Intelligent management and control system for 600MW unit air cooling energy-saving and efficiency-improving robot
By constructing a closed-loop management and control system, and using multi-source data fusion and thermal coupling simulation to generate virtual entities, the robot is driven to autonomously plan its path. This solves the problem of intelligent management of traditional air-cooled systems, realizes intelligent management of the robot-controlled system, solves the problem of improving the intelligence level of traditional air-cooled systems, and achieves deep integration of energy saving, efficiency improvement and unmanned operation and maintenance.
Patent Information
- Application Number
- CN202511389686.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-26
- Publication Date
- 2026-02-03
AI Technical Summary
Traditional air-cooled systems rely on manual inspections or periodic cleaning, which cannot keep track of the spatial distribution and evolution of fouling on the fins in real time. This leads to insufficient or excessive cleaning, affecting heat exchange efficiency and operational economy. Moreover, most existing monitoring systems are open-loop structures, and the predictive models cannot be dynamically corrected based on the actual cleaning effect. Over the long term, due to equipment aging and environmental changes, the models become inaccurate, making it difficult to continuously improve the system's intelligence level.
The system constructs a data fusion module, a digital twin module, a trend prediction module, an optimization decision-making module, a path planning module, a robot execution module, and a feedback correction module to achieve closed-loop management. It generates a high-fidelity virtual mapping body through multi-source data fusion and thermal coupling simulation, and accurately predicts the evolution of dirt accumulation and back pressure changes by embedding a neural network model with physical laws. This drives the flexible robot to autonomously plan its path and perform cleaning operations, and the prediction model is dynamically corrected through actual feedback.
It significantly improves the heat exchange efficiency of the air-cooled system, reduces plant power consumption and cleaning resource consumption, solves the problems of slow response, over-cleaning or under-cleaning in traditional manual cleaning, and achieves a deep integration of energy saving and efficiency improvement with unmanned operation and maintenance.
Smart Images
Figure CN121454976A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent robot control technology, and in particular to an intelligent robot control system for air-cooled energy-saving and efficiency-enhancing 600MW generating units. Background Technology
[0002] Robot intelligent control technology refers to a comprehensive intelligent control technology that integrates multi-source data perception, digital twin modeling, artificial intelligence prediction, optimization decision-making algorithms, autonomous path planning, intelligent robot execution, and closed-loop feedback correction to meet the operation and maintenance needs of complex industrial equipment. Therefore, how to utilize advanced technologies to improve the intelligence level and safety of robot intelligent control has become one of the most pressing issues to be addressed.
[0003] In the field of intelligent robot management, traditional air-cooled systems rely on manual inspections or timed cleaning, which cannot grasp the spatial distribution and evolution trend of fin fouling in real time, making it difficult to accurately determine the timing of cleaning. This leads to insufficient or excessive cleaning, affecting heat exchange efficiency and operational economy. Moreover, most existing monitoring systems are open-loop structures, and the predictive models cannot be dynamically corrected based on the actual cleaning effect. Over the long term, due to equipment aging and environmental changes, the models become inaccurate, making it difficult to continuously improve the system's intelligence level. Summary of the Invention
[0004] In view of the aforementioned existing problems, the present invention is proposed.
[0005] Therefore, this invention provides a 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent management and control system to solve the problems of traditional air-cooled systems relying on manual inspection or timed cleaning, which cannot grasp the spatial distribution and evolution trend of fin fouling in real time, making it difficult to accurately judge the cleaning time, resulting in insufficient or excessive cleaning, affecting heat exchange efficiency and operating economy. Moreover, existing monitoring systems are mostly open-loop structures, and the prediction model cannot be dynamically corrected according to the actual cleaning effect. In the long run, due to equipment aging and environmental changes, the model becomes inaccurate, and the system's intelligence level is difficult to continuously improve.
