470 results about "Setpoint" patented technology

The invention discloses a multi-parameter MPC two-phase liquid cooling energy efficiency optimization distribution control system and method, and relates to the technical field of liquid cooling energy efficiency optimization. The multi-parameter MPC two-phase liquid cooling energy efficiency optimization distribution control system and method comprises the following steps: S1, collecting and preprocessing two-phase liquid cooling control data; s2, the thermal load level of each area is evaluated periodically, a discrete time linear state space model is constructed, and the thermal load change rate is extracted; s3, the cold distribution amount of each area is quantified periodically, the pump speed and the valve opening target value are calculated, and an adjustment control instruction is generated; and S4, high heat risk and low heat stability areas are recognized, the cooling capacity compensation requirement is evaluated, a cooling capacity enhancement and recovery distribution value is output, and a corresponding control instruction is generated. The problems that an existing two-phase liquid cooling system generally only depends on temperature feedback for adjustment, so that cooling capacity distribution response lags behind, a short-time overheating phenomenon is prone to occurring in a high-heat area, and cooling capacity waste exists in a low-heat area at the same time are solved.

The invention relates to the technical field of automatic driving, in particular to an automatic driving obstacle intention prediction and avoidance method and system. The method comprises the following steps: acquiring image data and three-dimensional point cloud data in real time, respectively extracting static features and dynamic features, weighting by adopting an attention mechanism, fusing to form a fused feature vector, inputting the fused feature vector into a plurality of independently trained deep learning models, and obtaining a prediction intention by using a model difference weighted fusion output result. The method comprises the following steps of: setting an association constraint of an operation state and input control based on a prediction intention, establishing an avoidance strategy model by taking path length and time cost as double optimization targets, continuously updating input control in a prediction time domain until an optimization target value is iterated to be minimum, outputting a time sequence of an avoidance strategy, and generating a final execution instruction in combination with a real-time operation state. According to the invention, the bottleneck of a traditional system based on static characteristics and preset rules can be broken through, and the road traffic efficiency is obviously optimized while the traffic safety redundancy is improved.

The invention provides a multivariable decoupling control method and system for an air inlet system of a high-altitude simulation cabin, and the method comprises the steps: firstly building a precombustion chamber cavity thermodynamic differential equation, and building a state equation in combination with a regulating valve first-order inertia model and an engine model; constructing a decoupling matrix by using a differential flatness theory, and converting a pressure and temperature coupling system into an independent control channel; monitoring a system state in real time by using an extended state observer, uniformly estimating internal and external disturbances as total disturbances, and generating a compensation signal; the model prediction controller performs rolling optimization on future time domain control quantity according to deviation between a set value and an actual value, generates an initial instruction in combination with the compensation signal, and converts the initial instruction into an independent valve control signal through a decoupling matrix; the high / low-temperature airflow mass flow is controlled through the opening degree of the adjusting valve, the engine model outputs actual parameters to form closed-loop feedback, and the controller updates the control quantity in each cycle. The multi-variable strong coupling, the dynamic disturbance and the model uncertainty are effectively overcome, the control precision and the parameter robustness are both achieved, and the complex working condition requirements are met.

The invention discloses a combined heat and power generation system control strategy based on an AME-TD3 algorithm, and belongs to the technical field of combined heat and power generation system optimization control. Comprising the following steps: S1, collecting system parameters and initializing a CHP running state; s2, when a target # imgabs0 # value is calculated, an entropy reward mechanism is introduced; s3, realizing exploration and utilization of an adaptive noise adjustment mechanism and a dynamic balance strategy based on Critic network evaluation; and S4, optimizing the control strategy in real time. Aiming at the problem that an existing TD3 algorithm is insufficient in exploratory performance, an entropy correction item is introduced into calculation of a target # imgabs1 # value, structural correction is conducted on a target value function, strategy updating is more efficient, and the effectiveness of overall strategy optimization is improved; aiming at the problem that random noise in a traditional TD3 algorithm cannot adapt to a complex environment, a dynamic self-adjusting noise generation function is added in a strategy updating process, so that the convergence speed is increased, the stability is enhanced, and the self-adaptive adjustment capability of a system to load fluctuation and environmental condition change is enhanced. The method has good expandability.

