54 results about "Nonlinear optimization problem" patented technology

The invention discloses a composite material storage life prediction method based on variable transformation and a genetic algorithm, and relates to the technical field of composite material performance evaluation and life prediction, and the method comprises the steps: obtaining performance degradation data at different temperatures through an accelerated aging test, and building a semi-empirical performance degradation model; specific variable transformation is introduced, a nonlinear optimization problem is converted into a linear regression problem, partial model parameters are rapidly solved by adopting a least square method, and meanwhile, key nonlinear parameters are globally optimized by using a genetic algorithm, so that the defect that a traditional method is easy to fall into local optimum is overcome. And establishing a temperature-service life relationship by combining with an archenius model, and realizing storage life prediction at normal temperature. The method also integrates sensitivity analysis and uncertainty quantification functions, and provides reliability evaluation for a prediction result. The method fully considers the influence of the environment temperature, meets the requirement of efficient prediction, and improves the precision and stability of life prediction.

The invention relates to a priority-based high-visibility stream-by-stream data packet sampling method and system. Comprising the following steps: step 1, flow initiation and priority marking; step 2, flow establishment and controller decision making; step 3, rule issuing and strategy execution; and step 4, data forwarding, scheduling and sampling: a priority scheduling mechanism of a data plane and global optimization of a control plane are decoupled and coordinated. In a data plane, a terminal host marks a degradation priority for a data packet according to the number of bytes sent by a stream, and a switch performs strict queue scheduling and probability sampling based on the priority. In a control plane, a controller carries out collaborative modeling on the configuration of routing, sampling probability and degradation threshold as a nonlinear optimization problem with maximization of sampling utility as a target, and an efficient online primitive-dual interior point method is adopted for solving. According to the method, high visibility and load balancing are finally ensured, and meanwhile, the control and calculation overhead is remarkably reduced.

The application discloses a flexible load intelligent optimization scheduling method for improving the economy of photovoltaic access distribution networks. A multi-type flexible load grading and scheduling optimization model is established. On the one hand, according to the demand side response demand, the flexible load is divided into three categories, namely interruptible load, translatable load and adjustable load, a flexible load operation optimization model is constructed, the reasonable allocation of various flexible loads in different periods is realized, the peak regulation pressure and resource waste problems in the distribution network are effectively alleviated, the purchase of electricity during the power consumption peak can be effectively avoided, and the economy of the distribution network is improved. On the other hand, for the discrete and continuous mixed data distribution environment, a Latin square quantum-inspired genetic algorithm (LSQGA) is used to process the nonlinear optimization problem in the environment, so that the optimal feasible region in the macro data space can be quickly searched, and the feasibility of the optimization model under the extreme scenario is improved.

The invention relates to the technical field of new energy automobile intelligent driving, and discloses a new energy automobile multi-mode sensing trajectory planning method and system, and the method comprises the steps: obtaining multi-mode original sensor data, and carrying out the preprocessing; feature extraction and cross attention fusion are carried out; performing dynamic obstacle prediction, and setting a dynamic safety buffer area; constructing a behavior decision input state vector, and reasoning and adjusting a behavior decision; generating a candidate trajectory curve and performing feasibility check on the candidate trajectory curve; comprehensively evaluating the candidate track set; predicting a nonlinear optimization problem by taking the initial optimal trajectory curve as a reference trajectory, and solving and verifying the nonlinear optimization problem; according to the method, robust fusion and adaptive weight adjustment of multi-modal sensor data are realized by establishing a sensor health monitoring mechanism, a dynamic weight distribution strategy and a cross attention fusion network.

The invention discloses a vehicle three-way cooperative auxiliary control method fusing PINN tire transverse and longitudinal force prediction. Comprising the following steps: constructing a dynamic model of a vehicle based on a force balance equation and a moment balance equation of the vehicle; constructing a physical information neural network model; training the physical information neural network model by taking the dynamic model as a dynamic constraint and taking the minimum physical residual error loss as a target to obtain a physical prediction model; the physical prediction model is fused, a cost function with the longitudinal resultant force, the additional yawing moment, the slippage loss power and the motor loss power of the vehicle under the action of the auxiliary control quantity as factors is constructed, the cost function is solved with the minimum cost function as the target, and the minimum cost function is obtained. The auxiliary control quantity used for conducting three-way cooperative auxiliary control on the vehicle is obtained; and adopting the auxiliary control quantity pair. According to the method, the problems that tire model parameters are difficult to calibrate and the cost is high during vehicle three-way cooperative auxiliary control, and key performance indexes of a cost function in a nonlinear optimization problem have characterization defects are solved.