[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system, which includes: The module includes a data fusion module, a digital twin module, a trend prediction module, an optimization decision-making module, a path planning module, a robot execution module, and a feedback correction module. The data fusion module is used to collect unit operating parameters, environmental meteorological data and historical cleaning information, perform data alignment and formatting processing, and generate a unified time-series dataset. The digital twin module is used to construct a three-dimensional geometric model based on the air-cooled island structure data, establish a dynamic simulation system by combining the thermodynamic and fluid coupling relationship, and input the unified time series dataset into the simulation system to generate a real-time mapped virtual operating state. The trend prediction module is used to predict future back pressure change trends and pollution distribution characteristics based on the virtual operating state and the pollutant deposition evolution law, using a time-series neural network model. The optimization decision module is used to calculate the exhaust steam loss based on the back pressure change trend, combine the cleaning energy consumption and the fan operation energy consumption to construct a comprehensive energy cost objective function, solve the multi-objective optimization problem, and generate a set of cleaning strategies. The path planning module is used to extract the target area location from the characteristics of the accumulated dirt distribution according to the selected cleaning strategy, combine the robot motion constraints, generate the optimal movement path from the initial pose to the target area, and form a control instruction sequence containing execution parameters. The robot execution module is used to receive the control command sequence, drive the flexible mobile robot with a composite adsorption mechanism to move along the optimal movement path, and after reaching the designated position, activate the gas-liquid synergistic cleaning device to complete the fin cleaning, and collect status data during the operation. The feedback correction module is used to receive the status data, analyze the changes in heat exchange performance before and after cleaning, calculate the deviation between the actual effect and the predicted result, and dynamically correct the model parameters in the trend prediction module when the deviation exceeds the set range.
[0007] As a preferred embodiment of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system of the present invention, the unit operating parameters collected by the data fusion module include turbine exhaust flow rate, exhaust temperature, condenser back pressure, fan speed and inlet and outlet air temperature of each section of the air-cooled island. The environmental meteorological data includes ambient temperature, wind speed, wind direction, relative humidity, and atmospheric particulate matter concentration. The historical cleaning information includes the start and end times of each cleaning session, the cleaning area, water consumption, power consumption, and back pressure changes after cleaning. The data alignment is achieved through timestamp matching, and the formatting process unifies data from different sources into data record units with a preset structure.
[0008] As a preferred embodiment of the 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system of the present invention, the digital twin module is used to construct a three-dimensional geometric model based on the air-cooled island structure data, establish a dynamic simulation system by combining the thermodynamic and fluid coupling relationship, and input the structured dataset into the simulation system to generate a real-time mapped virtual operating state; the specific steps are as follows: Three-dimensional laser scanning point cloud data of the air-cooled island is acquired. Based on the point cloud data, a spatial geometric model of the air-cooled tube bundle, support structure, and fan is reconstructed. A thermodynamic coupling equation including steam-side condensation heat transfer and air-side convection heat transfer is established, wherein the heat transfer per unit area on the steam side is... The expression that satisfies this condition is: ; Air-side heat transfer per unit area The expression that satisfies this condition is: ; in, The heat transfer coefficient on the steam side is... The steam saturation temperature For pipe wall temperature, The air-side heat transfer coefficient is... The air inlet temperature; Under steady-state conditions, the heat exchange on both sides is equal, and the simultaneous solution yields... ; The structured dataset is input as a boundary condition into the thermo-coupling equation, and the pipe wall temperature distribution and air flow field are updated at a fixed time step to generate a virtual mapping body that reflects the current operating state.
[0009] As a preferred embodiment of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system of the present invention, the trend prediction module is used to predict the future back pressure change trend and pollution distribution characteristics based on a virtual mapping volume and combined with the pollutant deposition evolution law, through a time-series neural network model. The specific steps are as follows: The depositional evolution equation of pollutants on the fin surface is established, and the expression is as follows: ; in, The mass of accumulated dirt per unit area The equivalent diffusion coefficient is... It is the air velocity vector. Net deposition rate; Will The initial value is set to the residual dirt amount after the last cleaning, based on the wind field in the virtual mapping volume. Temperature field Using the concentration of particulate matter in the environment as input, the equation is solved to obtain the spatial distribution of pollution accumulation in future time periods; Incorporating fouling distribution into the thermal resistance model, the thermal resistance of fin fouling is... and Proportional, the expression is: ; in, For contamination thermal resistance coefficient, total thermal resistance Thermal resistance in a clean state; The heat exchange capacity is updated based on the total thermal resistance, and the future back pressure change trend is calculated in combination with the exhaust steam heat load. The physical model is embedded as prior knowledge into a long short-term memory neural network. The current back pressure, wind speed, ambient temperature, and pollution gradient are input, and the predicted back pressure and pollution intensity map at multiple future time points are output.
