warning module;

[0023] Fault prediction and warning module: According to the failure probability provided by the data analysis module and the change trend of the device operation data, use prediction algorithms to predict the failure development trend of the device in the next period of time, determine the possible failure time point, and combine the failure prediction results with the preset warning threshold. When the failure probability exceeds the threshold or it is predicted that the failure is about to occur, send a digital warning signal and transmit the warning information to the maintenance decision-making module;

[0024] Maintenance decision-making module: Receive the warning information from the fault prediction and warning module, comprehensively consider the device characteristics to formulate a maintenance plan, which will be displayed on the man-machine interface to remind the operation and maintenance personnel. At the same time, track and adjust the maintenance process and record the maintenance data.

[0025] Further, the data analysis module specifically constructs an SVM classification model, marks the normal operation state of the device as one class and the fault state as another class, and uses the device's own factors, operating environment factors, and historical operation factors as feature vectors.

[0026] Let the training data set be {(x 1 ,y 1 ),(x 2 ,y 2 ),...,(x m ,y m ),}, where x i is the feature vector and y i is the corresponding class label (normal is -1, fault is 1). The optimization objective function of SVM is: The constraint condition is: y i (w·x i +b)≥1-ξ i ,ξ i ≥0,i=1,2,···,m, where w is the normal vector of the hyperplane, b is the intercept, C is the penalty parameter, ξ is the slack variable. By solving this optimization problem, the model parameters are obtained, and then new data is classified and predicted to obtain the device failure probability P.

[0027] Further, the fault prediction and warning module specifically, for the device fault probability sequence {P t}, first conducts a stationarity test, transforms the non-stationary sequence into a stationary sequence through differencing operations, and then determines the order (p, d, q) of the model according to the characteristics of the autocorrelation function and partial autocorrelation function. p is the autoregressive order, d is the differencing order, and q is the moving average order. The historical data is used to estimate and train the prediction model so that the model can capture the time series law and trend change of the device operation data;

[0028] Estimate the parameters of the ARIMA model and establish the model where is the autoregressive coefficient polynomial, θ(B)=1+θ 1 B+θ 2 B 2 +···+θ q B q is the moving average coefficient polynomial, Δ d =(1-B) d is the differencing operator, B is the lag operator, and ε t is the white noise sequence. Use the established ARIMA model to predict the fault probability in a future period of time, and determine the fault time point T f .

[0029] Further, the maintenance decision-making module specifically formulates a scientific and reasonable maintenance plan by receiving the early warning information from the fault prediction and early warning module, and combining the importance level of the equipment, the current operating load of the unit, and the availability of maintenance resources;

[0030] Determine the equipment to be repaired, the maintenance items, the maintenance time point, and the resource allocation plan required for maintenance, ensure that the maintenance is carried out at the best time point, and finally display it on the man-machine interaction interface to remind the operation and maintenance personnel to handle it in time, avoid over-maintenance or under-maintenance of the equipment, minimize the unit shutdown time, and ensure the maximization of the unit operation benefit;

[0031] Transmit the maintenance plan to the maintenance execution and monitoring module, track and adjust the maintenance process, and record the maintenance data for subsequent equipment management and system optimization.

[0032] Further, the monitoring equipment module further includes a data division module: First, the thermal power plant equipment is initially divided into three major factor types: its own factors, equipment operation environment factors, and equipment historical operation factors, and the division is transmitted to the data acquisition module. The data acquisition module starts to execute until the fault prediction and early warning module determines the factor type to which the fault belongs, and then transmits the data to the data division module again. The data division module conducts a secondary division of the factor type to which it belongs. The data division module conducts a secondary division of the factor type to which it belongs at least into three major factor types, and then transmits it to the data acquisition module again. After multiple levels of division and multiple levels of analysis, the most accurate cause of the equipment failure and the best fault occurrence time point are finally determined, not only achieving the purpose of the most accurate maintenance time and the most accurate fault point, but also the early warning judgment process in the way of gradual prediction avoids the problem of data confusion caused by a one-time judgment of a batch of factors.

