179 results about "Active control" patented technology

Systems, methods, apparatuses, and computer program products for enhancing dynamic CORESET adaptation with limited or no control channel overhead. One method may include receiving, by a UE, a plurality of PDCCH CORESET configurations associated with at least one SS; determining, by the UE, at least one of the PDCCH CORESET configurations is associated with a symbol or slot format from a corresponding set of symbols; and performing, by the UE, PDCCH monitoring on an active control resource set configuration associated with the set of determined slots or symbols.

The invention discloses an intelligent network connection automobile active safety teaching control method based on digital twinning. The method comprises the steps that S1, multi-source sensor data are collected and preprocessed to generate a driving state data set; s2, inputting the state data into a digital twin model to drive a traffic scene and outputting a synchronous state; s3, based on Dueling-DDQN, executing strategy learning to generate an active safety control instruction; s4, the instruction response effect is simulated and verified in the virtual environment; s5, collecting driver operation behaviors, inputting the improved MHA-BiLSTM model to extract time sequence features, and outputting behavior features; and S6, comparing the driving behavior with a standard instruction item by item, calculating an operation deviation and a response difference, and generating a personalized active safety teaching task. According to the invention, efficient comparison and teaching feedback of the driving behavior and the active control strategy can be realized, and the intelligent level of driving training is improved.

The invention relates to the technical field of underground engineering construction monitoring and control, and discloses a multi-source monitoring data-based settlement risk early warning system for shield wall grinding undercrossing of a station, and the system comprises a data collection module which obtains shield and structure response data; the preprocessing module executes space-time correction to generate a synchronous data stream; the spectrum energy analysis module solves a mechanical-structural impedance coupling index; the rigidity inversion module uses the index to correct the specific energy and invert the residual thickness of the wall body; the active control module updates the kinetic model based on the thickness to calculate a frequency detuning factor; the decision module constructs a multi-dimensional feature vector, calculates a dynamic time warping distance, generates a control instruction according to a detuning factor, and achieves the active closed-loop control of the wall grinding risk. According to the method, a spectrum energy coupling analysis and dynamic stiffness inversion mechanism is introduced, and the apparent mechanical specific energy is corrected by using the mechanical-structural impedance coupling index, so that the accurate identification of the real physical state of the cutting medium and the real-time inversion of the wall residual thickness are realized.

The invention relates to a self-adaptive drilling method and system based on hole wall stability and deep learning, and belongs to the technical field of coal mine underground drilling. Aiming at the technical problems that the drilling efficiency is low and the construction safety risk is high due to the fact that an existing drilling control method is difficult to effectively predict and avoid hole wall instability, the method adopts the core technical scheme that a drilling parameter sequence is collected in real time through an orifice sensor; calculating a hole wall instability risk probability in real time by using a pre-trained lightweight hole wall instability prediction network; further, the scheme evaluation network predicts future multi-target performance based on the current state and the candidate control set; and finally, constructing a scoring function by taking the hole wall stability as a primary constraint, preferentially generating a control instruction and issuing the control instruction to a drilling machine for execution. According to the method, intelligent prediction and active control over the stability of the hole wall are achieved, the safety and reliability of drilling construction are remarkably improved, meanwhile, the drilling efficiency and the hole forming quality are both considered, the system calculation load is low, and on-site deployment and application are easy.

The invention discloses an anti-overturning active control method and device for a single-column pier bridge and a medium. The method comprises the steps that a thundersight displacement monitoring system and a hydraulic driving system are deployed; leighting multi-mode fusion data are obtained, and the transverse overturning angle of a beam body is calculated; constructing a digital twinborn body for predicting the overturning state of the bridge structure, and predicting the overturning states of the bridge at multiple moments in the future; and constructing an intelligent agent, and controlling the hydraulic driving system to execute an anti-overturning action based on the predicted bridge overturning state. The device and the medium are both used for implementing the method. The beneficial effects of the invention are that the structure displacement monitoring method based on Reighting heterogeneous fusion is provided for a traditional mode of a single structure displacement detection means, interference factor identification and weighted fusion in different environments are realized, and the reliability of monitored structure displacement data is ensured.