[0010] As a preferred embodiment of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system of the present invention, the optimization decision module is used to construct a comprehensive cost objective function based on the back pressure prediction value, combined with the cleaning energy consumption and the fan operating cost, solve the multi-objective optimization problem, and generate a set of cleaning strategies; the specific steps are as follows: The expression for calculating exhaust gas loss is: ; in, The exhaust steam mass flow rate, , For the specific enthalpy and specific entropy of exhaust steam, , Enthalpy and specific entropy under environmental baseline conditions Ambient temperature; Calculate the cleaning volume consumed during the cleaning process. The expression is: ; in, To reduce the power consumption of the cleaning water pump, For the quality of the cleaning water, The γ value of the deionized water; Calculate the driving force consumed during the operation of the wind turbine group Construct a comprehensive objective function The expression is: ; in, , , The weighting coefficients are used; the start time, duration, and intensity level of the cleaning are used as decision variables, and the solution is obtained under the condition of satisfying the back pressure safety limit. The minimum value is used to generate a set of Pareto optimal cleaning strategies.
[0011] As a preferred embodiment of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system of the present invention, the path planning module is used to extract the target area location from the accumulated dirt distribution characteristics according to the cleaning strategy selected in the cleaning strategy set, combine the robot motion constraints, generate the optimal movement path from the initial pose to the target area, and form a control instruction sequence containing execution parameters. The specific steps are as follows: Select the optimal strategy from the set of cleaning strategies and extract the corresponding cleaning area; Based on the dirt accumulation intensity map, identify continuous areas where the dirt accumulation intensity exceeds a preset threshold as target cleaning areas; Obtain the current position of the flexible mobile robot and the entrance to the target area ; Establish a three-dimensional path search space that includes obstacles in the air-cooled tube bundle; Define path cost function The expression is: ; in, This is the total path length. For the number of posture adjustments, To predict energy consumption, , , These are the weighting coefficients; Using the A* algorithm to search Minimum path ; Will Discretize the path points into a sequence and generate a control command sequence, which includes the coordinates of each path point, the cleaning mode, the injection pressure, and the residence time.
[0012] As a preferred embodiment of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system of the present invention, the robot execution module is used to receive a sequence of control commands, drive a flexible mobile robot with a composite adsorption mechanism to move along an optimal movement path, and after reaching a designated position, activate the gas-liquid synergistic cleaning device to complete the fin cleaning, and collect status data during the operation. The specific steps are as follows: The flexible mobile robot is controlled to move from its current position along a path sequence; when it reaches the steel tube bundle area, the electromagnetic chuck is energized to generate an adsorption force. The expression is: ; in, Magnetic flux density The suction cup area is... The vacuum permeability; When it reaches a non-magnetic connection area, the vacuum suction cup activates, and the internal pressure drops to... , generating adsorption force The expression is: ; in, Atmospheric pressure, The effective area of the suction cup; The robot collects angular velocity and acceleration through an inertial measurement unit, and extracts the displacement of feature points by combining visual odometry to update its current pose; Once the device reaches the designated location, it activates the gas-liquid synergistic cleaning device and sprays the gas-liquid mixture according to the set pressure and flow rate for cleaning. During the operation, water consumption, power consumption, and infrared thermal image data are recorded in real time.
[0013] As a preferred embodiment of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system of the present invention, the feedback correction module is used to receive status data during the operation, analyze the changes in heat exchange performance before and after cleaning, calculate the deviation between the actual effect and the predicted result, and dynamically correct the model parameters in the trend prediction module when the deviation exceeds the set range. The specific steps are as follows: Extract the average surface temperature before cleaning from infrared thermal image data Temperature after cleaning Calculate the improvement in heat transfer performance The expression is: ; Obtain the actual back pressure under the same operating conditions after cleaning, compare it with the predicted back pressure for the corresponding time period, and calculate the prediction deviation. The expression is: ; If | If the error exceeds a preset threshold, the model is deemed to have a significant bias. Will The error signal is fed back to the Long Short-Term Memory Neural Network, which adjusts the connection weights of the hidden layer units, updates the model parameters, and makes the subsequent trend prediction results closer to the actual response characteristics, thereby achieving closed-loop optimization of the system.
[0014] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements any step of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system as described in the first aspect of the present invention.
[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system as described in the first aspect of the present invention.