[0033] Further, the production management unit includes:

[0034] Fuel management module: A weighbridge and an automated metering system are installed at the fuel inlet to accurately record the incoming weight and input it into the system in real time. The level gauge in the warehouse monitors the inventory level with the help of sensors and transmits the level information. Coal quality analysis instruments are equipped to sample and detect before entering the factory and during use and upload the data. A reasonable procurement strategy is formulated based on the incoming quantity and inventory in combination with the production plan. At the same time, the boiler combustion parameters are adjusted according to the coal quality analysis results to adapt to different coal qualities, thereby optimizing the entire fuel management process;

[0035] Unit operation management module: Monitoring sensors are installed at key parts of the unit to collect real-time data, which is connected to the unit control system to obtain start-stop time, operation mode, and load change data and transmit it to the database. The start-stop plan and load distribution plan can be formulated based on the equipment fault analysis results of the equipment operation unit;

[0036] Production Planning and Scheduling Module: Integrate the load demand of the power grid dispatching center, power market trading data, and the information of internal units and equipment in the power plant to construct a production planning and scheduling management system. Use scheduling software with intelligent algorithms to generate optimization plans based on constraints such as unit capacity, maintenance time, and energy consumption. At the same time, formulate long-term and short-term production plans according to the collected data, clarify the power generation tasks, start-stop times, and load curves of each unit and issue them. Recalculate and adjust the plan with the help of the equipment failure analysis results of the equipment operation unit;

[0037] Personnel Management Module: Establish an employee information database to enter basic information, electronic shift scheduling information, and training records. At the same time, set up work performance collection points at each position, reasonably allocate personnel according to shift scheduling and production tasks, allocate maintenance forces during major overhauls, and conduct reasonable shift rotations during daily operations to avoid fatigue operations. Based on performance data evaluation and analysis, reward outstanding employees and provide promotion opportunities. Provide training and guidance or job adjustments for employees who do not meet the standards.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] An intelligent monitoring system for the operating status of a thermal power plant proposed by the present invention collects detailed real-time device data through a big data modeling module. After classification, sorting, and feature extraction, a neural network algorithm is used to construct a model. The data acquisition module in the device monitoring module collects sensor data and historical operation factor data, which is transmitted to the data preprocessing module for cleaning, standardization, classification, and sorting, and then stored in the data buffer. The data analysis module reads the data, inputs it into the big data model, comprehensively judges the operating status and failure probability of the device, and transmits it to the failure prediction and early warning module. The failure prediction and early warning module performs processing such as stationarity test on the failure probability sequence, and then uses the ARIMA model to predict the development trend and time point of the failure. When the threshold is exceeded, an early warning signal is sent to the maintenance decision-making module. The maintenance decision-making module formulates a maintenance plan in combination with the importance level of the device, the operating load of the unit, and the availability of maintenance resources, determines the maintenance equipment, projects, time points, and resource allocation plans, displays them on the man-machine interface, and transmits them to the maintenance execution and monitoring module. It also tracks and adjusts the maintenance process record data. In addition, the data division module in the device monitoring module first preliminarily divides the device factors, and then divides them again after the failure is determined. Through multi-level division and analysis, the exact cause and best time point of the failure are determined. Each module cooperates with each other, can clarify the direction of maintenance required, reduce the maintenance time, reduce the unit shutdown time, and moreover, through system calculation, the equipment to be maintained and the maintenance time point are determined, and maintenance is carried out at the best time point, avoiding excessive maintenance of equipment resulting in increased equipment shutdown time, or lack of maintenance leading to direct scrapping and damage of equipment, and ultimately unable to ensure the maximum benefit of unit operation. Brief Description of the Drawings

[0040] Figure 1 It is a schematic diagram of the equipment operation unit module of the intelligent monitoring system for the operating status of the thermal power plant of the present invention;

[0041] Figure 2 This is a schematic diagram of the production management unit module of the intelligent monitoring system for the operating status of the thermal power plant of the present invention. Specific embodiments

[0042] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0043] As Figure 1 - Figure 2 shown, an intelligent monitoring system for the operating status of a thermal power plant includes:

[0044] Equipment operation unit: used to comprehensively collect numerous operation parameters of the thermal power plant equipment, establish a large equipment data model, judge whether the operation condition of the equipment is normal by real-time monitoring of the equipment parameters and comparing and analyzing with the large model, and when it is found that the parameters deviate from the set threshold, give an early warning and perform maintenance in a timely manner.