The invention discloses a multi-axis linkage machining precision optimization control method adopting real-time detection driving, and relates to the technical field of machining intelligent control. The method comprises the following steps: establishing a multi-source real-time detection system comprising a laser displacement sensor, a force sensor and an industrial camera, and collecting multi-source data; performing time-space synchronization calibration on multi-source data, and improving a fusion vector of a feature by borrowing an attention mechanism convolutional neural network; constructing a machining precision dynamic compensation model by combining the kinematics forward and inverse solution sub-model and the tool thermal deformation compensation sub-model, and iteratively optimizing parameters by adopting an improved particle swarm optimization algorithm; the compensation amount is converted into a numerical control instruction through a data conversion module; and after compensation, data feedback updates the feature vector, and compensation parameters are dynamically adjusted until machining is finished. According to the invention, an active control system of multi-source detection, intelligent analysis, dynamic compensation and dynamic feedback is constructed, precise control of the whole processing flow is realized, and a general solution is provided for precision upgrading in the field of high-end equipment manufacturing.

The invention discloses a fault diagnosis and self-healing control method, system, equipment and medium for a flexible interconnection device of a power distribution network, and belongs to the technical field of fault diagnosis and control of the power distribution network, and the method comprises the steps: building a system state vector, collecting operation parameters of the power distribution network, and carrying out the weighted fusion; performing quality inspection on the original data based on the acquired operation parameters, and dynamically adjusting model parameters through digital twin synchronization and model parameter updating; deep learning is carried out to carry out multi-level fault feature extraction, and multi-scale feature fusion is carried out; a fault mode identification module is started, and fault probability is calculated for fault classification; evaluating the severity of the fault according to a fault diagnosis result, selecting an optimal control strategy, and carrying out self-healing control; and fault early warning is started to estimate a future fault occurrence probability, and early warning information of different levels is issued to perform preventive maintenance. According to the invention, full-process automation and intelligentization from data perception and intelligent diagnosis to active control and prospective maintenance are realized.

The invention discloses an intelligent shock absorption and noise reduction control method for rail transit, and belongs to the technical field of rail transit noise and vibration control. The method aims at solving the problems that traditional passive control measures are poor in adaptive capacity and limited in frequency band, and effective prospective offset is difficult to achieve due to system delay and inaccurate reference signals in existing active control. The method comprises the following steps: synchronously acquiring vibration and noise signals through a sensor array; calculating the contribution weight of vibration to noise by using a cross-domain modal correlation module based on an attention mechanism, and fusing the features; predicting an evolution sequence of vibration and noise in a specific time period in the future by adopting a long-short-term memory network model; synthesizing a target reference signal according to the prediction sequence, and generating an anti-phase control signal through an adaptive filtering algorithm; finally, it is ensured that the total delay of the system is smaller than the feedforward advance time, and an actuator is driven to achieve accurate foresight counteracting. The method is mainly used for improving the self-adaptive capacity and the control effect of the rail transit damping and noise reduction system.

The invention provides a direct disturbance suppression method for a vibration isolation system based on driving force feedforward and linear active disturbance rejection control, and belongs to the technical field of active vibration isolation. The problems that existing active control methods such as feed-forward control and iterative learning control are limited by model dependence, nonlinear adaptation is poor, and aperiodic disturbance processing capacity is weak are solved. The method comprises the steps that a dynamic model of a vibration isolation system is established, the vibration isolation system comprises a vibration isolation platform and a motion platform, and the dynamic model is used for quantifying direct disturbance force caused by motion of the motion platform; a feedforward compensator is designed, and feedforward compensation force is generated in real time according to a motion instruction of the motion platform; introducing a linear active disturbance rejection controller; and superposing the feed-forward compensation force with a compensation force generated by the linear active disturbance rejection controller to obtain a total compensation force, and applying the total compensation force to the vibration isolation system to realize disturbance suppression. The method is mainly used in the field of precision manufacturing and measurement.

The invention relates to the technical field of glass production process intelligent control, in particular to a glass production process intelligent control system based on a PLC. The system comprises a data acquisition module, a first processing module, a second processing module, a damage evaluation module, a state determination module and a parameter adjustment module. The system predicts a temperature field in the kiln and calculates thermal stress by collecting PLC and sensor data and using a reduced order model; the core is to track the historical fluctuation of thermal stress and determine the accumulated fatigue damage degree based on the fatigue accumulation theory; converting the temperature into the kiln health degree to determine a health perception regulation factor and a control mode; when the health degree is reduced, the system adaptively adjusts the control parameters of the PLC, so that the control response tends to be gentle and conservative; according to the invention, health perception adjustment of the control strategy is realized, and excessive active control behaviors causing physical damage can be actively avoided.