[0016] The beneficial effects of this invention are as follows: By constructing a closed-loop management and control system that integrates data fusion, digital twins, trend prediction, optimization decision-making, path planning, robot execution, and feedback correction, the entire chain of intelligent management of the 600MW unit's air-cooling system, from state perception to autonomous intervention, is realized. A high-fidelity virtual mapping body is generated using multi-source data fusion and thermo-coupling simulation. Combined with a neural network model embedded with physical laws, the evolution of dirt accumulation and back pressure changes are accurately predicted. An optimized cleaning strategy is generated with the goal of minimizing costs, and a flexible robot is driven to autonomously plan paths and execute cleaning operations. The prediction model is dynamically corrected through actual feedback, significantly improving the heat exchange efficiency of the air-cooling system, reducing plant power consumption and cleaning resource consumption, and solving the problems of delayed response, over-cleaning, or under-cleaning in traditional manual cleaning. This achieves a deep integration of energy saving, efficiency improvement, and unmanned operation and maintenance. Attached Figure Description
[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a flowchart of the intelligent control system for the air-cooled energy-saving and efficiency-enhancing robot of the 600MW unit in Example 1. Detailed Implementation
[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.
[0023] Example, refer to Figure 1 As an embodiment of the present invention, this embodiment provides a 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent management and control system, comprising: The module includes a data fusion module, a digital twin module, a trend prediction module, an optimization decision-making module, a path planning module, a robot execution module, and a feedback correction module. The data fusion module is used to collect unit operating parameters, environmental meteorological data and historical cleaning information, perform data alignment and formatting processing, and generate a unified time-series dataset. Furthermore, the unit operating parameters collected by the data fusion module include turbine exhaust flow rate, exhaust temperature, condenser back pressure, fan speed, and inlet and outlet air temperature of each section of the air-cooled island. Environmental meteorological data include ambient temperature, wind speed, wind direction, relative humidity, and atmospheric particulate matter concentration; Historical cleaning information includes the start and end times of each cleaning session, the area to be cleaned, water consumption, power consumption, and changes in back pressure after cleaning. Data alignment is achieved through timestamp matching, and formatting processes unify data from different sources into data record units with a preset structure. It should be noted that the data fusion module solves the heterogeneity problem of multi-source sensor data in terms of sampling frequency, transmission delay and communication protocol by establishing a unified time benchmark, ensuring that the input data relied upon by subsequent modules has spatiotemporal consistency and engineering availability, and providing a basic guarantee for the overall prediction accuracy and decision reliability of the system.
[0024] The digital twin module is used to build a three-dimensional geometric model based on the air-cooled island structure data, establish a dynamic simulation system by combining the thermodynamic and fluid coupling relationship, and input a unified time series dataset into the simulation system to generate a real-time mapped virtual operating state. Furthermore, three-dimensional laser scanning point cloud data of the air-cooled island is acquired. Based on the point cloud data, a spatial geometric model of the air-cooled tube bundle, support structure, and fan is reconstructed. A thermodynamic coupling equation including steam-side condensation heat transfer and air-side convection heat transfer is established, where the heat transfer per unit area on the steam side is... The expression that satisfies this condition is: ; Air-side heat transfer per unit area The expression that satisfies this condition is: ; in, The heat transfer coefficient on the steam side is... The steam saturation temperature For pipe wall temperature, The air-side heat transfer coefficient is... The air inlet temperature; Under steady-state conditions, the heat exchange on both sides is equal, and the simultaneous solution yields... ; The structured dataset is used as the boundary condition input to the thermo-coupling equation, and the pipe wall temperature distribution and air flow field are updated at a fixed time step to generate a virtual mapping volume that reflects the current operating state. It should be noted that the thermo-mechanical coupling equations constructed by the digital twin module fully consider the unsteady heat transfer characteristics of the air-cooled island under complex meteorological conditions. By using actual operating data as boundary conditions to dynamically drive the simulation system, it achieves high-precision reconstruction of the pipe wall temperature field and air flow field, enabling the virtual mapping body to truly reflect the thermo-mechanical distribution state under the current operating conditions and providing reliable physical field input for pollution accumulation prediction.