[0045] Environmental protection monitoring unit: responsible for separately collecting and monitoring the environmental protection-related data of the thermal power plant based on the equipment operation unit, real-time monitoring the concentration and total amount of pollutants emitted by the thermal power plant into the atmosphere, as well as wastewater pollutant indicators, solid waste generation and disposal information, and plant boundary noise values, timely detecting changes in the operation efficiency of environmental protection facilities through data feedback, reminding to adjust in a timely manner when the emissions exceed the standard, and at the same time providing a basis for the enterprise to formulate environmental protection strategies, optimize environmental protection processes, and improve the resource recovery and utilization efficiency.

[0046] Production management unit: collect data around multiple management dimensions of the thermal power plant production operation, covering the intake, storage, consumption and coal quality of fuel, unit start-stop and operation time, production plan and scheduling arrangements, and personnel management information, and coordinate from multiple aspects to ensure the orderly and efficient development of production.

[0047] System and equipment status unit: responsible for monitoring the operation status between each independent system and equipment of the thermal power plant, ensuring the accurate and stable transmission of control instructions, timely detecting and handling control abnormalities, and enabling the normal automatic operation control of the unit.

[0048] Furthermore, the equipment operation unit includes:

[0049] Big data modeling module: collect real-time data of historical equipment self-fault factors, equipment operation environment fault factors, and equipment historical operation fault factors, classify and sort out the data and extract features, and build a model using neural network algorithms based on the extracted features.

[0050] Among them, the factors of the equipment itself are those that can cause damage during the use of the equipment itself. For example:

[0051] 1. Mechanical wear

[0052] Wear of rotating components: There are a large number of rotating equipment in thermal power plants, such as the rotors of steam turbines, generators, and the impellers of fans. During long-term high-speed operation, due to mechanical friction, the surface materials of these components will gradually wear. For example, the blades of steam turbines are in long-term contact with steam. Under high-speed rotation, the edges of the blades will be worn due to the erosion of steam and their own vibration. As the wear intensifies, the shape of the blades changes, which will affect the efficiency of the steam turbine and may even lead to serious failures such as blade fracture.

[0053] Wear of transmission components: Transmission components such as gears and couplings will also wear during the power transmission process. Taking gears as an example, during long-term meshing, due to the pressure and relative movement between teeth, fatigue wear, abrasive wear, etc. will occur. After wear, problems such as tooth profile deformation and increased backlash will occur in the gears, resulting in a decrease in transmission accuracy, increased vibration and noise, and affecting the normal operation of the equipment.

[0054] 2. Material aging

[0055] Aging of metal materials: Metal components in thermal power plant equipment, such as boiler pipes and pressure vessels, work in an environment of high temperature, high pressure, and corrosive media for a long time, and the metal materials will age. For example, the water-cooled wall tubes of boilers are filled with high-temperature and high-pressure steam-water mixtures inside and are irradiated by the heat of the flame outside. After long-term operation, creep will occur in the metal pipe walls. Creep is a process in which materials slowly deform under high temperature and constant stress. As creep develops, the pipe walls will become thinner and the strength will decrease, eventually possibly leading to pipe rupture.

[0056] Aging of insulating materials: Insulating materials in electrical equipment such as generators and transformers will age due to factors such as temperature, electric field, and chemical substances. For example, the insulating materials of the stator windings of generators will gradually degrade in insulation performance during long-term operation due to the high temperature and corona discharge inside the motor. When the insulation resistance drops to a certain level, insulation breakdown will occur, resulting in a short-circuit fault and affecting the normal power generation of the generator.

[0057] 3. Equipment fatigue

[0058] Thermal fatigue: During the frequent start-up and shutdown processes of boiler equipment, or when the load changes significantly, due to the rapid temperature changes, thermal stresses will be generated. These thermal stresses can cause thermal fatigue in equipment components. For example, during the start-up process of a boiler's steam drum, the temperature of the water inside rises rapidly, while the temperature of the outer wall of the steam drum rises relatively slowly, thus generating thermal stresses within the steam drum wall. After multiple such temperature cycles, thermal fatigue cracks will appear on the steam drum wall, reducing the strength and sealing performance of the steam drum.

[0059] Mechanical fatigue: Some equipment components that bear alternating loads, such as the shafts and connecting rods of steam turbines, will exhibit mechanical fatigue. Taking the steam turbine shaft as an example, during rotation, the shaft not only bears its own gravity and the centrifugal force generated by rotation, but also bears the steam forces transmitted from the blades. These forces are alternating, and after long-term action, fatigue cracks will be generated on the surface or inside of the shaft. If the cracks continue to expand, it may ultimately lead to the fracture of the shaft, triggering serious accidents.