The invention discloses a three-dimensional positioning method and system for a bimodal target, and relates to the technical field of indoor positioning, and the method comprises the steps: collecting the phase ranging data of a connected state base station and the received signal strength indication data of a broadcast state base station, and forming a bimodal perception data set; selecting a static positioning model or a dynamic positioning model according to the motion state of a to-be-positioned point, and converting a positioning problem into a nonlinear optimization problem based on the bimodal perception data set; and solving the nonlinear optimization problem by adopting a CBS algorithm to obtain the position estimation of the point to be positioned in the three-dimensional space. According to the method, three-dimensional positioning of a static target and smooth track positioning of a dynamic target are realized at the same time, the optimization range is effectively constrained, the system deployment complexity and the calculation load are reduced, the precision and robustness of indoor three-dimensional positioning are effectively improved, and the method is suitable for application scenes with strict requirements for quick response and low-cost deployment.

This application belongs to the field of unmanned surface vessel (USV) control technology, specifically relating to a method and apparatus for predictive control of USV trajectory tracking using dynamic reference linearization. The method includes: establishing and discretizing a nonlinear dynamic model of the USV to construct a predictive model; constructing an online nonlinear optimization problem including terminal stability components based on this model; generating a dynamic reference trajectory in each control cycle based on the optimal solution of the previous moment, and linearizing the nonlinear model point-by-point along this trajectory to obtain a high-fidelity linear time-varying predictive model; further transforming the original nonlinear programming problem into an efficiently solvable quadratic programming problem; and outputting control quantities in real time by solving this problem and integrating reference commands. This application ensures the theoretical stability of the closed-loop system, achieves a synergy of high precision, high real-time performance, and high reliability, and effectively improves the trajectory tracking performance of USVs in complex marine missions.

The application provides a multi-parameter joint calibration method based on graph optimization of inertial and ultra-short baseline, applies the graph optimization technology to the multi-parameter joint calibration of inertial / ultra-short baseline, and the whole process comprises the following steps: establishing a multi-error parameter state model, constructing a multi-error parameter calibration factor, calculating based on a nonlinear optimization problem of an incremental interval balance, and finally obtaining accurate multi-error parameters. Compared with the prior art, the application can simultaneously estimate the responder position error, the rod arm error and the installation angle error, sufficiently utilizes historical observation data, solves the nonlinear problem of the optimization equation through repeated iteration and relinearization, and has the advantages of high calibration precision and fast calibration speed.

The application belongs to the technical field of data product operation and power data asset optimization, and discloses a power data asset value dynamic application method and device based on multi-scene coupling, which comprises the following steps: constructing the interaction relationship between power generation data assets, power consumption data assets, power grid data assets and multiple application scenes, and constructing system operation core constraint conditions; quantifying the value linkage relationship between different application scenes and the adaptability of various data assets to each application scene; taking the maximization of comprehensive value as an objective function, fusing multi-scene benefits, costs and data constraints, and forming a high-dimensional nonlinear optimization problem; designing a solution method based on an adaptive improved whale optimization algorithm to output an optimal value application strategy under the premise of meeting real-time requirements. The application realizes the optimal value application and maximum benefit of power data assets in a multi-scene environment.

The application discloses a kind of hypersonic vehicle multi-source energy management methods, belong to the technical field of calculation, reckoning or counting.For the nonlinear optimization problem of multi-source energy distribution in mode conversion process, the method first establishes the mechanism model of TBCC engine and multi-source power generation system, quantifies the correlation between power generation and fuel consumption;Subsequently, an adaptive sparse Kriging surrogate model is proposed, which improves the generalization ability of large-scale data through low-rank approximation of covariance matrix and dynamic selection strategy of inducing points;Further, an optimization function is designed to minimize the cumulative fuel consumption, and an improved DQMR-DDPG algorithm is integrated to solve the problems of policy oscillation and Q-value overestimation through state hierarchical experience pool, double Q-network and adaptive soft update mechanism.The application realizes real-time optimization of energy management under complex conditions, while meeting the thrust, power and safety constraints, significantly reducing fuel consumption during mode conversion phase.