The application discloses a kind of arbitrary pulsating wind field active control wind tunnel physical simulation method and system, comprising: establishing generalized pulsating wind field spectrum model, it has 6 controllable pulsating wind spectrum in the energy distribution model parameter of frequency domain;Based on generalized pulsating wind field spectrum model, inter-spectrum matrix is calculated using coherence function;Based on harmonic superposition method, inter-spectrum matrix is converted into pulsating wind field space-time information;Pulsating wind field space-time information is introduced into the control program of active control wind tunnel, through real-time monitoring and iterative adjustment model parameter, until the target wind field generated reaches expectation.The application can simulate wind field with arbitrary frequency energy distribution by generalized pulsating wind field spectrum model with 6 model parameters, combined with iterative closed-loop control, significantly improve the simulation accuracy and adaptability of complex wind field such as mountain valley wind, typhoon.

The invention provides an energy storage type double-fed phase modifier transient synchronous supporting method and device and a medium, and relates to the technical field of electric power. The method comprises the following steps: when voltage drop and frequency disturbance occur in a power system, determining an adaptive reactive current proportionality coefficient according to a voltage deviation normalized value, a frequency deviation normalized value and a frequency change rate normalized value of a machine end of the energy storage type doubly-fed phase modifier by adopting a control strategy of reactive power priority and active power amplitude limiting by the energy storage type doubly-fed phase modifier; determining a target reactive current according to the adaptive reactive current proportionality coefficient so as to carry out reactive control on the energy storage type double-fed phase modifier; and in combination with the amplitude limiting effect of the virtual inertia control and the adaptive reactive current proportionality coefficient on the active current, obtaining a target active current so as to carry out active control on the energy storage type double-fed phase modifier. According to the invention, when voltage drop and frequency disturbance occur in a power system, the active and reactive output proportion can be dynamically adjusted, and the overall stability of voltage and frequency is improved.

This application discloses a method, device, medium, and product for active control of aerodynamic noise in the target area during tiltrotor transition flight, relating to the field of noise control. The method includes: during tilt transition conditions, constructing a target noise control area within a set region based on flight mission and environmental requirements, discretizing it into multiple spatial observation points and assigning weights; parameterizing the controllable aerodynamic loads on the blades to obtain control load variables, establishing the relationship between the control load variables and the acoustic response of the target noise control area, determining unknowns, and establishing a global acoustic optimization objective function; using the DDQN algorithm as a solver, optimizing the global acoustic optimization objective function under the premise of satisfying actuator physical constraints to obtain the optimal control law, performing feedforward execution and feedback correction to achieve active control of aerodynamic noise in the target area. This application can solve the main shortcomings and derived problems caused by focusing on single-point noise control.

This invention provides a machine learning-based method for active shape control in laser additive manufacturing. The method includes: simulating temperature field distribution data and microstructure images perpendicular to the printing direction during the additive manufacturing process using a physical model, and preprocessing these to obtain the mechanical properties of the formed part corresponding to the microstructure images; predicting the forming quality at the next moment using a trained neural network; and when the predicted forming quality does not meet requirements, actively adjusting printing process parameters to achieve real-time active control of the macroscopic temperature field, thereby regulating the microstructure and mechanical properties of the formed part online. Compared to calculation methods that directly construct physical constitutive models, this method can significantly shorten the solution time. Furthermore, this invention establishes a temporal and spatial relationship between printing process parameters, temperature field distribution, microstructure images, and mechanical properties, which is beneficial for promoting comprehensive technological innovation in the field of additive manufacturing.

The invention discloses a dynamic evaluation and intelligent compensation system and method for contact rigidity of a machine tool joint part. The system comprises a digital twin and multi-source data interface module, a contact interface multi-scale morphology acquisition and characterization module, a joint part contact mechanical state online simulation module, a contact rigidity dynamic calculation model library, a performance evaluation and intelligent compensation decision-making unit and a compensation strategy execution and feedback module. According to the method, dynamic and high-precision contact stiffness analysis is achieved, the problem that a traditional static model is insufficient in precision is solved by deeply fusing microstructure measured data, a macrostructure model and a dynamic working condition load, and the time-varying characteristic of the joint part under the complex working condition can be reflected more truly. A perception-evaluation-compensation active control closed loop is constructed, and the digitization level of machine tool design and operation is improved. According to the method, the contact state of the joint part is actively optimized, so that the dynamic and static performance and machining quality of the machine tool are improved.

