1093 results about "Plant" patented technology

The invention belongs to the technical field of robot control, particularly provides a force feedback teleoperation control system and method for the tail end of a mechanical arm of a quadruped robot, and aims at the key challenges that control errors are caused by communication time delay and soft obstacles are difficult to recognize in a dynamic environment. Modeling and judgment are conducted on the contact state of the tail end of the mechanical arm in advance, and feedforward control and buffer adjustment oriented to communication time delay are achieved. And meanwhile, a dynamic semantic map is constructed in combination with multi-source sensing information, and soft obstacle reasoning and path optimization are performed by fusing a tail end force sense change trend, so that the recognition and avoidance capabilities of the system in a complex and invisible obstacle environment are remarkably improved. The system has good perspectiveness, self-adaptability and high redundancy safety characteristics, is suitable for multi-task inspection operation of industrial sites such as a thermal power plant, and is especially suitable for a remote man-machine cooperative operation scene in a narrow space.

The invention relates to a power plant temperature control system safety protection device and system. The power plant temperature control system safety protection device comprises a redundant temperature acquisition module, a temperature control safety processing module, an actuator control protection module, an intelligent strategy recovery module and a communication and alarm module. Through multi-point temperature data acquisition, abnormal trend intelligent identification and local linkage control, the device realizes quick response and protection control of temperature control abnormity of a key area of a boiler thermodynamic system. The system supports data interconnection with a DCS platform and an alarm analysis platform, has local independent operation and remote centralized management capabilities, and is suitable for temperature control safety protection of high heat load areas such as boilers of coal-fired power plants.

The invention relates to the technical field of boiler control, and discloses a gas-fired boiler starting control method and system. The method comprises the following steps: acquiring initial operation parameters such as hearth pressure, water inlet temperature, flue gas oxygen content and fuel valve opening of the gas-fired boiler, establishing a thermodynamic state model in a starting stage according to the initial operation parameters, and dividing dynamic response thresholds of different load intervals; and calculating a theoretical fuel supply rate of the combustor based on the model, generating and dynamically updating a fuel supply correction coefficient in combination with a threshold value, synchronously marking key nodes according to abnormal distribution of hearth temperature, and adjusting a spatial air distribution strategy of the combustor. And after the correction coefficient is subjected to multi-stage buffer smoothing, a fuel control instruction is output to an executing mechanism. And switching to steady state control after the boiler reaches the preset load. The method adapts to the characteristics of a power plant unit, so that the starting process is more stable, the damage to a heated surface and pollutant emission are reduced, and rapid grid connection is assisted.

The invention relates to the technical field of power plant intelligent adjustment, and discloses a power plant load intelligent adjustment method and system. The method comprises the following steps: distributing and deploying sensors according to field equipment of a power plant, and constructing a distributed sensor network based on the sensors; establishing a single time scale control framework according to the distributed sensor network; a heterogeneous delay compensation mechanism is constructed based on a distributed sensor network and a single time scale control framework; constructing a local linear matrix inequality constraint condition of each node, obtaining an optimal solution of the local linear matrix inequality constraint condition according to a heterogeneous delay compensation mechanism, and then generating an optimal control strategy of each node; the method comprises the following steps: constructing a global Lyapunov function based on a distributed sensor network and a single time scale control framework; and the change rate of the global Lyapunov function is monitored in real time, and corresponding equipment is intelligently and dynamically adjusted according to the optimal control strategy. The method achieves the intelligent regulation and control of the power plant load, is good in stability, is high in adaptability, and is short in regulation and control period.

The invention discloses a thermal power plant inspection robot path planning method, device and equipment and a storage medium, and the method comprises the steps: carrying out the semantic analysis and information extraction of an inspection task instruction, an equipment document and a historical inspection log through a large language model; obtaining a semantic importance score, a perception modal demand, recommended residence time and a shooting angle of each inspection target; establishing a multi-target path optimization model based on an improved ant colony optimization algorithm, and inputting the semantic importance score, the perception modal demand, the recommended residence time and the shooting angle into the multi-target path optimization model to obtain a joint scheduling sequence; the thermal power plant inspection robot is controlled to execute the joint scheduling sequence to perform autonomous navigation and perception tasks, the current path planning is adjusted according to the task completion condition, the intelligence, autonomy and adaptability level of the inspection robot can be remarkably improved, and the speed and efficiency of path planning of the thermal power plant inspection robot are improved.

