56 results about "Stability constraints" patented technology

The invention relates to an unmanned aerial vehicle trajectory tracking control method based on model prediction and preset performance constraint, and the method proposes to introduce Lyapunov stability constraint into a model prediction control framework and integrate a preset performance control mechanism, thereby achieving the unification of performance constraint and system stability analysis. Comprising the following steps: establishing a nonlinear system model based on unmanned aerial vehicle dynamics; position errors and attitude errors are defined, a preset performance function is constructed, and errors with performance constraints are converted into unconstrained errors through error normalization and nonlinear transformation; establishing a model prediction optimization problem on the premise of considering input saturation and stability constraints; designing an auxiliary control law based on the transformation error to construct a stability constraint; it is proved that the control strategy can ensure that errors meet preset performance constraints and system local asymptotic stability. According to the invention, stable and reliable trajectory tracking control of the unmanned aerial vehicle system can be realized, and the method has high tracking precision and good dynamic performance.

The invention provides an optimal primary frequency control method based on reinforcement learning. The method comprises the following steps: establishing a dynamic model of a power system; designing a frequency control strategy based on a deep reinforcement learning algorithm; the system frequency deviation is used as a reward signal, and a control strategy is trained through interaction with a power system; lyapunov function stability constraints are introduced into the reinforcement learning algorithm; a neural network is adopted to carry out parameterization design on the controller; discretizing the frequency dynamic equation, and proposing a multi-node collaborative optimization target; and deploying the trained control strategy to an actual power system, and adjusting control parameters in real time through an online optimization technology to adapt to the dynamic change of the system operation state. By means of the method, the frequency control accuracy and stability of the power system are improved, and the global optimal performance of frequency control of the whole power grid is achieved in actual deployment.

The invention discloses a space-ground integrated resource scheduling method and system, electronic equipment and a medium, and the method comprises the steps: training strategy parameter prediction models of all agents based on a first loss function and a second loss function, and outputting a new strategy parameter set after the training is completed; the new strategy parameter set is used as a current strategy parameter set in a transfer learning stage, so that a strategy parameter set after transfer updating is calculated, and a third loss function is constructed; constructing a stability constraint term according to the migrated and updated strategy parameter set and the current strategy parameter set, and adding the stability constraint term to a third loss function to obtain a total objective function; circularly training the strategy parameter prediction model and the transfer learning stage until the total objective function is converged to a preset threshold value, and obtaining a target strategy parameter set output by the transfer learning stage; and deploying the target strategy parameter set to the space-ground integrated network. According to the invention, the real-time performance of resource scheduling can be realized and the reliability of resource scheduling can be improved.

The invention discloses an SCR denitration ammonia escape and NOx intelligent control method based on machine learning. The method comprises the following steps that boiler operation data are collected and preprocessed; constructing a feature input tensor set; inputting the feature input tensor into a prediction model based on a neural control differential equation, and outputting a predicted NOx and ammonia escape concentration sequence in combination with a Koopman observable space and an energy gating mechanism; constructing a candidate ammonia injection distribution set meeting the valve constraint based on the prediction result; multi-objective optimization is carried out under emission limitation and ammonia injection stability constraint, and an optimal ammonia injection distribution vector is generated; projecting to a feasible region barrier set and then generating an ammonia spraying execution instruction; and collecting feedback data for incremental training and model adaptive updating. The ammonia spraying control precision and the denitration efficiency are improved, and intelligent prediction of ammonia escape and self-adaptive control of NOx are achieved.

The invention relates to the technical field of space optimization, and discloses a container loading space optimization method based on robot collaboration, and the method comprises the steps: building a multi-dimensional physical model of a package through obtaining the point cloud data and weight distribution of the package; generating a loading scheme containing a target pose, a pressure tolerance threshold and a loading sequence based on a reinforcement learning algorithm; the mechanical arm generates a motor current-position composite control instruction according to a pre-trained grabbing dynamical model, and accurate grabbing is achieved; in the loading process, the pose of the parcel is monitored in real time through a Kalman filtering algorithm, when deviation exceeds a threshold value, a local space KD tree is constructed, a collision probability gradient field is calculated, and a compensation scheme meeting stability constraints is generated; and finally, updating the quad-tree index and feeding back to the reinforcement learning algorithm for online optimization. The problem that in the robot loading process, the grabbing force control is not accurate, and the efficiency is reduced due to execution deviation accumulation is effectively solved, and efficient and accurate automatic loading is achieved.