The invention discloses an air conditioner temperature non-sensitive control method based on online learning, and the method specifically comprises the steps: S1, collecting indoor environment parameters, outdoor meteorological parameters, equipment operation parameters and energy price parameters, and forming a multi-source parameter time sequence; s2, constructing a Gaussian process state space model in which a multi-source reliability gating factor is introduced, and generating a cooling and heating load state prediction sequence; s3, constructing a nonlinear optimization problem based on the cooling and heating load state prediction sequence; s4, setting a switching penalty serialization item in the nonlinear optimization problem; s5, solving a nonlinear optimization problem by adopting a sequential quadratic programming algorithm, and generating a control variable updating sequence; s6, an air conditioner operation control strategy is generated based on the control variable updating sequence; and S7, calculating a prediction deviation and updating a multi-source reliability gating factor after the control period is finished, and completing an online learning process. According to the method, through reliability gating prediction and continuous optimization, air conditioner operation stability and somatosensory stability are both considered.

The invention relates to the technical field of robots, and discloses a robot, a recharging control method and device thereof and a storage medium. The method comprises the steps that the relative pose of the robot relative to a charging pile is acquired; based on the relative pose, solving the constructed nonlinear optimization problem with the operation constraint, and generating a reference trajectory from the position of the robot to the target preparation point; wherein at the target preparation point, the terminal motion state of the robot comprises a non-zero linear speed suitable for the robot to directly enter a charging docking state; performing real-time tracking on the reference trajectory by adopting model prediction control, and outputting a control instruction based on a tracking result to drive the robot to move along the reference trajectory; and after the robot arrives at the target preparation point and moves to be connected with the charging pile, the robot stops moving and is charged. According to the method, seamless charging in the robot recharging process can be achieved, namely seamless connection from advancing to charging is achieved, and the recharging efficiency and stability are improved.

The invention discloses a twinborn control-oriented magnetic drive rotor key parameter online identification method, which comprises the following steps of: constructing a parameterized digital twinborn model of a magnetic drive rotor, and defining time-varying key internal parameters in the model as parameter vectors to be identified; in a set time window, an objective function is defined, and a parameter identification problem is converted into a nonlinear optimization problem which minimizes the objective function; aiming at the problem, system parameter self-adaptive identification is carried out based on an LM optimization method, and a parameter updating vector is obtained; and transmitting the parameter update vector to a system matrix of the parameterized digital twinborn model for state prediction at the next moment, and outputting the parameter update vector to complete online identification of the key parameter. According to the method, the precision and real-time performance of parameter identification and the adaptability of an embedded platform are effectively improved, and a reliable model basis is provided for high-precision control, performance degradation monitoring and predictive maintenance of the magnetic drive rotor.

The invention discloses a power distribution network low-carbon high-quality scheduling method, system, equipment and medium considering distributed robust optimization segmented affine, and belongs to the technical field of power distribution network scheduling, and the method comprises the steps: taking the minimization of the carbon emission of the product of the main network carbon potential and the net load power as a target function; a Wasserstein distance is adopted to measure the deviation between historical data and real distribution, and multiple types of operation constraints are considered at the same time; the above distribution robust optimization model is converted and solved by adopting a piecewise affine theory, an original nonlinear optimization problem is converted into a linear optimization problem based on a simplex dominant uncertainty set, and the model solving efficiency and robustness are improved. According to the method, the uncertainty of new energy output and carbon potential is effectively dealt with through Wasserstein distance constraint, and the robustness is high; by means of segmented affine transformation, nonlinear robust optimization is simplified into linear optimization, and the solving speed is remarkably increased.

The invention discloses a beam forming optimization method based on sum rate maximization. According to the method, a multivariable coupling nonlinear optimization problem is constructed, and an iterative optimization algorithm based on an alternating optimization framework is provided to jointly optimize a digital beam matrix, a holographic beam matrix and an RIS phase shift diagonal matrix. According to the method, the IOA based on the Dinkelbach algorithm is utilized to solve the holographic beam forming problem; aiming at a joint optimization problem of a digital beam matrix and an RIS phase shift diagonal matrix, an original problem is converted into a positive semidefinite programming problem with a concave difference form by introducing a positive semidefinite matrix, a penalty term is introduced to process non-convex rank-one constraint, and an IOA solving method based on positive semidefinite programming and continuous convex approximation is provided. The method provided by the invention is superior to a traditional MISO SWIPT system and an RIS-assisted SWIPT system. In addition, through holographic beam forming, higher system sum rate and resource utilization rate can be realized on the premise of satisfying energy collection constraints.