[0025] The trend prediction module is used to predict future back pressure change trends and pollution distribution characteristics based on virtual operating status and the evolution law of pollutant deposition, using a time-series neural network model. Furthermore, an equation for the depositional evolution of pollutants on the fin surface is established, expressed as: ; in, The mass of accumulated dirt per unit area The equivalent diffusion coefficient is... It is the air velocity vector. Net deposition rate; Will The initial value is set to the residual dirt amount after the last cleaning, based on the wind field in the virtual mapping volume. Temperature field Using the concentration of particulate matter in the environment as input, the equation is solved to obtain the spatial distribution of pollution accumulation in future time periods; Incorporating fouling distribution into the thermal resistance model, the thermal resistance of fin fouling is... and Proportional, the expression is: ; in, For contamination thermal resistance coefficient, total thermal resistance Thermal resistance in a clean state; The heat exchange capacity is updated based on the total thermal resistance, and the future back pressure change trend is calculated in combination with the exhaust steam heat load. The physical model is embedded as prior knowledge into a long short-term memory neural network. The current back pressure, wind speed, ambient temperature, and pollution gradient are input, and the predicted back pressure and pollution intensity map at multiple future time points are output. It should be noted that the trend prediction module embeds the physical evolution law of pollutant deposition as prior knowledge into the neural network structure, effectively constraining the model training process and avoiding prediction instability of pure data-driven methods in scenarios with sudden changes in operating conditions or sparse data. This significantly improves the generalization prediction ability of long-term pollution accumulation trends and back pressure changes, and enhances the system's adaptability in complex environments.
[0026] The optimization decision module is used to calculate the exhaust steam loss based on the back pressure change trend, combine the cleaning energy consumption and the fan operation energy consumption to construct a comprehensive energy cost objective function, solve the multi-objective optimization problem, and generate a set of cleaning strategies. Furthermore, the exhaust gas loss is calculated using the following expression: ; in, The exhaust steam mass flow rate, , For the specific enthalpy and specific entropy of exhaust steam, , Enthalpy and specific entropy under environmental baseline conditions Ambient temperature; Calculate the cleaning volume consumed during the cleaning process. The expression is: ; in, To reduce the power consumption of the cleaning water pump, For the quality of the cleaning water, The γ value of the deionized water; Calculate the driving force consumed during the operation of the wind turbine group Construct a comprehensive objective function The expression is: ; in, , , The weighting coefficients are used; the start time, duration, and intensity level of the cleaning are used as decision variables, and the solution is obtained under the condition of satisfying the back pressure safety limit. The minimum value is used to generate a set of Pareto optimal cleaning strategies; It should be noted that the optimization decision module uses efficiency as the core evaluation index, and uniformly quantifies exhaust energy loss, cleaning resource consumption and fan operating energy consumption. It breaks through the limitations of traditional methods that only focus on a single energy consumption or back pressure threshold. Through multi-objective optimization, it generates cleaning strategies that take into account both economy and energy saving, and realizes the transformation from experience-driven to energy-efficient decision-making.
[0027] The path planning module is used to extract the target area location from the characteristics of the accumulated dirt distribution according to the selected cleaning strategy, combine the robot motion constraints, generate the optimal movement path from the starting pose to the target area, and form a sequence of control instructions containing execution parameters. Furthermore, the optimal strategy is selected from the set of cleaning strategies, and the corresponding cleaning area is extracted. Based on the dirt accumulation intensity map, identify continuous areas where the dirt accumulation intensity exceeds a preset threshold as target cleaning areas; Obtain the current position of the flexible mobile robot and the entrance to the target area ; Establish a three-dimensional path search space that includes obstacles in the air-cooled tube bundle; Define path cost function The expression is: ; in, This is the total path length. For the number of posture adjustments, To predict energy consumption, , , These are the weighting coefficients; Using the A* algorithm to search Minimum path ; Will Discretize the path point sequence to generate a control command sequence, which includes the coordinates of each path point, cleaning mode, injection pressure and residence time; It should be noted that the path planning module comprehensively considers the robot's kinematic constraints and operational energy consumption in a complex three-dimensional structure. By balancing path length, attitude adjustment frequency, and energy consumption through a weighted cost function, the generated optimal path not only meets the accessibility requirements but also effectively extends the robot's single-operation endurance and improves the overall execution efficiency of the cleaning task.