[0060] Operating environment factors are those factors in the equipment's surrounding environment that can cause damage, such as:

[0061] 1. Temperature influence

[0062] High-temperature environment: Equipment such as boilers and steam turbines in thermal power plants operate in a high-temperature environment. Excessive temperature will accelerate the aging and damage of equipment materials. For example, in the high-temperature superheater area of a boiler, the steam temperature inside the pipe is as high as 500 - 600 °C, and the outside of the pipe is directly radiated by the flame. This high-temperature environment will reduce the strength of the pipe material and deteriorate its oxidation resistance. At the same time, high temperature will also cause the aging of the equipment's sealing materials, resulting in leakage. For example, high temperature will make the shaft seal material of the steam turbine soften and deform, losing its sealing function, causing steam leakage and reducing the efficiency of the unit.

[0063] Low-temperature environment: In thermal power plants in some northern regions, the temperature is relatively low in winter. The low-temperature environment may cause some liquid media in the equipment to freeze, resulting in pipeline rupture. For example, if the open-air fire-fighting water pipes, cooling water pipes, etc. do not have good heat insulation measures, the water in the pipes will freeze in the low temperature, and the volume expansion will burst the pipes. In addition, low temperature will also affect the performance of some equipment. For example, the capacity of a battery will decrease at low temperature, affecting its power supply ability in an emergency.

[0064] 2. Corrosion effect

[0065] Chemical corrosion: The equipment in thermal power plants is subject to corrosion by various chemical substances. In boilers, dissolved oxygen, carbon dioxide and other gases in water, as well as acidic or alkaline substances in water, will corrode metal pipes. For example, dissolved oxygen in water will react with the metal pipe wall to form rust. If the corrosion is severe, it will cause the pipe wall thickness to decrease and even result in perforation and leakage. In the flue gas system, acidic gases such as sulfur dioxide and sulfur trioxide will corrode equipment such as flues and chimneys, especially when the humidity is high, the corrosion rate will accelerate.

[0066] Electrochemical corrosion: When different metals come into contact in an electrolyte solution, a primary battery will be formed and electrochemical corrosion will occur. For example, in the seawater cooling system of a thermal power plant, when a copper alloy heat exchanger is connected to a carbon steel pipe, since seawater is an electrolyte solution, a primary battery will be formed between the two metals, leading to accelerated corrosion of the carbon steel pipe. Electrochemical corrosion is usually more concealed and destructive than chemical corrosion, and poses a serious threat to the structural integrity of the equipment.

[0067] 3. Effects of dust and impurities

[0068] Wear effect: After dust particles in the air enter the equipment, they will cause wear to the moving parts of the equipment. For example, during the operation of a fan, a large amount of dust enters the fan along with the air, and the dust particles will scour the fan impeller and casing, wearing the surface of the impeller and reducing the efficiency of the fan. In a coal mill, impurities in the coal will also cause wear to the grinding components of the coal mill, shortening the service life of the grinding components.

[0069] Blocking problems: Dust and impurities may also block the pipes and filters of the equipment. In the air preheater of a thermal power plant, dust in the flue gas is likely to deposit on the surface of the heat exchange elements of the preheater, forming ash deposits. The ash deposits will reduce the heat exchange efficiency of the preheater and increase the ventilation resistance. If the ash deposits are severe, it may also lead to poor ventilation and affect the normal combustion of the boiler. In some precision instrument equipment, such as the sampling pipelines and filters of instruments, the blockage of dust and impurities will cause inaccurate measurement and affect the control and operation of the equipment.

[0070] Operating operation factors are factors that can cause damage due to improper operation by personnel, such as:

[0071] 1. Improper operation

[0072] Improper start-stop operation: During the start-up and shutdown processes of the equipment, if the operation sequence is incorrect or the operation speed is too fast, it will cause damage to the equipment. For example, when starting a boiler, if the operation is not carried out in accordance with the specified heating and pressure increase procedures, it may cause damage to various components of the boiler due to excessive thermal stress.

[0073] Improper load regulation: The equipment in a thermal power plant needs to adjust the load according to the requirements of the power grid. If the load adjustment speed is too fast or the adjustment range exceeds the allowable range of the equipment, it will have an adverse impact on the equipment. For example, when quickly increasing the boiler load, if the adjustment of fuel quantity and air volume does not match, it will lead to incomplete combustion, generating black smoke. At the same time, it may also cause drastic fluctuations in the boiler pressure and temperature, affecting the safe and stable operation of the equipment.