A control device (100) for controlling a number N, with N ≥ 1, of actuators (200) for actuating N optical elements of an optical system, comprising a control node (K2) coupling to the actuator (200) for providing a control voltage (V2) for the actuator (200), an amplifier (110) configured to receive a supply voltage (V1) provided at an input node (K1) and to provide an output voltage (V3) at an output node (K3), wherein the amplifier (110) comprises a plurality M, with M ≥ 2, of current sources (111, 112) comprising at least one charge-controlled current source (112) and a holding capacitor (113) for maintaining the level of the provided output voltage (V3), and a buffer coupled between the output node (K3) and the control node (K2). (120), which is designed to,to decouple the control node (K2) from the output node (K3) and to provide the control voltage (V2) for the actuator (200) based on the level of the output voltage (V3) of the holding capacitor (113).

Systems and methods for controlling industrial process automation and control systems can automatically, through the use of machine learning (ML) models and algorithms, extract plant assets from engineering diagrams and other plant engineering data sources. The systems and methods can establish asset relationships to create a plant asset registry and build an asset hierarchy from the plant assets. The systems and methods can generate an ontological knowledge base from the plant asset hierarchy, and provide an HMI for controlling the industrial process based on the plant asset hierarchy and the ontological knowledge base.

The invention relates to a multi-modal oscillation suppression method and system for a network construction type power supply grid-connected system, belongs to the technical field of power system control, and solves the problem that multi-modal oscillation of a large-scale inversion type power supply grid-connected system cannot be cooperatively and rapidly suppressed. The method comprises the following steps: when a system generates broadband oscillation, collecting grid-connected point voltage and current signals of each power station to extract a dominant oscillation mode of each power station; constructing a modal coupling matrix based on the dominant oscillation mode to extract modal coupling strength; identifying a key path of oscillation propagation in the dominant oscillation mode, and extracting a path weight and a power station participation degree; calculating the comprehensive power regulation and control weight of each power generation station in multiple modes so as to determine the power regulation quantity; and distributing the power regulation quantity to an energy storage device, a static var generator (SVG) and each power generation station according to oscillation characteristics of different frequency bands so as to cooperatively suppress multi-modal oscillation. The active / reactive oscillation energy is cooperatively adjusted, so that the oscillation energy of each frequency band is effectively suppressed.

The invention discloses a regulation and control method for stably improving the water quality of water produced by a reverse osmosis system of a power plant, and relates to the field of water quality regulation and control, which comprises the following steps: configuring a water inlet flow regulation device, performing initial setting, configuring the position of an online sensor, performing data acquisition, setting judgment rules of turbidity, differential pressure and transmembrane pressure difference, and determining the water quality. And configuring a flocculant adding device and adding control logic, configuring a backwashing and adding linkage control state machine, and carrying out data smoothing, trend analysis and self-adaptive threshold adjustment. When the water quality of the raw water fluctuates for a short time or the pollution load is increased, dynamically adjusting the water inlet flow, the backwashing intensity and the flocculant adding concentration based on parameters such as turbidity, pollution index, conductivity and pressure difference which are acquired by an online sensor in real time; and meanwhile, the concentrated water reflux proportion and the high-pressure pump frequency are automatically corrected by continuously calculating the desalting rates of the first-stage and second-stage reverse osmosis devices, so that the running state of the membrane component is kept in an optimal interval.

The invention discloses a wind power plant inertia damping adaptive regulation and control method and system fused with a deep Q network, and belongs to the technical field of power systems. The method comprises the following steps: establishing a doubly-fed wind turbine generator, a converter mathematical model and an interaction model of a wind power system connected to a power grid, and analyzing the grid-connected stability of the wind power system; the dynamic response characteristic of providing inertia support by the VSG system under different conditions is researched, and the primary frequency modulation capability is improved; a depth Q network is adopted to select rotational inertia and a damping coefficient, a damping inertia adaptive control strategy is improved, cooperation of inertia support and frequency response capability improvement of a wind turbine generator and continuous and smooth adjustment of active power is realized, the stability of a power grid is improved, and the operation performance of a wind power plant is optimized. According to the method, dynamic optimal matching of inertia response, primary frequency modulation and the running state of a unit is achieved, auxiliary services are provided, meanwhile, the running safety and service life of a draught fan are guaranteed, and smooth and impact-free adjustment of active power is achieved.