The application discloses a wind power plant control method and device considering rotor speed dynamic constraint, and the method comprises the following steps: obtaining the operation parameters of synchronous units in a power grid; constructing an additional active control model of the wind power plant containing power optimization control; linearizing the power grid frequency and the synchronous unit output power according to the operation parameters and the additional active control model combined with the piecewise function principle to obtain a total power increment piecewise linear expression of the wind power plant; performing piecewise fitting according to the wind power coefficient and the rotor speed to obtain a first equation; obtaining a rotor speed piecewise expression according to the wind power plant rotor motion equation, the first equation and the total power increment piecewise linear expression; adjusting the control parameters of the wind power plant in each speed section according to the rotor speed piecewise expression; the control parameters at least include primary frequency modulation parameters; and controlling the wind power plant according to the control parameters.

The invention relates to the technical field of high and low temperature all-in-one machines, in particular to a high and low temperature all-in-one machine safety control method and system. Firstly, operation parameters of the high and low temperature all-in-one machine are collected in real time through a multi-mode sensor system; performing time synchronization and data fusion on the acquired operation parameters to generate a multi-dimensional time sequence fusion data sequence; calculating a real-time parameter boundary of the dynamic security domain model based on a current set working condition; inputting the multi-dimensional time sequence fusion data sequence and the real-time parameter boundary into a risk assessment engine, and calculating a comprehensive risk score; and finally executing a hierarchical active control strategy according to the comprehensive risk score. According to the method, a complete technical closed loop from multi-source data fusion, dynamic risk assessment to intelligent hierarchical control is realized, and the safety and reliability of equipment are remarkably improved.

An orientation fault detection system is configured to detect improper installation of the dynamic active control system (DACS) controller / software module by receiving data output from a plurality of sensors to determine whether an orientation fault is detected corresponding to an improper installation of the controller / software module. Such determination is made based upon sensor data when corresponding environmental and operational conditions are met, so the orientation fault detection system is further configured to determine whether the requisite conditions are met for orientation fault detection based upon the sensor data. If an orientation fault is detected, the orientation fault detection system initiates a safe state protocol that renders the vessel safe to operate.

The application belongs to the technical field of power electronic equipment and converter control, and specifically discloses a network-constructed converter control method and system for improving small disturbance stability. The application constructs an adaptive threshold judgment mechanism based on a deviation index: the maximum deviation index of the system in the stable parameter domain is determined according to various typical active control strategies, and the adaptive threshold is set by taking the intermediate value; in actual control, the maximum deviation index corresponding to the current control strategy is compared with the threshold, and the converter model considering or ignoring the reactive dynamic coupling is adaptively selected for controller parameter setting. Since the mechanism can dynamically evaluate the influence degree of the reactive dynamic coupling on the stability, the small disturbance stability analysis accuracy is ensured, the model complexity and the calculation efficiency are effectively balanced, the stability misjudgment caused by excessive simplification of the model is avoided, and the system stability margin misjudgment problem caused by ignoring the reactive dynamic coupling in the prior art is solved.

The invention belongs to the technical field of ship structure manufacturing and welding processes, and discloses a geometric dimension control and parameter optimization method for ship body angle welding and butt welding, which comprises the following steps: S1, constructing a classified welding process database and carrying out data preprocessing; s2, constructing and training a neural network forward prediction model special for a welding seam; s3, the process parameters are intelligently and optimally adjusted through the neural network forward prediction model special for the welding seam, and an optimal process parameter combination is output; and S4, performing experimental verification on the optimal process parameter combination output in the step S3, and performing closed-loop evolution on the optimized fillet weld time sequence enhanced neural network and butt weld time sequence enhanced neural network. According to the method, the design variable, which directly determines the material consumption, of the geometric dimension of the welding seam is used as an accurate control target, and direct and active control over the welding cost is achieved on the premise that the welding quality is guaranteed through intelligent reverse optimal adjustment of the technological parameters.