The invention belongs to the technical field of artificial intelligence, and particularly relates to a cross-platform data interaction sharing method and system based on artificial intelligence. Comprising the following steps: constructing an edge cloud collaborative architecture for health degree perception, and establishing a dynamic health degree baseline for nodes; constructing a correction reward function taking the node health degree as a stability constraint, and training by adopting a multi-agent reinforcement learning model to obtain a Nash equilibrium optimal strategy; decoding the optimal strategy into an executable cross-platform data interaction planning graph through a strategy interpreter network, and coordinating resource conflicts by utilizing a game theory; on the basis of the planning graph, proxy re-encryption and hierarchical homomorphic encryption of context awareness are executed, and an authority token chain is established to realize fine-grained authority tracking; closed-loop optimization of strategies and system self-evolution are carried out through multi-dimensional quality scoring, multi-layer traceability analysis and an elastic weight solidification technology; cooperative gains of cross-platform data interaction in three dimensions of efficiency, security and long-term stability are realized.

The invention discloses a stability constraint LQR single-rotor unmanned aerial vehicle trajectory tracking control method based on GWO-PSO optimization. The method comprises the following steps: constructing a nonlinear dynamic model of a single-rotor unmanned aerial vehicle; defining a hovering point as a unique balance point for stable work of the single-rotor unmanned aerial vehicle, and obtaining a linearization model surrounding the balance point according to the nonlinear dynamic model based on the balance point; establishing an optimization model considering model uncertainty and external interference according to the linearization model so as to design an initial feedforward feedback LQR control law based on stability constraint; taking a weight matrix Q and a weight matrix R in the initial feedforward feedback LQR control law as a group of parametric variables, and optimizing the parametric variables based on the constructed GWO-PSO hybrid optimization algorithm to obtain an optimal variable; and obtaining an optimal feedforward and feedback LQR control law according to the optimal variable, thereby realizing trajectory tracking control of the single-rotor unmanned aerial vehicle. According to the invention, the problem that the existing method is insufficient in trajectory tracking precision, anti-interference capability and robustness is solved.

This invention discloses a method for joint evaluation and optimization of buckling and bending of arbitrary quadrilateral laminates based on physical information neural networks, relating to the field of composite material structure design technology. The method, for arbitrary quadrilateral laminates, first constructs a method based on classical laminate theory... USA The stiffness matrix explicitly reflects the influence of ply parameters, and a physical domain is transformed to a reference domain through a bitriangular affine mapping to establish a physical information neural network mechanical model. The model is trained with the goal of minimizing the total potential energy or Rayleigh quotient, solving for the buckling factor and layer-by-layer strain limits, and adapting to four typical boundary conditions through a boundary embedding function. Subsequently, using ply angle and thickness as variables, optimization is performed using a genetic algorithm, with the goal of minimizing mass, and the buckling factor and strain limits as constraints. The final output is the optimal ply scheme that satisfies specific boundary conditions. This invention can optimize ply angle and thickness while satisfying buckling stability and strain strength constraints.

The invention discloses a large model reasoning stability enhancement method and system in a text scene, and relates to the technical field of knowledge enhancement deep learning, and the method comprises the steps: building a structured text knowledge graph based on a text knowledge base, and generating a dynamic text knowledge embedding matrix through a graph attention network; inserting a text knowledge gating cross attention module into a decoder selection layer of the pre-trained large model, taking the text knowledge gating cross attention module as an external knowledge source, obtaining a knowledge enhanced hidden state after gating fusion, and constructing a transformation model; semantic equivalent perturbation is carried out on an input text to obtain a perturbation sample, the perturbation sample is input into the transformation model in parallel to obtain extraction probability distribution, and divergence and gradient direction consistency loss between two distributions are calculated; and combining cross entropy loss and gradient direction consistency loss to train and transform the model, and updating parameters to convergence to obtain a final large model. According to the method, the fact consistency of output can be improved, common optimization of knowledge guidance and stability constraint is realized, and the result is accurate and reliable.