State of the art systems used for industrial plant monitoring have the disadvantage that they fail to correctly assess reason for dip in performance of the plant and in turn trigger appropriate corrective measures. The disclosure herein generally relates to industrial plant monitoring, and, more particularly, to a system and method for development and deployment of self-organizing cyber-physical systems for manufacturing industries. The system monitors and collects data with respect to various parameters, from the industrial plant. If any performance dip is detected, the system determines corresponding cause, and also triggers one or more corrective actions to improve performance of the plant and different plant components to a desired performance level.

The invention relates to the technical field of coal-fired power plant environmental protection, in particular to a coal-fired power plant intelligent environmental protection island control system based on multi-source data fusion. Comprising a multi-source data acquisition module, a data transmission module, a data preprocessing module, a multi-source data fusion module, an operation state analysis module, an intelligent control decision module, a control instruction execution module and a man-machine interaction module. According to the coal-fired power plant intelligent environmental protection island control system based on multi-source data fusion, equipment operation, environmental monitoring, power grid, fuel and meteorological data are widely acquired through the multi-source data acquisition module, the limitation of a single data source of a traditional system is broken through, a distributed Kalman filtering algorithm is combined with a neural network algorithm for data fusion, and the data fusion efficiency is improved. According to the method, the incidence relation between data can be deeply mined, temperature and pressure parameters of desulfurization equipment and flue gas sulfur dioxide concentration data are subjected to fusion analysis, the reason of desulfurization efficiency reduction can be accurately positioned, and compared with traditional single data source judgment, the fault diagnosis accuracy is improved.

The invention relates to the technical field of space life science, in particular to a multi-axis control and acoustic levitation combined microgravity environment simulation closed cultivation device and a control system.The multi-axis control and acoustic levitation combined microgravity environment simulation closed cultivation device comprises two vertical sliding rail mechanisms, a first rotating servo motor is arranged on each vertical sliding rail mechanism in a sliding mode, and an outer rotating frame is driven to rotate in the first direction; the second rotating servo motor is fixed to the outer rotating frame and drives the cultivation chamber connected with the inner rotating frame to rotate in the second direction, and the acoustic levitation device is installed in the cultivation chamber. According to the invention, random gravity disturbance is introduced through random sliding on the sliding rail, and a microgravity environment with coupled random gravity and rotating speed double factors is constructed in combination with random rotation of double shafts; according to the method, the influence of microgravity on plant leaf respiration under the condition of no heat convection and air shear force can be deeply researched, the influence of air heat convection and shear force can be inhibited, the space microgravity cultivation condition can be truly simulated, and a reliable platform is provided for ground plant space cultivation research.

The invention provides an analysis method and device for flow, efficiency and water consumption rate of a hydraulic power plant generator set. A distributed flow measurement unit based on a domestic PLC controller and a domestic operating system is constructed, and each unit is independently configured with the flow measurement unit to realize decentralization of data acquisition and processing; active power, guide vane opening, water head difference, instantaneous flow and volute inlet pressure data are collected in real time through a distributed flow measurement unit, and a unit efficiency and water consumption rate calculation model is dynamically fitted based on the data; sending an output result of the calculation model to a monitoring system through an anti-interference communication network, wherein the communication network adopts layered wiring and a double-shielded twisted pair to reduce the signal bit error rate; and the IMC integrated management and control platform is used for remote monitoring and diagnosis of the flow measurement unit, and seamless connection with a smart power plant data medium station is realized. The system reliability and the maintenance flexibility are improved, and the autonomous controllable requirements of domestic substitution and intelligent power plants are comprehensively met.

The invention belongs to the field of metallurgy and coal, and relates to an automatic control model for feeding and discharging of a coal bunker. The coal bunker feeding and discharging automatic control model is designed according to the blast furnace coal injection procedure of an iron making plant of an iron and steel enterprise, and automation and intelligentization of the whole coal bunker feeding and discharging process are achieved. A sensor, an automatic control system and a data analysis technology are integrated in the model, the material level of the coal bunker is monitored in real time through a material level indicator, and feeding and discharging strategies are automatically adjusted in combination with multi-level priority control logic. The system is provided with a dynamic switching mechanism, feeding targets can be automatically switched according to the material level of the coal bunker, the coal grinding speed is accurately controlled, and material breakage and material overflowing are prevented. In addition, the model has a safety interlocking protection function, and production safety is ensured. Through application of the model, the labor intensity of workers is remarkably reduced, the production efficiency and stability are improved, resource waste and equipment abrasion are reduced, the requirements of the modern industry for automatic and intelligent production are met, and sustainable development of enterprises is assisted.