[0074] 2. Maintenance mistakes

[0075] Maintenance quality problems: During the equipment maintenance process, if the technical level of maintenance personnel is not up to standard or the maintenance process does not meet the requirements, it will cause potential hazards in the maintained equipment. For example, when welding boiler pipes, if the welding quality is poor, with defects such as pores and slag inclusions, it will reduce the strength of the pipes and is prone to leakage accidents.

[0076] Irrational maintenance cycle: If the maintenance cycle is too long, some potential problems of the equipment cannot be discovered and handled in time, which will lead to the accumulation of problems and eventually cause failures. On the contrary, if the maintenance cycle is too short, it will increase unnecessary maintenance costs, and at the same time, it may also damage the equipment due to frequent disassembly and installation.

[0077] Monitoring equipment module: It collects data on the equipment's own factors, equipment operating environment factors, and equipment historical operation factors in real time and inputs them into the big data model. Through model algorithms, it analyzes the data in real time. Once the equipment shows signs of failure, the system can quickly capture the abnormal changes in the data.

[0078] Furthermore, the monitoring equipment module includes:

[0079] Data acquisition module: It is responsible for collecting sensor data on the thermal power plant equipment for monitoring the equipment's own factors and equipment operating environment factors. At the same time, it receives the equipment historical operation factor data recorded by the equipment and transmits the collected data to the data preprocessing module.

[0080] Data preprocessing module: It cleans the collected raw data, removes noise data, outliers, and duplicate data. At the same time, it standardizes the data to make data of different types and magnitudes comparable. It classifies and organizes the processed data according to the equipment's own factors, operating environment factors, and historical operation factors, and stores it in the data buffer for the data analysis module to call.

[0081] Data analysis module: It reads the preprocessed data from the data buffer and inputs it into the pre-constructed big data model. It uses model algorithms to analyze the data in real time, comprehensively judges the operating state of the equipment, identifies whether the equipment shows signs of failure or potential risks, calculates the failure probability, and transmits the analysis results to the fault prediction and early warning module.

[0082] The specific model algorithm is as follows: By constructing an SVM classification model, the normal operating state of the device is marked as one class, and the fault state is marked as another class. The device's own factors, operating environment factors, and historical operation factors are used as feature vectors.

[0083] Let the training data set be {(x 1 ,y 1 ),(x 2 ,y 2 ),...,(x m ,y m ),}, where x i is the feature vector, and y i is the corresponding class label (normal is -1, fault is 1). The optimization objective function of SVM is: The constraint condition is: y i (w·x i +b)≥1-ξ i ,ξ i ≥0, i = 1, 2, ···, m, where w is the normal vector of the hyperplane, b is the intercept, C is the penalty parameter, ξ is the slack variable. By solving this optimization problem, the model parameters are obtained, and then new data is classified and predicted to obtain the device failure probability P.

[0084] Fault prediction and early warning module: According to the failure probability provided by the data analysis module and the change trend of the device operation data, use the prediction algorithm to predict the failure development trend of the device in the future period of time, determine the possible failure time point, and combine the failure prediction result and the preset early warning threshold. When the failure probability exceeds the threshold or it is predicted that the failure is about to occur, a digital early warning signal is sent, and the early warning information is transmitted to the maintenance decision-making module.

[0085] The specific prediction algorithm is as follows: For the device failure probability sequence {P t}, first perform a stationarity test, convert the non-stationary sequence into a stationary sequence through differencing operations, and then determine the order (p, d, q) of the model according to the characteristics of the autocorrelation function and partial autocorrelation function. p is the autoregressive order, d is the differencing order, and q is the moving average order. Use historical data to estimate and train the prediction model so that the model can capture the time series law and trend change of the device operation data.

[0086] Estimate the parameters of the ARIMA model and establish the model where is the autoregressive coefficient polynomial, θ(B) = 1 + θ 1 B + θ 2 B 2 + ··· + θ q Bq is the moving average coefficient polynomial, Δ d =(1 - B) d is the difference operator, B is the lag operator, ε t is the white noise sequence. The established ARIMA model is used to predict the failure probability in the future for a period of time, and the failure time point T is determined according to the prediction result f .