The invention belongs to the technical field of power system dynamic safety analysis and renewable energy source grid connection, and provides a method for determining the permeation limit of high renewable energy sources under multi-dimensional stability constraint. The method takes physical mechanism depth modeling-multi-dimensional stability coupling theory-intelligent data driven screening as a core, constructs a quantitative correlation model of frequency stability, transient stability and small disturbance stability through rigorous theoretical derivation, proposes a global-region double-layer dynamic inertial constraint mechanism, and combines an AI-driven critical scene screening and clustering method to obtain a critical scene clustering algorithm. And finally determining the safe RE penetration limit. According to the method, the limitation of traditional single-dimension evaluation and static constraint is broken through, all theoretical derivation is based on the dynamic characteristics and the statistical learning principle of the power system, moderate data verification is assisted, and a complete theoretical system and an engineering tool are provided for RE safety integration of the low-inertia power grid.

The invention relates to the technical field of power grids, in particular to a distributed new energy acceptance capacity dynamic calculation method and system, and the method comprises the steps: constructing a combined small-disturbance dynamic model for distributed new energy access, and obtaining a stable constraint boundary; establishing an optimization model by taking distributed new energy acceptance capacity maximization as a target function; and solving the optimization model by adopting a self-adaptive particle swarm algorithm to obtain the maximum acceptance capacity of the distributed new energy in the current operation state of the power distribution network. According to the invention, based on the equivalent dynamic model of the network-forming inverter and the state-space equation of the power distribution network, the combined small-disturbance dynamic model is constructed and the stability analysis is carried out, the stability constraint is directly introduced into the acceptance capacity optimization, and the adaptive particle swarm algorithm is adopted for solving, so that the maximum access capacity of the distributed new energy can be accurately evaluated. Meanwhile, small-interference stability of the system and operation constraints of the power distribution network are considered, and calculation precision, efficiency and engineering application value are improved.

The application discloses a power transformer temperature monitoring method, device and equipment, and relates to the technical field of power equipment state monitoring and intelligent modeling. First, according to the load factor, the average winding temperature and the position of the tap switch of the power transformer at the current moment, the physical model of the power transformer is used to calculate the heat generation power at the current moment; the heat generation power at the current moment is brought into a heat balance equation for solving to obtain the top-layer oil temperature simulation value at the current moment; then, the physical characteristics at the current moment and a preset number of moments before the current moment are combined to form an enhanced feature sequence; the enhanced feature sequence is input into a deviation prediction model to obtain the dynamic deviation of the top-layer oil temperature at the current moment, the top-layer oil temperature simulation value at the current moment is calibrated, and the top-layer oil temperature monitoring value is obtained. The application provides prior knowledge through the physical model, combines data-driven learning of the deviation dynamic characteristics, and applies stability constraints, so that high-precision and stable oil temperature prediction is realized.

The invention provides a new energy power system reliability evaluation method based on environment variables, and relates to the technical field of power systems. According to the method, a non-parametric self-adaptive quantile regression forest model is adopted to perform modeling on wind speed and sunlight intensity, and online incremental updating is triggered based on Kolmogorov-Smirnov statistics; setting a cross-scale state coupling interface layer in the double-layer nested reliability evaluation architecture, defining a shared state vector including an energy storage charge state, an equipment aging index and an available reserve capacity, and applying Lyapunov stability constraint; a period economic value dynamic pricing module based on Q-learning is integrated in a short-term evaluation layer, and a real-time electricity price, a user interruption contract and a load elastic coefficient are used as state spaces to correct a load shedding strategy. According to the scheme, through adaptive modeling of environment variable distribution and cross-scale state stable transmission, the problem of evaluation result lag is effectively solved, and the real-time performance and economical efficiency of reliability evaluation are improved.

The application discloses a kind of low-loss proportional amplification fluxgate current sensor measurement methods, it is related to current measurement technical field, including, obtaining the observation set of synchronous demodulation by phase-locked;Based on the scheduling optimization function of prediction set construction, apply soft opening constraint, amplitude stability constraint, phase consistency constraint and saturation depth constraint, output corresponding control variable;According to control variable setting excitation signal parameter and executing excitation drive and compensation injection, extract the harmonic amplitude-phase response of current cycle in demodulation window position;Harmonic amplitude-phase response is converted into linear measurement output of current cycle, while updating observation set.The application implements proportional closed-loop conversion in I / Q rectangular coordinates, combines amplitude error and phase error into compensation current update amount, and combines calibration proportionality coefficient and output amplitude limiting, so that closed-loop converges amplitude-phase deviation in linear interval quickly, improves compensation current stability and response speed and reduces energy consumption.