The present application relates to the technical field of thermal power plants, and discloses an intelligent operation method and system for a thermal power plant, a device, and a storage medium. The method comprises: measuring operation parameters of a thermal power plant in real time by means of advanced detection and soft sensing techniques; establishing a fault monitoring model to monitor the operation process of the thermal power plant on the basis of the real-time detected operation parameters, and generating an early warning signal when a fault occurs; performing fault diagnosis, and generating a fault self-healing instruction; and evaluating the operation status of the thermal power plant by means of a performance evaluation model, and then obtaining optimal operation modes and optimal operation parameter setpoints by means of a multi-objective optimization algorithm. The present application achieves high-performance real-time intelligent closed-loop control, improving the adaptability to complex working conditions and the disturbance rejection capability of a thermal power plant production process control system, and the overall performance of a thermal power plant production process; and efficient human-computer interaction is achieved by means of artificial intelligence-based early warning diagnosis and closed-loop self-healing.

The invention relates to the technical field of dressing plant control, in particular to a LightGBM-based one-key ore grinding grading control method, which comprises the following steps of: preprocessing collected iron ore grinding production data through an industrial internet of things to obtain production data, constructing a data-driven dynamic prediction model by utilizing a LightGBM algorithm, and combining time-space lag cross-correlation and flattening data input to obtain a one-key ore grinding grading control model. And deploying the trained lightweight model to a PLC control layer, and completing optimization and adjustment of corresponding key process parameters. The method has the advantages that by collecting production data in real time and utilizing the LightGBM algorithm to construct the dynamic prediction model, key variables such as the rotating speed and the ore feeding amount of the mill can be accurately and dynamically adjusted, the ore grinding process is optimized, no-load and overload operation of the mill is reduced, and compared with a traditional control method, the ore grinding efficiency is remarkably improved; operators only need to set target parameters, key parameters such as the rotating speed and the ore feeding amount of the mill are optimized in real time through the LightGBM model, it is ensured that the ore grinding fineness stably reaches the standard, and frequent intervention and adjustment are not needed.

The invention belongs to the crossing field of artificial intelligence and industrial automation, particularly relates to an aeration fan frequency conversion group control energy-saving method based on an intelligent optimization algorithm, and aims to solve the problems of high energy consumption, response lag and large equipment loss of a traditional control mode. According to the method, parameters such as dissolved oxygen, flow and pressure are collected in real time, a nonlinear optimization model with minimum energy consumption, accurate oxygen supply and minimum start and stop as multiple targets is constructed, an optimal rotating speed distribution scheme is solved online by adopting an improved particle swarm algorithm, and a smooth slope instruction output and closed-loop feedback correction mechanism is combined, so that the optimal rotating speed distribution scheme is obtained. And dynamic self-adaptive adjustment of fan group control is realized. The energy-saving efficiency is remarkably improved by 18%-25%, the annual electricity is saved by more than 800000 kilowatt-hour, meanwhile, the load unevenness is reduced to below 8%, fan start and stop is reduced by 60%, the service life of equipment is prolonged by 2-3 years, hardware does not need to be newly added, the deployment period is short, investment recovery is fast, and the method is successfully applied to multiple sewage plants.

The invention discloses a thermal power plant limited space operation safety and environment multi-parameter monitoring system, and relates to the technical field of operation safety monitoring. A structure parameter model is constructed based on a space volume structure, a gas exchange path, a risk type and an operation position, and multi-parameter sensing nodes are configured; extracting abnormal fluctuation characteristics of the environmental parameters through edge computing nodes; a gas flow model and a heat distribution model are combined to carry out space-time inversion, a risk source is positioned, a space risk heat map is dynamically generated, the position and action state of an operator can be obtained in real time, the exposure probability is calculated, and automatic identification and graded early warning of a high-risk area are realized; the system has high-precision, low-delay and full-coverage monitoring capability, and the intrinsic safety of limited space operation of the thermal power plant is improved.

The invention discloses a novel quick-change mechanism with a flow control function for a tundish. The novel quick-change mechanism comprises a mounting part, a sliding part, a pressing part, a first driving part and a second driving part. And the functions of the stopper rod opening and closing device are integrated on the tundish quick-change nozzle device, so that the investment, the installation site, the maintenance cost, the spare part site and the cost of a set of equipment are saved. The two sets of equipment are combined into a whole, so that the number of operators in a steel mill is reduced, and meanwhile, preparation is made in advance for the subsequent steel mill if intelligent and automatic control over steel-making continuous casting is adopted, and the positive effect is achieved.