The invention discloses a displacement estimation method and system for transcranial ultrasonic shear wave viscoelastic fluid imaging, and belongs to the technical field of ultrasonic medical image processing, and the method comprises the steps: dividing ultrasonic radio frequency data after beam forming into a plurality of independently processed data blocks; the ultrasonic radio frequency data comprises two frames of ultrasonic radio frequency signals collected before and during shear wave propagation; constructing a cost function for each data block, wherein the cost function comprises a data similarity item and a mixed-order displacement continuity constraint item; performing linearization processing on the cost function, and converting a nonlinear optimization problem into a linear problem; establishing and solving a linear equation set to obtain local displacement of each data block; and splicing the local displacement of each data block to form a global displacement field. According to the method, high-precision estimation of the transcranial shear wave displacement and reliable evaluation of the viscoelastic fluid property of the brain tissue can be realized, relatively efficient and accurate displacement estimation is kept in a transcranial high-noise environment, and the performance of transcranial ultrasonic shear wave elastography is improved.

The invention relates to the technical field of power grid planning, in particular to a high-performance parallel automatic differential method, system, equipment and medium for main distribution micro power flow nonlinear optimization, and by adopting the method provided by the invention, modeling is carried out based on a main distribution micro power flow nonlinear optimization problem, and an expression diagram is constructed; considering that a large number of isomorphic constraints exist in the problem, the constructed expression graphs are subjected to grouping management, and an automatic differential calculation function pointer used for calculating each group of instances is generated for a plurality of obtained expression groups. And finally, based on the constructed automatic differential calculation function pointer, introducing an OpenMP parallel framework, and carrying out automatic differential parallel calculation on each expression group. Based on the experimental result of opfbenchmark, the method provided by the invention improves the calculation efficiency of automatic differentiation by 47% in the main distribution micro nonlinear optimization problem.

The application provides a sequential linearization solving method and device for network-constructed energy storage optimal control problems, comprising: establishing a system frequency dynamic characteristic representation model and an energy storage dynamic characteristic representation model; constructing a model prediction optimal control framework for network-constructed energy storage, considering the coupling characteristics of system frequency dynamics and energy storage dynamics, and optimizing the optimal control strategy under the minimum cost of network-constructed energy storage; analyzing the linearization constraint conditions of the model prediction optimal control framework, and realizing the linearization representation of the system frequency dynamics and the energy storage dynamics; analyzing the linearization objective function of the model prediction optimal control framework, and realizing the linearization mapping between the optimization objective and the system state and the energy storage state; constructing a sequential linearization solving framework, converting the nonlinear optimization problem of the optimal control framework into a group of iterative and progressive linear optimization problems, and realizing the fast and stable convergence of the algorithm through an acceptance rejection mechanism and a credible domain updating mechanism.

The invention provides a navigational speed optimization method and system based on a ship energy consumption prediction model, and relates to the technical field of industrial intelligent optimization. The method comprises the following steps: constructing a ship energy consumption prediction model; based on the ship energy consumption prediction model, establishing a navigational speed optimization model; and an SLSQP algorithm is adopted to solve the navigational speed optimization model. According to the invention, dynamic coupling of environmental factors is realized; and an SLSQP algorithm is adopted to convert a nonlinear optimization problem into a quadratic sub-problem for iterative solution, so that the calculation efficiency is remarkably improved, the blank of an existing method in the aspects of environmental adaptability, algorithm efficiency and real-time performance is filled up, and a key tool is provided for energy conservation and emission reduction of the shipping industry.

This application relates to a collaborative control method and system for high and low voltage equipment under complex operating conditions, and falls within the field of high and low voltage equipment control technology. The method includes: acquiring real-time operating data of the target prefabricated substation; calculating the comprehensive disturbance intensity and determining the severity level of the fluctuation; based on the severity level, invoking a collaborative protection risk assessor to obtain the risk probability of protection over-level operation under the current voltage fluctuation condition; when the risk probability exceeds a set risk threshold, constructing a nonlinear optimization problem with minimizing the risk probability as the optimization objective, using the action time limits of the high-voltage and low-voltage sides as decision variables, and constrained by the physical limits of the equipment; solving the nonlinear optimization problem to obtain the optimal action time limit combination, and generating and issuing collaborative control commands. This invention solves the problems of slow response in traditional high and low voltage equipment fault handling, inability to cope with transient disturbances such as voltage fluctuations, and lack of effective coordination between high and low voltage side equipment, which easily leads to improper fault handling coordination.