The invention discloses a measurement method of a fluxgate current sensor with low loss and scale amplification, which relates to the technical field of current measurement and comprises the following steps of: acquiring an observed quantity set subjected to phase locking synchronous demodulation; constructing a scheduling optimization function based on the pre-measurement set, applying a soft opening constraint, an amplitude stability constraint, a phase consistency constraint and a saturation depth constraint, and outputting a corresponding control variable; excitation signal parameters are set according to the control variables, excitation driving and compensation injection are executed, and harmonic amplitude phase response of the current period is extracted in the position of the demodulation window; and converting the harmonic amplitude-phase response into linear measurement output of the current period, and updating the observed quantity set at the same time. Proportional closed-loop conversion is implemented in I / Q rectangular coordinates, amplitude errors and phase errors are synthesized into compensation current updating quantity, and a calibration proportionality coefficient and output amplitude limiting are combined, so that a closed loop quickly converges amplitude-phase deviation in a linear interval, the stability and response speed of compensation current are improved, and energy consumption is reduced.

This application discloses a method, system, electronic device, and medium for integrated space-ground resource scheduling. The method trains a policy parameter prediction model for all agents based on a first loss function and a second loss function until a new policy parameter set is output after training. This new policy parameter set is used as the current policy parameter set in the transfer learning phase to calculate the updated policy parameter set and construct a third loss function. Based on the updated and current policy parameter sets, a stability constraint term is constructed and added to the third loss function to obtain the overall objective function. The policy parameter prediction model and the transfer learning phase are trained iteratively until the overall objective function converges to a preset threshold, yielding the target policy parameter set output by the transfer learning phase. The target policy parameter set is then deployed to the integrated space-ground network. This application enables real-time resource scheduling and improves the reliability of resource scheduling.

The invention belongs to the field of airborne power system design, and particularly relates to a stability constraint method for an MEA variable-frequency alternating-current power system. The method comprises the following steps: firstly, establishing a port impedance / admittance model of each device in a system; secondly, analyzing the stability of each subsystem, and redesigning if the stability margin does not meet the requirement; then, on the premise that all subsystems are stable, the overall stability of the system is analyzed according to a generalized Nyquist stability criterion; if the whole system is unstable or underdamped, quantitatively identifying a high-influence negative subsystem with the maximum stability deterioration degree in a mode of setting admittance of a load-side subsystem to be zero; and finally, the stability margin of the high-influence negative subsystem is improved by redesigning the high-influence negative subsystem. The MEA variable-frequency alternating-current power system is ensured to have enough stability margin in the design stage, so that the stability problem is avoided.

The application provides a deep network-based MSWI process ignition point temperature control method, which comprises the following steps: taking a key manipulated variable as an input and an ignition point temperature as an output, establishing an ignition point temperature controlled model by using a linear regression decision tree with interpretability, and simulating and predicting the ignition point temperature in an actual combustion system; calculating an error between a predicted value of the ignition point temperature and a set value, taking the error and an error change as inputs, using an LSTM controller to extract time sequence characteristics of the error information, and outputting a corrected feeding speed, so that the ignition point temperature output by the model can track the set value; meanwhile, each gate weight and a bias parameter of the LSTM controller are adaptively updated by introducing Lyapunov stability constraints, so that stable calculation of a control increment of the feeding speed is realized at each sampling time. The method can ensure stable adjustment of the ignition point temperature without an accurate system model.

A constructive certificate-based neural network multi-agent consensus control method includes: 1) constructing a distributed controller based on a residual neural network, employing parallel linear shortcut branches and nonlinear residual branches to enhance nonlinear transient regulation capability while maintaining baseline stability; 2) establishing a stability constraint mechanism based on constructive certificates, calculating the upper bound of the Lipschitz constant of the residual branches in real time during training, and dynamically adjusting the shortcut gain accordingly to ensure that the stability margin of the closed-loop system is always positive; 3) performing end-to-end training based on differentiable simulation, combining spectral pruning techniques and a comprehensive loss function to jointly optimize controller parameters. Experiments show that this invention achieves comprehensive performance comparable to the optimal linear controller without requiring manual grid search for parameters, significantly reducing control energy consumption, and providing rigorous Lyapunov stability theory guarantees throughout the training process.