The invention discloses a landscape plant precise maintenance control method based on multi-source perception and a growth model, particularly relates to the technical field of automatic control systems, and is used for solving the problem that a set value in an existing landscape plant maintenance control system is kept constant and cannot adapt to dynamic growth requirements of plants. The optimal environment set value of the current growth stage is calculated by collecting growth environment parameters and plant physiological state parameters, a multivariable time sequence causal network is constructed based on the parameters, the time sequence evolution entropy is calculated, and the change trend of the set value is monitored. Identifying a conflict relationship among different maintenance targets, analyzing a coordination priority through a key causal path to generate a coordinated maintenance strategy, and generating a set value smooth switching track when a time sequence evolution entropy exceeds an adaptive threshold value and a set value change trend is consistent with a phenological period prediction direction; and finally, the controller is adjusted to output an instruction to drive an execution mechanism to carry out maintenance operation according to the updated set value, and optimization of the control system on the whole life cycle of plant growth is achieved.

The invention belongs to the technical field of industrial control, and discloses a thermal power plant coal conveying system intelligent monitoring system based on multi-source perception and edge calculation. Comprising the steps of collecting multi-modal sensing data through a heterogeneous sensor array, performing deep collaborative extraction on cross-modal features such as vision, acoustics, vibration, temperature and current by using a physical constraint correlation model, and dynamically identifying an equipment abnormal mode according to the cross-modal features; and a hierarchical fusion analysis architecture is introduced, so that multi-level confidence guarantee is provided for fault diagnosis. Through a simulation deduction mechanism driven by digital twinning, system response can be pre-evaluated before a control strategy is executed, a scheme with the optimal comprehensive performance is optimized, and the reliability of a control decision is ensured; an increment correction mechanism of state deviation real-time feedback is also realized, and control deviation caused by model mismatch is avoided; and each abnormal symptom of the coal conveying link can be accurately captured and adaptively handled, so that the system safety and the operation and maintenance intelligent level are remarkably improved.

The invention provides a power plant combustion optimization method, system and device and a program product. The method comprises the steps that historical operation data of a power plant are collected; constructing a self-adaptive physical information coupling model and obtaining a target state parameter calculation value; training a coupling model based on historical operation data; wherein the data loss item of the loss function is constructed based on the target state parameter measured value and the target state parameter predicted value, the physical loss item of the loss function is constructed based on the target state parameter measured value and the target state parameter calculated value, and the coupling model is trained based on the loss function; acquiring real-time operation data of the power plant, and inputting the real-time operation data into the coupling model to obtain a target state parameter predicted value output by the coupling model at the next moment; and obtaining an optimal control quantity based on the target state parameter predicted value at the next moment, and controlling the working state of the actuator dynamic cluster based on the optimal control quantity. According to the invention, fine tuning of the power plant is realized.

A computer-implemented method for generating control code for actuating an asset in an industrial plant to perform a given task includes obtaining candidate control code from an automatic code generator based at least in part on the given task; performing code validation on the candidate control code to determine whether it is capable of being executed, and / or capable of being compiled for execution; after successful code validation, determining whether execution of the candidate control code is capable of actuating the asset to perform the given task by executing the candidate control code in a simulation environment; and when this determination is positive, determining the candidate control code as the sought control code.

The invention belongs to the technical field of power plant thermodynamic and control, and particularly relates to a fractional order model of an ultra-supercritical coal-fired unit coordination control system and an identification method. The fractional order model comprises a coal pulverizing system, a boiler steam-water system and a steam turbine power generation system; the modeling method comprises the following steps: S1, acquiring and storing data of an ultra-supercritical coal-fired unit coordinated control system DCS; s2, constructing a fractional order model of the coordination control system; and S3, fitting a to-be-solved function through acquired data, and identifying dynamic parameters of the fractional order model offline by adopting a segmented time-sharing adaptive particle swarm method. According to the method, fractional order modeling is carried out on the coordinated control system of the coal-fired unit, the dynamic parameters of the system are identified by adopting a segmented time-sharing adaptive particle swarm identification method, the optimal model parameters are obtained through iterative calculation, the precision of the unit coordinated control system model under the wide load is greatly improved, and the system performance is improved. And implementation of a control strategy based on a system model is facilitated.