This invention relates to a high-fidelity torque measurement device and method based on resonance suppression and measurement optimization, belonging to the field of high-precision detection technology for spacecraft, and includes a closed-loop measurement scheme in both hardware and software. The device comprises a high-precision triaxial air-bearing stage, a high-frequency torque measurement sensor system, a fiber optic inertial measurement unit, an active control actuator system, and a host computer control and data processing system. The method sequentially achieves precise locking of inertial parameters with variable weights for measurement innovations through pre-test calibration, completes wideband resonance active suppression based on inertial linkage during testing, and achieves mode-free signal purification through variational mode decomposition using multidimensional measurement optimization after testing. This invention overcomes the physical limits of traditional mechanical structures in broadening the measurement bandwidth, eliminates dynamic calibration model mismatch errors, and achieves high-fidelity and robust measurement of wideband dynamic torque, suitable for precise detection of high-frequency micro-vibration dynamic torque in spacecraft actuators.

A hybrid active controller for practically asymptotically stabilizing the DC-DC boost converter under parameter uncertainty. The controller uses an estimation algorithm that identifies the input voltage and output load of the converter in finite time. Using these estimates, the control algorithm “unites” global and local control schemes. The global control scheme induces practical asymptotic stability of a desired output voltage and corresponding current, and the local control scheme maintains industry-standard PWM behavior during steady state. Stability properties for the resulting hybrid closed-loop system are established and simulation results illustrating the main results are provided.

The invention relates to the technical field of third-generation grid-controlled power semiconductor devices, and discloses a digital closed-loop grid active real-time control driving device, which comprises a sampling element, a control element and an execution element, the input end of the sampling element is connected with a SiC MOSFET (Metal-Oxide-Semiconductor Field Effect Transistor) device, the output end of the sampling element is connected with the control element, the control element is connected with the execution element, and the sampling element is connected with the control element. The output end of the execution element is connected with the SiC MOSFET device, the control element receives an analog sampling signal of the sampling element, the analog sampling signal is processed to obtain an analog modulation signal, the analog modulation signal is sent to the execution element, and the execution element processes the analog modulation signal to regulate and control the gate-source voltage of the SiC MOSFET device. According to the digital closed-loop grid active real-time control driving device, in the submicrosecond transient process of a SiC MOSFET device, high-speed response and accurate control of grid active control driving are achieved based on existing software and hardware conditions, and therefore quantitative regulation and control of peak stress and switching loss in the transient process of the device are achieved.

The invention relates to a natural gas boiler noise reduction method and a vibration suppression device in the technical field of natural gas boilers, the noise reduction method comprises the steps S1 to S5, and the vibration suppression device comprises a sensing and diagnosis module, an active control module, a passive treatment module and a central processing and closed-loop optimization unit. According to the technical scheme, dynamic, multi-target collaborative and closed-loop optimization is realized based on a data model and a collaborative technology; the problem of sound source accuracy is solved through multi-source sensing and intelligent diagnosis in the steps S1-S2, the problems of effectiveness and collaboration are solved through active and passive multi-physics field cooperative control in the steps S3-S5, and the problems of adaptability and long-term effectiveness are solved through a device-based central closed-loop optimization unit. Therefore, fundamental, intelligent and systematic treatment of noise and vibration problems of the natural gas boiler is realized.

The invention relates to an intelligent water level control device and method based on cosmic ray muon flux. The device comprises a muon detector, a liquid level meter, water level adjusting equipment and a central processing unit which is in communication connection with the muon detector, the liquid level meter and the water level adjusting equipment. The central processing unit executes a closed-loop control process, firstly sets a target water level, controls the adjusting equipment, regulates the target water level according to feedback of the liquid level meter, then collects and monitors real-time muon flux data, judges whether collection meets preset conditions or not, sets a next target water level if the collection meets the preset conditions, regulates and controls the next target water level repeatedly, and monitors continuously if the collection does not meet the preset conditions until a preset measurement sequence is completed. According to the invention, a data acquisition system and a water level control system are integrated, accurate water level regulation and control are realized, dynamic decision is made through a muon flux monitoring result, dynamic coupling feedback of data acquisition and water level control is achieved, data acquisition sufficiency can be automatically judged, a measurement state can be intelligently switched, water depth-flux model experiment efficiency and automation level can be improved, and the water level can be accurately controlled. And the requirements of active control variables and intelligent experiments are met.