A method for automated guidance of a vehicle along a specified setpoint trajectory at an actual speed, wherein the setpoint trajectory includes geometric trajectory variables, the method including ascertaining an actual deviation of the vehicle from the setpoint trajectory and outputting the ascertained actual deviation and generating manipulated variables such that the vehicle, in the case of automated control of a drive system and / or a braking system and / or a steering system of the vehicle, moves closer to the setpoint trajectory. The method also includes ascertaining whether an undesirable driving state is present when the vehicle moves closer to the setpoint trajectory, wherein the undesirable driving state is ascertained from the specified setpoint trajectory based on the geometric trajectory variable, and, if the presence of the undesirable driving state is identified, automatically controlling the vehicle based on generated stability variables and / or adapting the setpoint trajectory.

The invention relates to the technical field of environment regulation and control, and discloses a multi-parameter environment closed-loop regulation and control method and system based on an FPGA, and the method comprises the steps: unifying monitoring data to a same space coordinate system through the FPGA according to the monitoring data collected by a sensor network disposed in an environment, analyzing the coupling relation between different parameter data, and obtaining fusion data; based on the fusion data, analyzing the environment crop state through a preset crop state recognition model, simulating the influence condition of different parameter adjustment processes on the crop state, and generating a first control instruction set; analyzing the degree of deviation between the environmental parameters in the fused data and a preset target value, calculating the priority of each control instruction in the first control instruction set, and merging and optimizing the control instructions; the environmental parameters are corrected; according to the application, the regulation and control efficiency can be improved, the environmental disturbance and the crop demand change can be quickly responded through FPGA parallel processing, and the environmental regulation and control efficiency and the regulation and control effect are improved.

The invention discloses a serial cold rolling system thickness and tension control method based on data driving, and relates to the technical field of intelligent predictive control. According to the method, the neural network model is driven by non-autonomous data for prediction, dynamic mapping of tension and thickness parameters is realized through a self-learning mechanism, and high-precision prediction can be completed without depending on real-time measurement data. According to the method, the problems of hysteresis and error accumulation of a traditional measurement link are effectively solved, and deviation transfer in multi-rack cooperative control is effectively inhibited while the prediction efficiency is remarkably improved. A Laguerre function is integrated, parameter dimensionality reduction of a network weight space is achieved through orthogonal projection, the calculation amount is reduced, and the time for reaching a target value is shortened. A multivariable coupling mathematical model of the tension and the thickness of the cold-rolled strip steel is constructed, a Nash equilibrium game optimization strategy is introduced based on a distributed optimal control theory framework, and multi-target collaborative optimization control of the tension-thickness is realized.

The invention discloses an AGV navigation system used in a storage dynamic environment, relates to the technical field of storage robot control, and is used for solving the problem of poor stable navigation in dynamic storage. According to the method, a closed-loop navigation link of state convergence, confidence evaluation, corridors, setting generation and cooperative gating is constructed, state frames are packaged in unified time, and positioning confidence and availability marks are generated based on the consistency of pose predictive quantity and pose observed quantity, so that credible hierarchical control is realized in shielding, light reflection and personnel forklift passing scenes, and the positioning accuracy is improved. A time sequence occupation constraint set is formed through obstacle tracking, a forbidden area mark is fused, a passable corridor boundary is generated, a speed set value sequence and a curvature set value sequence are generated in a corridor, acceleration change rate constraint and steering change rate constraint are applied, and stable driving control and steering control instructions are output. And the way-giving sequence and the recovery sequence are matched, and the execution state and the occupation state are written back, so that the navigation safety, the stability and the operation continuity are improved.

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 machine learning, and discloses a learning path generation method based on reinforcement learning, and the method comprises the steps: obtaining the state deviation data of a target object relative to a target value, and constructing an evolution trajectory; extracting a first-order derivative and a second-order gradient of the state deviation to determine a response phase acceleration vector; when the acceleration vector exceeds a threshold value, judging a response phase sudden change and locking a time point; triggering asymmetric sampling space reconstruction of the empirical pool, and executing sampling probability attenuation on historical samples before the time point by using an exponential attenuation operator; according to the method, the logic viscosity of the strategy network to the old response phase is forcibly relieved by using an asymmetric sampling mechanism, the strategy updating lag generated by high-reward historical experience is eliminated, and the strategy remodeling efficiency of the system under the working condition of phase migration is improved.