The invention provides a factory equipment control method and device, equipment and a storage medium, and relates to the technical field of industrial manufacturing. In some embodiments of the invention, the method comprises the following steps: obtaining an equipment operation state parameter, a process operation parameter, a global operation parameter and an external environment parameter; inputting the equipment operation state parameters, the process operation parameters, the global operation parameters and the external environment parameters into a scheduling model, and solving through multi-objective optimization by taking the minimum comprehensive cost as an objective function to obtain high-collaboration operation regulation and control parameters; controlling factory equipment to execute regulation and control operation according to the operation regulation and control parameters; according to the invention, cross-level parameter linkage is realized, global optimum is taken as a regulation and control policy, the control efficiency of factory equipment control is improved, and the intelligent degree is improved.

The invention belongs to the technical field of thermal generator set automatic control, and particularly discloses a BEST double-machine regenerative thermal power plant converter load shedding automatic control method which comprises the steps that the shaft power of a steam feed pump and the shaft power of a small generator are determined; according to the sum, the pre-action opening degree of a BEST feed pump turbine steam inlet adjusting valve after the converter is subjected to load shedding is determined; predicting the rising rotating speed of the BEST steam feed pump set according to the rotating speed of the BEST steam feed pump set; and a steam feed pump minimum flow recirculation valve control method and a BEST feed pump turbine backpressure control method are started, and a rotating speed control method of the BEST feed pump turbine is determined according to the control method. According to the method, the rotating speed fluctuation process of the BEST steam feed pump set is shortened, the fluctuation amplitude of the boiler inlet feed water flow is reduced, and the stability and safety of the whole set after load shedding of the converter are guaranteed.

The invention relates to the technical field of offshore wind power grid connection, in particular to a low-frequency oscillation suppression method for an offshore wind power synchronous machine grid connection system, which comprises the following steps: acquiring inherent parameters of the synchronous machine grid connection system and active power reference instructions of each fan of an offshore wind power plant; according to the inherent parameters, control parameters of a preset damping controller are determined, and the damping controller is configured based on the control parameters; acquiring an actual operation frequency signal of the offshore low-frequency power grid in real time, and comparing the actual operation frequency signal with a preset rated frequency of the offshore low-frequency power grid to obtain a frequency deviation signal; inputting the frequency deviation signal into a configured damping controller, and outputting a damping power instruction; and superposing the damping power instruction and the active power reference instruction to generate an active power control instruction. According to the invention, the low-frequency oscillation of the new energy synchronous machine grid-connected system can be effectively suppressed, and meanwhile, the situation that the grid-connected safety is threatened due to the great fluctuation of the terminal voltage of the motor caused by suppression measures is avoided.

A computer-implemented method and system for performing one or more AI techniques on data captured from an Industrial Automation Control Systems (IACS) system, wherein the IACS is associated with at least one industrial plant. Data is captured from the IACS relating to each of a plurality of assets associated with the at least one industrial plant. The captured IACS data is analyzed to determine a plurality of assets associated with the at least one industrial plant contained in the captured IACS data, and metadata respectively associated with each determined asset. An electronic data repository is generated that provides an electronic inventory for each of the plurality of assets and associated metadata. An AI engine coupled to the electronic data repository, performs one or more AI techniques on the plurality of assets and the metadata respectively associated with each determined asset, to perform a certain task.

The invention discloses a multi-target optimization unit adaptive combustion and control real-time optimization method and system. The method comprises the following steps: acquiring unit operation and decision parameters; constructing a target function with net power; an online Gaussian mixture process regression and uncertainty quantization network (OH-GPR-UQN) is adopted to online identify a dynamic relationship between a decision parameter and a net power objective function, a gradient is estimated, and the uncertainty of the gradient is quantized; on the basis of the gradient and the uncertainty thereof, an adaptive exploration and utilization constraint hill climbing algorithm (BOA-CEHC) based on Bayesian optimization guidance is adopted, and the adjustment step length and direction disturbance of decision parameters are determined by maximizing an acquisition function which considers the utilization gradient and the exploration uncertainty at the same time; processing operation constraints by combining gradient projection and a penalty function to obtain optimized parameters; the optimized parameters are used for closed-loop control and operation data feedback so as to update the OH-GPR-UQN online. According to the method, accurate identification, intelligent optimization and robust processing constraint can be realized, and the comprehensive benefits of the unit are remarkably improved.