[0028] The robot execution module is used to receive control command sequences, drive the flexible mobile robot with a composite adsorption mechanism to move along the optimal movement path, and after reaching the designated position, activate the gas-liquid synergistic cleaning device to complete the fin cleaning, and collect status data during the operation. Furthermore, the flexible mobile robot is controlled to move from its current position along a sequence of path points; when it reaches the steel tube bundle area, the electromagnetic chuck is energized to generate an adsorption force. The expression is: ; in, Magnetic flux density The suction cup area is... The vacuum permeability; When it reaches a non-magnetic connection area, the vacuum suction cup activates, and the internal pressure drops to... , generating adsorption force The expression is: ; in, Atmospheric pressure, The effective area of the suction cup; The robot collects angular velocity and acceleration through an inertial measurement unit, and extracts the displacement of feature points by combining visual odometry to update its current pose; Once the device reaches the designated location, it activates the gas-liquid synergistic cleaning device and sprays the gas-liquid mixture according to the set pressure and flow rate for cleaning. During operation, water consumption, power consumption, and infrared thermal image data are recorded in real time. It should be noted that the robot's execution module adopts a composite adsorption mechanism of electromagnetic and vacuum to ensure stable attachment on both steel surfaces and non-magnetic connection areas. Combined with the fusion positioning technology of inertial measurement and visual odometry, it achieves high-precision pose estimation in highly turbulent industrial environments, ensuring reliable execution of cleaning operations and accurate acquisition of process data.
[0029] The feedback correction module is used to receive status data, analyze the changes in heat exchange performance before and after cleaning, calculate the deviation between the actual effect and the predicted result, and dynamically correct the model parameters in the trend prediction module when the deviation exceeds the set range. Furthermore, the average surface temperature before cleaning was extracted from the infrared thermal image data. Temperature after cleaning Calculate the improvement in heat transfer performance The expression is: ; Obtain the actual back pressure under the same operating conditions after cleaning, compare it with the predicted back pressure for the corresponding time period, and calculate the prediction deviation. The expression is: ; If | If the error exceeds a preset threshold, the model is deemed to have a significant bias. Will The error signal is fed back to the long short-term memory neural network, which adjusts the connection weights of the hidden layer units of the network, completes the model parameter update, and makes the subsequent trend prediction results closer to the actual response characteristics, thereby achieving closed-loop optimization of the system. It should be noted that the feedback correction module establishes a closed-loop feedback mechanism between the cleaning effect and the model prediction. By correcting the neural network weights online, it continuously improves the accuracy of the trend prediction model in the actual operating environment, effectively copes with uncertainties in long-term operation such as equipment aging and environmental changes, and ensures that the system has the ability to continuously optimize and evolve intelligently.
[0030] This embodiment also provides a computer device applicable to the intelligent management and control system of a 600MW unit air-cooled energy-saving and efficiency-enhancing robot, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the intelligent management and control system of a 600MW unit air-cooled energy-saving and efficiency-enhancing robot as proposed in the above embodiment.
[0031] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.
[0032] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the intelligent control system for energy-saving and efficiency-enhancing robotic system for air-cooled units as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.
[0033] In summary, this invention achieves intelligent management of the entire chain of the 600MW unit's air-cooling system, from state perception to autonomous intervention, by constructing a closed-loop management system that integrates data fusion, digital twins, trend prediction, optimization decision-making, path planning, robot execution, and feedback correction. It utilizes multi-source data fusion and thermo-coupled simulation to generate a high-fidelity virtual mapping body, and combines a neural network model embedded with physical laws to accurately predict the evolution of dirt accumulation and back pressure changes. An optimized cleaning strategy is generated with the goal of minimizing costs, driving a flexible robot to autonomously plan paths and execute cleaning operations. Through dynamic correction of the prediction model based on actual feedback, the invention significantly improves the heat exchange efficiency of the air-cooling system, reduces plant power consumption and cleaning resource consumption, and solves the problems of delayed response, over-cleaning, or under-cleaning in traditional manual cleaning. This achieves a deep integration of energy saving, efficiency improvement, and unmanned operation and maintenance.