The invention discloses an AC excitation system self-starting control method and system, and the method comprises the steps: obtaining a rotor current rotation frequency and an actual rotor rotation frequency, and generating a slip frequency; the slip frequency is compared with a preset dynamic threshold value, and the rotating speed given value is dynamically adjusted; inquiring a preset multi-stage sectional type current control curve according to the actual rotor rotation frequency, and outputting a d-axis current given value; pI adjustment is carried out based on the slip frequency to generate a q-axis current given value, and corresponding preset current threshold limit is applied according to the frequency band where the actual rotor rotation frequency is located; performing independent parameter configuration on the magnetic flux current control loop and the torque current control loop by adopting a non-coupling parameter setting algorithm; and in the unit starting process, under the condition that the stable starting constraint condition is met, an optimal starting parameter combination is generated. According to the invention, by dynamically adjusting the rotating speed given value and controlling the current in a segmented manner, the stable self-starting of the alternating current excitation system is realized, and the reliability and efficiency of the starting process are improved.

The invention provides an excitation amplitude automatic closed-loop control system and method for a DLCP test. The system comprises a multi-frequency alternating current excitation and direct current bias voltage applying module which is used for automatically applying alternating current excitation signals and direct current bias voltage signals according to a preset excitation amplitude sequence under a plurality of test frequencies; the PID closed-loop control module is used for performing closed-loop correction on the applied voltage according to the response of the test port of the tested semiconductor device, so that the condition of the applied voltage at the test port meets a preset requirement; the capacitance measurement module is used for acquiring an equivalent capacitance response signal of the tested semiconductor device; the capacitance test data caching module is used for performing associated marking and classified caching on the capacitance test data according to the test frequency, the excitation level and the bias voltage set point; and the main control module is used for coordinating automatic test processes under the conditions of a plurality of test frequencies and different excitation amplitudes and outputting corresponding capacitance test data so as to improve the consistency, stability and automation degree of the DLCP test.

The invention provides a regulation and control method, system and device for a relay valve overrun test and a medium, and belongs to the technical field of air braking systems. The method comprises the steps that pressure response characteristics of a relay valve pipeline system under different opening degrees of a proportional pressure regulating valve are tested, a finite set representing pipeline dynamic response is constructed based on pressure data in a sampling period, and a dynamic matrix model is generated to describe the relation between future pressure output and control quantity; pipeline pressure output at the future moment is predicted based on the dynamic matrix model, an optimized target function is established in combination with the target boosting rate, and the optimal control quantity is solved to adjust the opening degree of the proportional pressure regulating valve in real time; comparing a predicted pressure value with an actual measurement value to obtain an error, dynamically correcting a predicted model parameter by using a weighted error sequence, and adjusting a subsequent control quantity to compensate system hysteresis and external interference; and circularly executing the processes until the pipeline pressure and the regulation and control time reach corresponding target values, and cutting off the gas source.

The invention relates to a shaft ECD intelligent regulation and control method and system based on machine learning and closed-loop control, and the method comprises the steps: obtaining original data of drilling parameters, and constructing an ECD prediction model based on the original data; acquiring real-time input data; inputting the real-time input data into the ECD prediction model for prediction to obtain an ECD prediction value; calculating according to the ECD predicted value and the target ECD value to obtain an ECD error value; and selecting a model intelligent regulation and control mode or a PID intelligent regulation and control mode according to the ECD error value to carry out ECD regulation and control. According to the method, the shaft ECD is predicted through the machine learning model, and the drilling parameters are adjusted in real time through the PID controller, so that the ECD is kept within the preset range. The double-layer control system combines the real-time feedback advantage of PID and the predictive capacity of machine learning, it is ensured that variables are accurately regulated and controlled in the drilling process, the ECD is maintained within the safety range, the risks such as stratum fracture and blowout are reduced, and the drilling efficiency and safety are improved.

The invention relates to the technical field of digital twinning, and discloses a process industry digital twinning platform system, which comprises a process mechanism modeling module used for constructing a multi-scale first principle model system according to physical and chemical laws of the process industry; the dynamic coupling engine is used for collecting process data in real time and carrying out coupling solution by combining the first principle model system and the equipment operation state model; the closed-loop optimization control interface is used for carrying out two-way communication with the distributed control system, generating an optimization set value based on a solving result and executing control; and the visual diagnosis module is used for carrying out multi-dimensional analysis, fault root cause tracing and early warning response on process parameter deviation. First principle equations of reaction kinetics, heat transfer theory and the like are directly embedded into the virtual model through the process mechanism modeling module, so that the model has physical and chemical interpretation capability on complex chemical reactions and energy transfer, and high precision is still kept under complex working conditions.