The invention relates to the technical field of open caisson construction, in particular to an active control type assembly type mechanized open caisson construction visualization system and method, and the method comprises the following steps: obtaining construction data of open caisson construction equipment in real time; based on construction data, obtained in real time, of the open caisson construction equipment, the corresponding robot three-dimensional model and the open caisson three-dimensional model are driven to synchronously move and are displayed; and in the intelligent control mode, the intelligent control module is in communication connection with the open caisson construction equipment, receives an operation instruction of a user and issues the received operation instruction to the open caisson construction equipment for execution so as to realize intelligent control. According to the method, the digital twinborn model is constructed, visualization and whole-process analog simulation of the whole operation process of the underwater robot and visualization of the open caisson posture and construction information are achieved, and an operator is helped to complete settlement construction work more accurately and efficiently.

The present application relates to the technical field of zero-carbon factory operation and maintenance central control, and particularly relates to a zero-carbon factory intelligent operation and maintenance central control system and method, the system comprising a data platform, a digital twin construction module, a carbon trajectory real-time measurement module, a multi-target collaborative optimization engine, an edge-side active controller, a virtual power plant aggregation interaction module, a twin deduction and backtracking diagnosis module and a visual human-computer interaction interface, the carbon trajectory real-time measurement module comprising a dynamic emission factor library, and the multi-target collaborative optimization engine comprising a source-side intelligent agent, a load-side intelligent agent, a storage-side intelligent agent and a coordination intelligent agent. The present application can realize real-time carbon footprint tracking and transparent management, through multi-agent game, realize multi-target dynamic collaborative optimization of cost, carbon emission and green electricity consumption, improve response speed and reliability in the face of unexpected situations, so that the zero-carbon factory can not only realize internal low-carbon operation, but also obtain additional income by providing flexible services for the power grid.

The invention relates to the technical field of ship sealing systems, in particular to a ship tail shaft sealing self-balancing air control device which comprises a double-channel redundancy integrated control valve set and a special intelligent control system, and the valve set adopts an integrated main block to integrate an active control channel and a standby safety channel. The intelligent control system collects working condition data through an environment sensing module, calculates a comprehensive environment index through a data fusion module, dynamically adjusts pressure based on a fuzzy PID algorithm, and meanwhile has the functions of automatic fault switching and predictive maintenance. The problems that in the prior art, a ship tail shaft sealing device is lagged in control and insufficient in reliability are solved, the pressure control precision is high, sealing gas supply is uninterrupted, sealing ring abrasion and operation and maintenance cost can be reduced, and the device is suitable for ship tail shaft sealing scenes.

The invention relates to the technical field of tunnel lining protection, and discloses a tunnel lining crack active control method and system based on SHCC. The method comprises the following steps: continuously acquiring lining surface crack dynamic image data through a sensing array deployed in a tunnel; inputting the data into a crack semantic analysis module, calling a crack history archive and a tunnel structure database by the crack semantic analysis module for joint reasoning, and outputting a crack regulation and control target vector; constructing a geometric topology network with lining segments as vertexes and mechanical dependency relationships between the segments as edges, mapping regulation and control target vectors to the network, calculating influence scores of the vertexes through a similarity diffusion algorithm, and screening a key regulation and control vertex set; a candidate regulation and control instruction list is generated according to the set operation multi-index decision model, instructions are executed to complete SHCC material high-precision injection, active control over tunnel lining cracks is achieved, and the regulation and control requirements of crack dynamic changes in the tunnel operation process are met.

This invention relates to an active control method for suppressing lift fluctuations in a sphere based on reinforcement learning. A three-dimensional flow environment around the sphere is constructed, and a formula for solving the lift fluctuations of the sphere is provided. The feedback signal of the flow environment at the current moment is the lift fluctuation F′. l The control effect (reward value) from the previous moment is input to the controller in the closed-loop active flow control process. The controller outputs the action signal (rotation parameter) of the controlled flow environment. After receiving the action signal (rotation parameter), the flow environment executes the rotation parameter command and then transmits the control effect of the signal to the agent. This process is repeated multiple times to find the rotation control rate with the optimal control effect. In this invention, the agent module utilizes reinforcement learning to output multiple control signals to actuators at different positions. By appropriately setting the action network and evaluation network, the complex goal of suppressing lift fluctuations around a sphere can be achieved. This invention achieves a lift fluctuation suppression effect of approximately 70%.