[0034] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system, characterized in that: include: The module includes a data fusion module, a digital twin module, a trend prediction module, an optimization decision-making module, a path planning module, a robot execution module, and a feedback correction module. The data fusion module is used to collect unit operating parameters, environmental meteorological data and historical cleaning information, perform data alignment and formatting processing, and generate a unified time-series dataset. The digital twin module is used to construct a three-dimensional geometric model based on the air-cooled island structure data, establish a dynamic simulation system by combining the thermodynamic and fluid coupling relationship, and input the unified time series dataset into the simulation system to generate a real-time mapped virtual operating state. The trend prediction module is used to predict future back pressure change trends and pollution distribution characteristics based on the virtual operating state and the pollutant deposition evolution law, using a time-series neural network model. The optimization decision module is used to calculate the exhaust steam loss based on the back pressure change trend, combine the cleaning energy consumption and the fan operation energy consumption to construct a comprehensive energy cost objective function, solve the multi-objective optimization problem, and generate a set of cleaning strategies. The path planning module is used to extract the target area location from the characteristics of the accumulated dirt distribution according to the selected cleaning strategy, combine the robot motion constraints, generate the optimal movement path from the initial pose to the target area, and form a control instruction sequence containing execution parameters. The robot execution module is used to receive the control command sequence, drive the flexible mobile robot with a composite adsorption mechanism to move along the optimal movement path, and after reaching the designated position, activate the gas-liquid synergistic cleaning device to complete the fin cleaning, and collect status data during the operation. The feedback correction module is used to receive the status data, analyze the changes in heat exchange performance before and after cleaning, calculate the deviation between the actual effect and the predicted result, and dynamically correct the model parameters in the trend prediction module when the deviation exceeds the set range.
2. The 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system as described in claim 1, characterized in that: The data fusion module collects unit operating parameters including turbine exhaust flow rate, exhaust temperature, condenser back pressure, fan speed, and inlet and outlet air temperatures of each section of the air-cooled island. The environmental meteorological data includes ambient temperature, wind speed, wind direction, relative humidity, and atmospheric particulate matter concentration. The historical cleaning information includes the start and end times of each cleaning session, the cleaning area, water consumption, power consumption, and back pressure changes after cleaning. The data alignment is achieved through timestamp matching, and the formatting process unifies data from different sources into data record units with a preset structure.
3. The 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system as described in claim 2, characterized in that: The digital twin module is used to construct a three-dimensional geometric model based on the air-cooled island structure data, establish a dynamic simulation system by combining the thermodynamic and fluid coupling relationship, and input the structured dataset into the simulation system to generate a real-time mapped virtual operating state. The specific steps are as follows: Three-dimensional laser scanning point cloud data of the air-cooled island is acquired. Based on the point cloud data, a spatial geometric model of the air-cooled tube bundle, support structure, and fan is reconstructed. A thermodynamic coupling equation including steam-side condensation heat transfer and air-side convection heat transfer is established, wherein the heat transfer per unit area on the steam side is... The expression that satisfies this condition is: ； Air-side heat transfer per unit area The expression that satisfies this condition is: ； in, The heat transfer coefficient on the steam side is... The steam saturation temperature For pipe wall temperature, The air-side heat transfer coefficient is... The air inlet temperature; Under steady-state conditions, the heat exchange on both sides is equal, and the simultaneous solution yields... ; The structured dataset is input as a boundary condition into the thermo-coupling equation, and the pipe wall temperature distribution and air flow field are updated at a fixed time step to generate a virtual mapping body that reflects the current operating state.
4. The 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system as described in claim 3, characterized in that: The trend prediction module is used to predict future back pressure change trends and pollution distribution characteristics based on a virtual mapping volume and combined with the pollutant deposition evolution law, using a time-series neural network model. The specific steps are as follows: The depositional evolution equation of pollutants on the fin surface is established, and the expression is as follows: ; in, The mass of accumulated dirt per unit area The equivalent diffusion coefficient is... It is the air velocity vector. Net deposition rate; Will The initial value is set to the residual dirt amount after the last cleaning, based on the wind field in the virtual mapping volume. Temperature field Using the concentration of particulate matter in the environment as input, the equation is solved to obtain the spatial distribution of pollution accumulation in future time periods; Incorporating fouling distribution into the thermal resistance model, the thermal resistance of fin fouling is... and Proportional, the expression is: ； in, For contamination thermal resistance coefficient, total thermal resistance Thermal resistance in a clean state; The heat exchange capacity is updated based on the total thermal resistance, and the future back pressure change trend is calculated in combination with the exhaust steam heat load. The physical model is embedded as prior knowledge into a long short-term memory neural network. The current back pressure, wind speed, ambient temperature, and pollution gradient are input, and the predicted back pressure and pollution intensity map at multiple future time points are output.