The invention discloses an internet-of-things central air conditioner control method based on a joint regulation and control algorithm model, which comprises the following steps: S1, periodically acquiring temperature, area and use purpose parameters through an internet-of-things temperature controller mounted in a tail end room, and uploading the parameters to an edge computing node; s2, collecting water temperature, water pressure and flow data in real time through an Internet of Things sensor arranged on a pipeline, and uploading the data to an edge computing node; s3, the edge calculation node calculates a load coefficient alpha and a water flow coefficient beta based on the collected information; s4, the control system takes the load coefficient alpha and the water flow coefficient beta as input, and determines a host outlet water temperature set value and a water pump frequency based on a host-water pump combined regulation and control algorithm model; and S5, the host and the water pump are controlled based on the host outlet water temperature set value and the water pump frequency, and dynamic regulation and control between cold and heat supply and energy consumption are achieved. According to the method, dynamic matching of cooling and heating loads and energy-saving optimized operation of the system are realized through accurate acquisition of terminal data and multi-dimensional load modeling.

The invention discloses a motor system energy-saving optimization automatic control system based on industrial Internet of Things, which relates to the technical field of automatic control, and comprises a data acquisition module, a load decoupling module, an aerobic prediction module, a biochemical load disturbance index calculation module and a control module, the system comprises an operating parameter prediction module, a disturbance index prediction module, a decision optimization module, a model correction module and an oxygen demand prediction module, wherein the operating parameter and the disturbance index are combined to predict an optimal dissolved oxygen set value; the decision optimization module generates an optimal motor rotating speed control instruction according to the disturbance index and the set value; and the model correction module calculates a deviation factor according to the set value and real-time concentration and performs online correction when a threshold value is exceeded. Through cooperation of multiple modules, accurate, energy-saving and self-adaptive regulation and control of the aeration process are realized, loads can be accurately distinguished, the treatment effect and energy consumption can be balanced, long-term stable operation is guaranteed, and the energy consumption of the system is remarkably reduced.

The invention discloses a cross-industry multi-source fault intelligent research and judgment method and system based on a large model, and relates to the technical field of large models, and the method comprises the steps: calling a historical fault record of a power transmission line, deploying a sensor on the power transmission line, and obtaining target data corresponding to each fault position in each physical type; obtaining an associated physical type and an associated function corresponding to each fault type; extracting all detection points on the target power transmission line to obtain first target values of the detection points; extracting a management and control area where personnel are limited to enter in the power transmission line, and obtaining a second target value of each detection point; setting a weight to obtain a total target value of each detection point; and obtaining a first judgment criterion and a second judgment criterion, and judging whether to carry out early warning prompt on the selected sampling point or not. According to the method, the fault record and the control area are analyzed, so that the current sampling point is more reasonable and representative, the power transmission line fault is intelligently researched and judged, and the reliable judgment on the power transmission line is effectively improved.

The invention relates to the technical field of automatic control theory application and data center thermal management, and discloses a data center emergency refrigeration control method based on state space model prediction, which comprises the following steps: establishing a thermodynamic cycle state space equation; the refrigeration flow set value change rate and the water return side pressure change amount are obtained, and the dynamic heat exchange sensitivity is determined; monitoring feedback current characteristics of the driving unit, inverting cooling path resistance distribution and updating gain matrix elements; calculating a state prediction information sequence according to the temperature vector prediction deviation, and adjusting a time domain step length; the dynamic heat exchange sensitivity is introduced into prediction logic, controlled variable dynamic pressure retardation is compensated, an emergency refrigeration frequency control instruction is output, fluid fluctuation is converted into sensitivity compensation of a logic level, non-linear dynamic pressure damping of the emergency working condition is effectively impacted, the problems of control overshoot and logic instability caused by physical lag are solved, and the reliability of the system is improved. And the refrigeration response precision at the emergency conversion moment is improved.

A computing system for functional integration with a distributed power grid having one or more inverters. The computing system includes processing device(s) to control operation of an inverter by executing a feedback controller with a grid-tied line, of the distributed power grid, treated as a plant and including a unified algebraic control system operating based on a pair of closed-loop variables carried in closed-loop signals of the unified algebraic control system. The processing device(s) control, using the feedback controller, transitions of the inverter between a plurality of operating modes by adjusting a magnitude of the closed-loop signals, which correspond to the pair of closed-loop variables, towards a setpoint of a plurality of setpoints. Each setpoint corresponds to a different operating mode of the plurality of operating modes.