[0087] The monitoring device module also includes a data partitioning module: First, the thermal power plant equipment is initially partitioned into three major factor types: its own factors, equipment operating environment factors, and equipment historical operation factors. The partitioning results are transmitted to the data acquisition module, and the data acquisition module starts to execute until the failure prediction and warning module determines the factor type to which the failure belongs and transmits the data back to the data partitioning module. The data partitioning module performs a secondary partitioning of the factor type to which it belongs. The data partitioning module performs a secondary partitioning of the factor type to which it belongs and at least partitions the factor type into three major factor types, and then transmits it back to the data acquisition module again. After multi-level partitioning and multi-level analysis, the most accurate cause of the equipment failure and the best failure time point are finally determined, not only achieving the purpose of the most accurate maintenance time and the most accurate failure point, but also the early warning judgment process in the way of step-by-step prediction avoids the problem of data confusion caused by a one-time judgment of a batch of factors

[0088] The maintenance decision-making module: By receiving the warning information from the failure prediction and warning module, combining the importance level of the equipment, the current unit operation load situation, and the availability of maintenance resources, a scientific and reasonable maintenance plan is formulated to determine the equipment to be repaired, the maintenance items, the maintenance time point, and the resource allocation plan required for maintenance, ensuring maintenance at the best time point, and finally displaying it on the man-machine interface to remind the operation and maintenance personnel to handle it in time, avoiding over-maintenance or under-maintenance of the equipment, minimizing the unit shutdown time, ensuring the maximization of the unit operation benefit, transmitting the maintenance plan to the maintenance execution and monitoring module, and tracking and adjusting the maintenance process, recording the maintenance data for subsequent equipment management and system optimization

[0089] Furthermore, the production management unit includes:

[0090] The fuel management module: A weighbridge and an automated metering system are installed at the fuel inlet to accurately record the incoming weight and input it into the system in real time. The level gauge in the warehouse monitors the inventory level with the help of sensors and transmits the level information. Coal quality analysis instruments are equipped to sample and detect before entering the factory and before use and upload the data. A reasonable procurement strategy is formulated based on the incoming quantity and inventory level in combination with the production plan. At the same time, the boiler combustion parameters are adjusted according to the coal quality analysis results to adapt to different coal qualities, thereby optimizing the entire fuel management process

[0091] Unit operation management module: Install monitoring sensors at key parts of the unit to collect real-time data, connect with the unit control system to obtain start-stop time, operation mode and load change data, and transmit them to the database. It can formulate start-stop plans and load distribution schemes based on the analysis results of equipment failure analysis in the equipment operation unit with the help of the analysis results.

[0092] Production plan and scheduling module: Integrate the load demand of the power grid dispatching center, power market trading data, and the information of the units and equipment within the power plant to build a production plan and scheduling management system. Use scheduling software with intelligent algorithms to generate optimization schemes according to constraints such as unit capacity, maintenance time, and energy consumption. At the same time, formulate long-term and short-term production plans based on the collected data, clarify the power generation tasks, start-stop times and load curves of each unit and issue them. Recalculate and adjust the plan with the help of the equipment failure analysis results in the equipment operation unit.

[0093] Personnel management module: Establish an employee information database to enter basic information, electronic shift scheduling information and training records. At the same time, set up work performance collection points at each position, reasonably allocate personnel according to shift scheduling and production tasks, allocate maintenance forces during overhauls, and arrange reasonable shifts during daily operations to avoid fatigue operations. Based on performance data evaluation and analysis, reward excellent employees and provide promotion opportunities, and provide training and counseling or job adjustments for employees who do not meet the standards.

[0094] It should be noted that in the description of this application, it should be understood that the orientation or positional relationship indicated by terms such as "length", "thickness", "inner", "outer", "axial", "radial", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention.