The invention relates to the field of ferrous metallurgy automation control, and discloses a self-adaptive regulation and control method and system for steel production equipment based on instantaneous energy efficiency characteristic feedback. The method comprises the following steps: synchronously acquiring equipment operation state and material flow data through a millisecond-level period, and resolving an instantaneous characteristic value representing unit effective output energy consumption intensity by using an energy consumption-material coupling model; and under the limitation of the process stability constraint model, based on a comparison result of the instantaneous characteristic value and a dynamic reference threshold value, automatically generating a physical correction instruction to drive an execution mechanism to execute closed-loop adjustment. According to the method, the energy efficiency index is converted into the real-time process control variable, dynamic balance between energy efficiency optimization and process stability is achieved, and irreversible material loss is dealt with through cross-process compensation.

The application discloses a transient stability constrained optimal power flow method based on an improved quadratic convex relaxation, which comprises the following steps: step 1, based on historical data, wind power and load are respectively subjected to uncertainty modeling by adopting Weibull distribution and normal distribution to obtain a probability density function of wind power output and a probability density function of load power; step 2, a transient stability constrained optimal power flow model considering source and load double-end uncertainty is established by taking minimum expected total fuel cost as an objective function; step 3, the opportunity constraint in the CC-TSCOPF model is converted into a deterministic constraint to obtain a deterministic optimal power flow model; step 4, for non-convex items in the power flow balance equation constraint in the deterministic optimal power flow model, an improved quadratic convex relaxation method IQC is adopted for processing, and then the objective function is linearized to obtain a linear programming model; and step 5, a solver is used to solve the linear programming model to obtain an optimal generator scheduling scheme meeting the transient stability constraint.

The invention relates to the technical field of medicinal liquor production simulation, and discloses a medicinal liquor production process simulation method. The method comprises the following steps: acquiring a physical parameter set and a production process parameter set of medicinal liquor production raw materials; constructing a spatial distribution model containing a solvent diffusion path and an effective component dissolution track through multi-scale molecular dynamics simulation; extracting a core reaction area meeting specific conditions in the model, and screening out key simulation nodes; the key simulation nodes are input into a pre-constructed multiphase flow coupling simulation system, the system generates optimized process parameters through iterative solution, and concentration field uniformity constraints and phase interface stability constraints are introduced into convergence conditions; establishing a dynamic production process model in combination with optimization process parameters and time stage division of a production process parameter set, and simulating spatio-temporal evolution behaviors of medicinal liquor components based on a non-equilibrium thermodynamic equation; and separating a concentration evolution field and a phase distribution field from the model, carrying out multi-stage reconstruction on the concentration evolution field and an original solvent ratio, and outputting a final simulation result.

A transient stability constraint optimal power flow acquisition method considering wind power uncertainty, step 1: taking the power system optimal power flow solution as the initial operation state and setting the expected fault set; step 2: using a typical wind power scene to describe the uncertainty of wind power output; step 3: for a series of expected faults under each typical wind power scene, time domain simulation is carried out, and for the case that the power system is transiently unstable, the process constraint of the transient stability of the power system is converted into an algebraic form of the transient stability constraint through constraint conversion; step 4: the obtained transient stability constraint is added to the power system optimal power flow model, and the equivalent power system transient stability constraint optimal power flow (TSCOPF) model containing wind turbines is constructed, and the model solution is recalculated. The purpose of the application is to expand the power system transient stability constraint optimal power flow model solving method to adapt to the uncertainty of wind power generation and guarantee the transient stability of the power system.

The invention discloses a fusion reactor ECRH system gyrotron power prediction method and a fusion reactor ECRH system gyrotron power prediction system. The method comprises the following steps: preprocessing collected gyrotron data; performing parameter selection and weight distribution on the preprocessed data; performing data length constraint, signal stability constraint and stable area proportion constraint on the selected and weighted parameters, and screening data meeting the triple engineering constraint conditions as steady-state operation data; grading the steady-state operation data according to the stability ratio and the relative standard deviation, and selecting a data set trained by the deep learning model according to the grade sequence of grading; inputting the data set into a deep learning model, adjusting hyper-parameters of the deep learning model by using a whale optimization algorithm, training the deep learning model, and inputting gyrotron data collected in real time into the trained deep learning model to obtain predicted gyrotron power; the method has the advantage that the prediction precision and efficiency are improved.