The invention provides a fuzzy self-adaptive Q learning control method and system for sewage treatment, and the method comprises the steps: obtaining the dissolved oxygen concentration as a system state, and constructing a nonlinear optimization problem; performing dynamic adjustment on proportion, integral and differential coefficients by adopting a Mamdani type fuzzy inference rule in combination with fuzzy logic, performing defuzzification through a centroid method to obtain adaptive PID parameters, and forming a fuzzy adaptive control strategy; an online Q learning framework is further constructed, a Q function is approximated by using the evaluation network, a network generation strategy adjustment amount is executed, and network weight is optimized based on a Bellman equation and a particle swarm algorithm; finally, fuzzy control and Q learning output are fused, control input is generated through a coupling coefficient to adjust an oxygen transfer coefficient, and tracking control over the dissolved oxygen concentration is achieved. According to the invention, the tracking control precision of the dissolved oxygen concentration and the system operation stability can be improved.

This invention discloses a method for dynamic autonomous navigation of ships based on domain-type gradient fields, comprising the following steps: establishing a domain-type gradient field decreasing model for ship encounter scenarios; establishing a ship navigation motion constraint model; after establishing the ship navigation motion constraint model, establishing corresponding domain-type gradient field models for different ship encounter scenarios; transforming the route design problem for different ship encounter scenarios into a constrained nonlinear optimization problem and solving it to achieve local route design and reach the target point. During ship operation, this invention utilizes an intelligent autonomous device on board to collect information about the ship itself, other ships, and obstacles. By judging the positions and states of all parties during navigation, it provides effective motion decisions to ensure ship safety.

The invention discloses a power distribution network topology reconstruction and distributed power supply output coordinated optimization method, system, device and medium, and belongs to the technical field of intelligent power distribution network operation optimization and control, and the method comprises the steps: collecting power distribution network data from a power distribution automation system for data preprocessing, collecting each distributed power supply data, and generating a scene set; topology reconstruction is carried out by changing the states of the interconnection switch and the section switch to obtain the optimal switch state, and the active power output and the reactive power output of the distributed power supply are optimized to obtain the optimal power supply output. Two-stage optimization is carried out, coordination is carried out through an iteration mechanism, a switch action sequence is generated according to the target topology and the current topology obtained through optimization, instruction issuing is carried out, and the network state verification optimization result is checked step by step. According to the method, global coordination optimization of topology reconstruction and distributed power supply output is realized, the adaptive capacity of the optimization scheme to the uncertainty of the distributed power supply is improved, and the solving efficiency of a mixed integer nonlinear optimization problem is remarkably improved.

The application discloses a composite material storage life prediction method based on variable transformation and a genetic algorithm, relates to the technical field of composite material performance evaluation and life prediction, and comprises the following steps: obtaining performance degradation data at different temperatures through an accelerated aging test, and establishing a semi-empirical performance degradation model. By introducing a specific variable transformation, a nonlinear optimization problem is converted into a linear regression problem, least square method is used to quickly solve part of model parameters, and a genetic algorithm is used to globally optimize key nonlinear parameters, so that the defect that a traditional method is prone to local optimization is overcome. The temperature-life relationship is established in combination with an Arrhenius model, and storage life prediction at normal temperature is realized. The method further integrates sensitivity analysis and uncertainty quantification functions, and provides reliability evaluation for the prediction result. The application fully considers the influence of environmental temperature, meets the demand of efficient prediction, and improves the precision and stability of life prediction.

The application provides a floating offshore wind turbine safety control method, steps comprising: constructing a mechanical-electrical coupling model based on a secondary problem description, converting into a control affine system, and studying the relationship between the blade pitch angle and the generator rotor angular velocity; constructing a control Lyapunov function candidate according to the speed reference value under the over-rating operation mode, constructing a control barrier function candidate according to the platform pitch angle threshold, and verifying the rationality; taking the control Lyapunov function and the control barrier function as constraints, constructing a control framework of a nonlinear optimization problem, and solving the blade pitch angle control output. The method can effectively improve the generator speed tracking performance under the over-rating condition on the premise that the platform pitch angle does not exceed the safety boundary, enhance the adaptability of the control strategy to the complex marine environment and model uncertainty, and has good engineering application value and popularization prospect.