[0095] In addition, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0096] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. An intelligent monitoring system for the operation status of a thermal power plant, characterized in that: include: Equipment operation unit: used to comprehensively collect numerous operating parameters of thermal power plant equipment and establish a large model of equipment data. Through real-time monitoring of equipment parameters and comparison and analysis with the large model, it can determine whether the operating condition of the equipment is normal. If parameters deviate from the set threshold, timely warning and maintenance can be carried out; Environmental monitoring unit: Based on the equipment operation unit, it is responsible for the independent collection and monitoring of environmental protection data of thermal power plants, real-time monitoring of the concentration and total amount of pollutants emitted into the atmosphere by thermal power plants, as well as wastewater pollutant indicators, solid waste generation and disposal information, and plant boundary noise values. Through data feedback, it can timely detect changes in the operating efficiency of environmental protection facilities, and promptly remind adjustments when emissions exceed the standards. At the same time, it also provides a basis for enterprises to formulate environmental protection strategies, optimize environmental protection processes, and improve resource recycling efficiency; Production management unit: collects data from multiple management dimensions of thermal power plant production and operation, including fuel intake, storage, consumption and coal quality, unit start-up and shutdown and operation time, production plan and scheduling, and personnel management information, and coordinates from multiple aspects to ensure orderly and efficient production; System and equipment status unit: responsible for monitoring the operating status of each independent system and equipment in the thermal power plant, ensuring accurate and stable transmission of control instructions, timely detection and processing of control anomalies, and ensuring normal automatic operation and control of the unit.

2. The intelligent monitoring system for the operation status of a thermal power plant according to claim 1, characterized in that: The equipment operation unit comprises: Big data modeling module: collects real-time data on historical equipment failure factors, equipment operating environment failure factors, and equipment historical operation failure factors, classifies and sorts the data, extracts features, and builds a model using a neural network algorithm based on the extracted features; Monitoring equipment module: collects data on the equipment's own factors, equipment operating environment factors, and equipment historical operation factors in real time and inputs them into the big data model. The data is analyzed in real time through the model algorithm. Once the equipment shows signs of failure, the system can quickly capture abnormal changes in the data.

3. The intelligent monitoring system for the operation status of a thermal power plant according to claim 2, characterized in that: The monitoring equipment module comprises: Data acquisition module: responsible for collecting sensor data on thermal power plant equipment used to monitor equipment factors and equipment operating environment factors, while receiving equipment historical operation factor data recorded by the equipment, and transmitting the collected data to the data preprocessing module; Data preprocessing module: cleans the collected raw data, removes noise data, outliers and duplicate data, and standardizes the data to make data of different types and magnitudes comparable. The processed data is classified and sorted according to the device's own factors, operating environment factors and historical operation factors, and stored in the data cache area, waiting for the data analysis module to call; Data analysis module: reads pre-processed data from the data buffer area and inputs it into the pre-built big data model, uses the model algorithm to analyze the data in real time, comprehensively judges the operating status of the equipment, identifies whether the equipment has signs of failure or potential risks, calculates the failure probability, and transmits the analysis results to the fault prediction and early warning module; Fault prediction and early warning module: Based on the fault probability provided by the data analysis module and the change trend of the equipment operation data, the prediction algorithm is used to predict the fault development trend of the equipment in the future, determine the possible fault time point, and combine the fault prediction results with the preset early warning threshold. When the fault probability exceeds the threshold or it is predicted that the fault is about to occur, a digital early warning signal is issued and the early warning information is transmitted to the maintenance decision module; Maintenance decision module: Receives warning information from the fault prediction and warning module, and formulates a maintenance plan based on comprehensive consideration of equipment characteristics. The plan will be displayed on the human-computer interaction interface to remind operation and maintenance personnel. At the same time, the maintenance process is tracked and adjusted, and maintenance data is recorded.

4. The intelligent monitoring system for the operation status of a thermal power plant according to claim 3, characterized in that: The data analysis module specifically constructs an SVM classification model to mark the normal operating state of the equipment as one category and the fault state as another category, using the equipment's own factors, operating environment factors and historical operation factors as feature vectors; Assume that the training data set is {(x1,y1),(x2,y2),...,(x m ,y m ),}, where x i is the feature vector, y i is the corresponding category label (normal is -1, fault is 1), and the optimization objective function of SVM is: The constraints are: i (w·x i +b)≥1-ξ i ,ξ i ≥0,i=1,2,···,m, where w is the normal vector of the hyperplane, b is the intercept, C is the penalty parameter, and ξ is the slack variable. The model parameters are obtained by solving this optimization problem, and then the new data are classified and predicted to obtain the equipment failure probability P.