A method for approximating a friction value includes: determining a load characteristic; determining a setpoint variable of the vehicle; determining a manipulated variable expected value specifying a predicted value of a manipulated variable to be provided to set the setpoint variable, wherein the determination of the manipulated variable expected value is performed using the load characteristic; determining an actual variable corresponding to the setpoint variable; determining a manipulated variable actual value, which is provided at the steering system, in order to modulate the actual variable; determining a manipulated variable deviation between the manipulated variable expected value and the manipulated variable actual value; and / or determining a setpoint-actual deviation between the setpoint variable and the corresponding actual variable; approximating the friction value based on the determined load characteristic and based on the determined manipulated variable deviation and / or the determined setpoint-actual deviation. A driver assistance system is configured to carry out the method.

A feedforward-feedback controller for integrated temperature and pressure control of a building space includes one or more processing circuits. The one or more processing circuits are configured to generate a supply air flow rate setpoint using a combined feedforward-feedback control process that (i) proactively compensates for a feedforward air flow disturbance in the building space and (ii) reactively compensates for a feedback air pressure error in the building space, generate a supply air temperature setpoint using a predictive temperature model that predicts the supply air temperature setpoint required to achieve a zone temperature setpoint for the building space when supply air is provided to the building space at the supply air flow rate setpoint and the supply air temperature setpoint, and operate building equipment to provide the supply air to the building space at the supply air flow rate setpoint and the supply air temperature setpoint.

An intelligent valve self-adaptive sealing method based on working condition perception belongs to the technical field of industrial process control and fluid transportation, and comprises the following steps: step 1, multi-dimensional working condition perception and data preprocessing; 2, modeling based on a sealing requirement of a physical model and data driving; 3, optimizing a sealing force set point based on an intelligent decision-making strategy; fourthly, closed-loop control and execution are carried out; and 5, carrying out sealing state health diagnosis and early warning. According to the intelligent valve self-adaptive sealing method achieved through the intelligent valve self-adaptive sealing system, a working condition sensing and preprocessing module, a sealing requirement calculation module, a sealing setting optimization module, an intelligent control and precision execution module and a state diagnosis and early warning module are included. According to the method, the sealing force set point can be adjusted in advance according to the predicted working condition change, and dynamic and real-time self-adaptive setting, prediction and closed-loop optimization of the valve sealing force are achieved; and meanwhile, real-time transparency and diagnosis early warning of the sealing state of the valve are achieved.

The invention relates to the technical field of flow metering, in particular to a temperature return control method of an intelligent control valve. The method comprises the following steps: receiving a temperature return control strategy sent by a user and a set value of specified temperature return, and judging an initial control parameter set according to the temperature return control strategy; real-time operation data are collected from the temperature sensing module and the flowmeter according to the sampling period, the return water temperature and flow at the current moment are obtained, and the return temperature deviation is calculated; calculating a control target value of the intelligent control valve by utilizing a model fuzzy logic predictive control algorithm based on the temperature return deviation; and the valve opening degree is adjusted based on the control target value, and the control parameters are dynamically corrected through the temperature return control strategy for self-adaptive temperature return control. According to the method, through a composite algorithm of fusing fuzzy logic control and model predictive control (MPC), on the basis that preliminary control signals are generated through fuzzy reasoning, a nonlinear system model is combined to carry out multi-step prediction on the future return water temperature change trend.

The invention provides a mill control loop set value optimization method based on empirical-driven adaptive enhancement decision. The method comprises the following steps: identifying operation indexes in the ore grinding process of a mill; constructing a performance index function based on the identified ore grinding granularity and the mill load; constructing a causal diagram of historical operation data in the ore grinding process of the mill; calculating the similarity between the causal diagram of the current working condition and each causal diagram of historical operation data; an Actor-Critic model is built, and the Actor-Critic model is trained; setting a similarity threshold value eta between the causal diagram of the current working condition and all causal diagrams in the historical operation data; based on the calculated similarity and a similarity threshold eta, screening a pre-trained Actor-Critic model which is most matched with the current working condition causal graph, and updating a strategy parameter theta k of the pre-trained Actor-Critic model in real time; and transmitting the set value generated in real time to a controller of a mill control loop. The problem of control lag caused by sudden change of working conditions in a traditional method is solved, dynamic and accurate regulation and control of the set value in the ore grinding process are achieved, and the production efficiency of the ore grinding process is improved.