5. The 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system as described in claim 4, characterized in that: The optimization decision module is used to construct a comprehensive cost objective function based on the predicted back pressure value, combined with the cleaning energy consumption and the fan operating cost, solve the multi-objective optimization problem, and generate a set of cleaning strategies. The specific steps are as follows: The expression for calculating exhaust gas loss is: ; in, The exhaust steam mass flow rate, , For the specific enthalpy and specific entropy of exhaust steam, , Enthalpy and specific entropy under environmental baseline conditions Ambient temperature; Calculate the cleaning volume consumed during the cleaning process. The expression is: ； in, To reduce the power consumption of the cleaning water pump, For the quality of the cleaning water, The γ value of the deionized water; Calculate the driving force consumed during the operation of the wind turbine group Construct a comprehensive objective function The expression is: ; in, , , The weighting coefficients are used; the start time, duration, and intensity level of the cleaning are used as decision variables, and the solution is obtained under the condition of satisfying the back pressure safety limit. The minimum value is used to generate a set of Pareto optimal cleaning strategies.
6. The 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system as described in claim 5, characterized in that: The path planning module is used to extract the target area location from the dirt distribution characteristics based on the selected cleaning strategy in the cleaning strategy set, combine it with the robot motion constraints, generate the optimal movement path from the initial pose to the target area, and form a control instruction sequence containing execution parameters. The specific steps are as follows: Select the optimal strategy from the set of cleaning strategies and extract the corresponding cleaning area; Based on the dirt accumulation intensity map, identify continuous areas where the dirt accumulation intensity exceeds a preset threshold as target cleaning areas; Obtain the current position of the flexible mobile robot and the entrance to the target area ; Establish a three-dimensional path search space that includes obstacles in the air-cooled tube bundle; Define path cost function The expression is: ; in, This is the total path length. For the number of posture adjustments, To predict energy consumption, , , These are the weighting coefficients; Using the A* algorithm to search Minimum path ; Will Discretize the path points into a sequence and generate a control command sequence, which includes the coordinates of each path point, the cleaning mode, the injection pressure, and the residence time.
7. The 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system as described in claim 6, characterized in that: The robot execution module is used to receive a sequence of control commands, drive the flexible mobile robot equipped with a composite adsorption mechanism to move along the optimal movement path, and after reaching the designated position, activate the gas-liquid synergistic cleaning device to complete the fin cleaning, and collect status data during the operation. The specific steps are as follows: The flexible mobile robot is controlled to move from its current position along a path sequence; when it reaches the steel tube bundle area, the electromagnetic chuck is energized to generate an adsorption force. The expression is: ; in, Magnetic flux density The suction cup area is... The vacuum permeability; When it reaches a non-magnetic connection area, the vacuum suction cup activates, and the internal pressure drops to... , generating adsorption force The expression is: ; in, Atmospheric pressure, The effective area of the suction cup; The robot collects angular velocity and acceleration through an inertial measurement unit, and extracts the displacement of feature points by combining visual odometry to update its current pose; Once the device reaches the designated location, it activates the gas-liquid synergistic cleaning device and sprays the gas-liquid mixture according to the set pressure and flow rate for cleaning. During the operation, water consumption, power consumption, and infrared thermal image data are recorded in real time.
8. The 600MW unit air-cooled energy-saving and efficiency-enhancing robotic intelligent control system as described in claim 7, characterized in that: The feedback correction module is used to receive status data during the operation, analyze the changes in heat exchange performance before and after cleaning, calculate the deviation between the actual effect and the predicted result, and dynamically correct the model parameters in the trend prediction module when the deviation exceeds the set range. The specific steps are as follows: Extract the average surface temperature before cleaning from infrared thermal image data Temperature after cleaning Calculate the improvement in heat transfer performance The expression is: ； Obtain the actual back pressure under the same operating conditions after cleaning, compare it with the predicted back pressure for the corresponding time period, and calculate the prediction deviation. The expression is: ； If | If the error exceeds a preset threshold, the model is deemed to have a significant bias. Will The error signal is fed back to the Long Short-Term Memory Neural Network, which adjusts the connection weights of the hidden layer units, updates the model parameters, and makes the subsequent trend prediction results closer to the actual response characteristics, thereby achieving closed-loop optimization of the system.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system as described in any one of claims 1 to 8.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the 600MW unit air-cooled energy-saving and efficiency-enhancing robot intelligent management and control system as described in any one of claims 1 to 8.