5. The intelligent monitoring system for the operation status of a thermal power plant according to claim 4, characterized in that: The fault prediction and warning module specifically performs the following steps to predict the equipment fault probability sequence {P t }, firstly, a stationarity test is performed, and the non-stationary series is converted into a stationary series through differential operation. Then, the order of the model (p, d, q) is determined according to the characteristics of the autocorrelation function and the partial autocorrelation function, where p is the autoregressive order, d is the differential order, and q is the moving average order. The historical data is used to estimate and train the parameters of the prediction model so that the model can capture the time series laws and trend changes of the equipment operation data; Estimate the parameters of the ARIMA model and build the model in is the autoregressive coefficient polynomial, θ(B)=1+θ1B+θ2B 2 +···+θ q B q is the moving average coefficient polynomial, Δ d =(1-B) d is the difference operator, B is the lag operator, ε t is a white noise sequence. The established ARIMA model is used to predict the failure probability in the future, and the failure time point T is determined according to the prediction results. f .

6. The intelligent monitoring system for the operation status of a thermal power plant according to claim 3, characterized in that: The maintenance decision module specifically formulates a scientific and reasonable maintenance plan by receiving the warning information from the fault prediction and warning module and combining the importance level of the equipment, the current unit operating load, and the availability of maintenance resources; Determine the equipment that needs maintenance, maintenance items, maintenance time, and resource allocation plan required for maintenance, ensure maintenance is carried out at the best time, and finally display it on the human-computer interaction interface to remind operation and maintenance personnel to handle it in time; The maintenance plan is transmitted to the maintenance execution and monitoring module, and the maintenance process is tracked and adjusted, and the maintenance data is recorded for subsequent equipment management and system optimization.

7. The intelligent monitoring system for the operation status of a thermal power plant according to claim 5, characterized in that: The monitoring equipment module also includes a data division module: first, the thermal power plant equipment is initially divided into three major factor types: its own factors, equipment operating environment factors, and equipment historical operation factors, and the division is transmitted to the data acquisition module. The data acquisition module starts to execute until the fault prediction and early warning module determines the factor type to which the fault belongs, and transmits the data to the data division module again. The data division module divides the factor type again, and then transmits it to the data acquisition module again. After multi-level division and multi-level analysis, the most accurate cause of the equipment failure and the best time point for the failure to occur are finally determined.

8. The intelligent monitoring system for the operation status of a thermal power plant according to claim 7, characterized in that: The data division module further divides the factor types into at least three major factor types.

9. The intelligent monitoring system for the operation status of a thermal power plant according to claim 1, characterized in that: The production management unit comprises: Fuel management module: A floor scale and an automated metering system are installed at the fuel entry point to accurately record the incoming weight and enter the information into the system in real time. The material level meter in the warehouse uses sensors to monitor the inventory level and transmit the material level information. Coal quality analysis instruments are equipped to conduct sampling and testing before entering the factory and before use, and upload the data. A reasonable procurement strategy is formulated based on the incoming quantity and inventory combined with the production plan. At the same time, the boiler combustion parameters are adjusted according to the coal quality analysis results to adapt to different coal qualities, thereby optimizing the entire fuel management process; Unit operation management module: Install monitoring sensors at key parts of the unit to collect real-time data, and connect with the unit control system to obtain start and stop time, operation mode and load change data to transmit to the database, and formulate start and stop plans and load distribution plans; Production planning and scheduling module: Integrate the load demand of the power grid dispatching center, power market transaction data, and the internal unit and equipment information of the power plant to build a production planning and scheduling management system, and use scheduling software with intelligent algorithms to generate optimization plans based on constraints such as unit capacity, maintenance time, and energy consumption. At the same time, formulate a long-term and short-term production plan based on the collected data, clarify the power generation tasks, start and stop times, and load curves of each unit, and issue them; Personnel management module: Establish an employee information database to enter basic information, electronic scheduling information and training records. At the same time, set up work performance collection points at each position, reasonably deploy personnel according to scheduling and production tasks, deploy maintenance forces during overhaul, and reasonably rotate shifts in daily operations to avoid fatigue. Based on performance data evaluation and analysis, reward outstanding employees and provide promotion opportunities, and provide training and counseling or job adjustments for employees who do not meet the standards.

10. The intelligent monitoring system for the operation status of a thermal power plant according to claim 9, characterized in that: The unit operation management module uses the equipment failure analysis results of the equipment operation unit to formulate a start-stop plan and a load distribution plan based on the analysis results; The production planning and scheduling module recalculates and adjusts the plan with the help of the equipment failure analysis results of the equipment operation unit.

Images