The invention relates to the technical field of industrial automation and energy management, in particular to a smoke waste heat utilization monitoring system based on big data. Comprising a multi-dimensional time sequence data acquisition and fusion module used for generating a unified data set after cleaning and time alignment processing; the heat source-load prediction module based on time sequence deep learning is used for generating a predicted supply and demand curve in a future scheduling period; the multi-target dynamic optimization and scheduling decision module is used for generating an optimal heat energy distribution scheduling scheme; and the real-time feedback and self-adaptive control module takes the optimal heat energy distribution scheduling scheme as a dynamic set value of the control system, and returns the actual distribution power measured and calculated in real time to the multi-dimensional time sequence data acquisition and fusion module for periodically retraining the attention mechanism long and short term memory network model. According to the system, the precision of supply and demand matching is remarkably improved, and the problem of supply and demand mismatching caused by untimely response in a traditional mode is effectively solved.

The invention discloses an embedded hardware sensing dynamic feedback and data acquisition control method, and particularly relates to the field of general control or regulation systems, which comprises the following steps of: establishing a complete sensing acquisition framework by initializing a hardware control parameter set, synchronously acquiring original physical data and environmental parameters of a sensor, and preprocessing the original physical data and environmental parameters into available measurement values; comprehensively evaluating the deviation between actual output and a target value based on a quantitative difference model, and realizing dynamic correction and real-time feedback optimization of software level control parameters through a gradient optimization algorithm; when the parameter optimization effect does not reach the expectation, a hardware-level calibration process is automatically triggered, and temperature drift, power supply fluctuation and other physical characteristic drifts are compensated; a closed-loop control link is formed through multiple rounds of acquisition verification, it is ensured that the data precision is stable and reaches the standard, the sensing precision stability and control reliability of an embedded system in a complex environment are remarkably improved, and the method is suitable for high-precision embedded sensing application scenes.

A method for controlling (S4) a drive motor (4) of a human-powered vehicle (2) is described. The method comprises determining (S1) a nominal setpoint value of a torque of a shaft (6) connected to the drive motor (4) and determining (S2) at least one adaptable parameter. The method comprises determining (S3) a threshold value for an amount of the setpoint value of the torque as a function of a specific temperature of an element (8) of the vehicle (2) and as a function of the specific parameter. The method further comprises determining (S4) a setpoint value of the torque of the shaft (6), wherein the specific nominal setpoint value is determined as the setpoint value if an amount of the nominal setpoint value is smaller than the threshold value for an amount of the setpoint value, and otherwise the setpoint value is determined such that an amount of the setpoint value corresponds to the specific threshold value.The method further comprises controlling (S5) the drive motor (4) as a function of the determined torque target value. A control device (10) configured to carry out such a method is also described. Furthermore, a drive train (12) having such a control device (10) and a muscle-powered vehicle (2) having such a drive train (12) are described.

The invention discloses a thermal power generating unit AGC performance online detection and optimization method. A programmable logic controller is arranged outside a distributed control system of each thermal power generating unit; the distributed control system transmits an AGC instruction target value, an AGC input signal, a unit variable load rate set value and a unit load of the thermal power unit to the programmable logic controller in real time; and the programmable logic controller calculates the real-time variable load rate of the unit, transmits the real-time variable load rate of the unit to the decentralized control system to correct the set value of the variable load rate of the unit, obtains key data points and AGC real-time performance indexes, and transmits the key data points and the AGC real-time performance indexes to the decentralized control system. According to the invention, the real-time variable load rate of the unit and the set variable load rate of the unit are calculated on line to obtain the variable load rate correction coefficient, the variable load rate of the unit can be corrected on line in real time, the adjustment performance of the unit is optimal, and power plant engineers can accurately master the AGC performance of a single unit.

In variants, an ML-based control system for an industrial facility can include a remote control plane, a client, a set of agents, data processing modules, an optional data store, a telemetry module, a learning module, and / or any other suitable components. In variants, the set of agents can include a facility agent, a set of local cooling agents, and / or any other suitable agents. In variants, the set of agents can include system state approximators and decision models learned via reinforcement learning. The ML-based control system can control components of the industrial facility by predicting and / or determining component setpoints based on power load and / or heat load of the industrial facility.