The application discloses a robot motion planning method based on rolling search and optimization and related equipment, which allows an autonomous mobile robot with sensor constraints to dynamically plan an obstacle avoidance trajectory in an unknown complex and dense obstacle environment. The method comprises the following steps: firstly, an initial feasible path satisfying the dynamics of the autonomous mobile robot is planned through two process searches; secondly, a nonlinear optimization problem considering path safety and smoothness is solved to improve the path smoothness, with path points as variables; then, the feasible path is time-parameterized by using a uniform B-spline rolling method, and control points are optimized to obtain a safe and dynamically feasible minimum time trajectory; finally, the trajectory risk is evaluated according to the distance relationship between the B-spline trajectory and the obstacle, so that the trajectory time is adjusted. The application is verified through simulation and physical experiment, and can realize risk perception of the autonomous mobile robot in the dense obstacle environment and adaptive speed adjustment.

This invention provides a predictive control method for laser processing mirrors based on an extended-dimensional mapping dynamic model, belonging to the field of laser processing mirror motion control. The method includes the following steps: S1, proposing a dynamic model based on an extended-dimensional mapping of a neural network, combining neural networks and linear models to characterize the dynamic characteristics of the dual-axis laser processing mirror; S2, identifying the parameters in the extended-dimensional mapping dynamic model by solving an optimization problem based on input-output data; S3, designing a linear quadratic programming problem based on the extended-dimensional mapping dynamic model and the control objective; S4, solving the quadratic programming problem to obtain a predictive controller with nonlinear feedforward compensation capability. This invention mainly addresses the trajectory tracking problem of dual-axis laser processing mirrors under high-frequency sampling, solving the problems of insufficient tracking accuracy caused by the nonlinearity of the mirror driver itself and the difficulty in solving nonlinear optimization problems under rapid sampling.

The invention provides a time optimal model prediction control method for a tendon-driven redundant mechanical arm, and belongs to the crossing field of prediction control and optimization control in the control theory. According to the method, a dynamic adjustable scale factor is introduced into a model predictive control (MPC) framework, time optimality is brought into an optimization objective, and a time optimal model predictive control (TOMPC) optimization problem including multiple physical constraints such as tendon length and speed is constructed; performing efficient iterative solution on the nonlinear optimization problem by using a sequential quadratic programming (SQP) algorithm to obtain an optimal control sequence and a scale factor in a future time domain; and an optimal control instruction at the current moment is extracted to drive the mechanical arm to move, and closed-loop rolling optimization is achieved through state feedback. According to the method, on the premise that system constraints are met, the tracking speed of the end effector is maximized, the task completion time is remarkably shortened, meanwhile, the tracking precision and operation safety are guaranteed, and the operation efficiency and economical efficiency of the redundant mechanical arm in an industrial automation scene are effectively improved.

The application provides a flexible deformable part visual tracking and robot pose control method, and belongs to the field of robot vision and automatic assembly. First, an ASLIC superpixel segmentation algorithm is proposed by fusing edge perception, adaptive compactness and adaptive grid step, and combined with feature fusion and global topology constraint optimization path search, a two-dimensional center line is accurately extracted. Secondly, a three-dimensional pose structured representation is constructed by projection error compensation and adaptive cylindrical constraint. Finally, in the deformation control stage, the stiffness decreasing mechanism is revealed, the bending and torsion deformation is decoupled based on the parallel transmission of the Frenet moving frame, and the torsion angle is estimated, and finally a bending and torsion coupled total potential energy model is constructed, the pose prediction is converted into a nonlinear optimization problem, and through a hierarchical controller, the position and torsion of the flexible deformable part under the robot control are cooperatively and high-precision closed-loop controlled. The application can realize stable visual capture, accurate three-dimensional reconstruction and bending and torsion cooperative control of the flexible deformable part.

This invention proposes a nonlinear model predictive control method based on GRU equivalent wind speed prediction. First, an objective function is established to achieve multiple control objectives, including maximizing wind energy capture, reducing power output fluctuations, and minimizing fatigue damage to the transmission system; this is the proposed nonlinear model predictive control method. Then, the GRU method is used to predict the equivalent wind speed on the impeller surface in advance, and this wind speed information is used as the control input for the nonlinear model predictive control. Finally, since traditional numerical solution methods are not suitable for solving nonlinear optimization problems, an efficient search algorithm based on iterative optimization and an efficient search solver is employed to solve for the nonlinear optimal solution. This nonlinear model predictive control method captures more wind energy than the classical optimal torque control method and exhibits better performance in suppressing fluctuations, while also extending the service life of unit components. This method demonstrates strong robustness and good control performance, providing a new approach and method for improving maximum power point tracking performance.