A transient stability constraint-based optimal power flow method based on semidefinite programming and convex relaxation includes the following steps: Step 1: Establish a Gaussian mixture model (GMM) for wind power output prediction based on historical wind speed and direction data, generate wind power output scenarios using Latin hypercube sampling (LHS), and reduce the scenarios using K-means++ clustering to obtain the final scenario set; Step 2: Based on the GMM constructed in Step 1 and the final scenario set, establish a transient stability optimal power flow model (TSCOPF), obtain the trajectory sensitivity matrix using the trajectory sensitivity method (TST), and construct linear inequality constraints based on the trajectory sensitivity matrix; Step 3: Reconstruct the model including the above linear inequality constraints into a semidefinite programming form (SDP), add a penalty term based on the matrix second-order minor to the objective function, and perform penalized SDP relaxation; Step 4: Solve using the MOSEK solver to obtain feasible power flow and unit output schemes.

The invention discloses a CACC controller parameter optimization method considering communication delay and a related device, and the method comprises the steps: building a CACC model, and determining the stability constraint and communication delay boundary of the CACC model; constructing an optimization objective function according to the CACC model, and constructing a Gaussian process model based on the optimization objective function; and based on the Gaussian process model, solving optimal parameters in the stable region by using a Bayesian optimization algorithm. On the basis of a stability optimization framework of the CACC model, the Bayesian optimization algorithm is introduced innovatively, accurate optimization of parameters of the CACC controller under communication delay constraints is realized, and queue stability and disturbance suppression capability are improved.

The application discloses a second-order cone relaxation-based optimal power flow solving method containing transient stability constraints of wind power, and comprises the following steps: step 1, wind speed is predicted, wind power output scenarios are generated by combining Latin hypercube sampling (LHS), and K-means clustering is used to reduce the scenarios to obtain typical scenarios of wind power output and corresponding occurrence probabilities, so as to represent and process the uncertainty of wind power; step 2, an initial architecture of a transient stability-constrained optimal power flow (TSCOPF) model containing an objective function, static constraints and transient stability constraints is constructed; step 3, a second-order cone relaxation (SOCR) is used for non-convex items in a power balance equation in the TSCOPF model, and a successive inscribed polygon method is used to reduce a relaxation gap, then, the relaxed power balance equation is integrated with linear inequality constraints obtained by transformation in step 2 and other static constraints, and the original TSCOPF model is overall converted into a mixed integer second-order cone programming problem; and step 4, an optimal power flow solving result satisfying the transient stability constraints is obtained.

The present application relates to a model prediction and preset performance constraint based unmanned aerial vehicle trajectory tracking control method, which introduces Lyapunov stability constraint in the model prediction control framework and integrates preset performance control mechanism, thereby realizing the unification of performance constraint and system stability analysis, including the following steps: establishing a nonlinear system model based on unmanned aerial vehicle dynamics; defining position error and attitude error and constructing preset performance function, converting error with performance constraint into unconstrained error through error normalization and nonlinear transformation; establishing model prediction optimization problem under the premise of considering input saturation and stability constraint; designing auxiliary control law based on transformed error to construct stability constraint; proving that the control strategy can guarantee that the error meets preset performance constraint and system local asymptotic stability. The present application can realize stable and reliable trajectory tracking control of unmanned aerial vehicle system, and has high tracking precision and good dynamic performance.

The invention discloses a brain-like adaptive optimal control method and system based on Lyapunov stability and information geometric constraints, and belongs to the technical field of adaptive control and intelligent control. According to the method, a multi-time-scale adaptive control model containing a fast adaptive state variable and a slow adaptive state variable is constructed, and control parameters are updated online according to an error signal between system output and reference input; in a parameter updating process, Lyapunov stability constraint is introduced to ensure global convergence of a system state under a random disturbance condition, and information geometric constraint is carried out on an updating direction and an updating amplitude of a control parameter by utilizing a Fisher information matrix constructed by a partial derivative of the system output to the control parameter, so that the control parameter is updated. The identifiability and the anti-noise performance of parameter estimation are improved; and on the premise that the stability constraint is met, the control input approaches the optimal feedback control solution under the preset performance index in the steady-state stage. According to the method, stable, robust and approximately optimal adaptive control can be realized under the conditions of parameter uncertainty and random disturbance, and the method is suitable for application scenes of robot control, intelligent equipment, embedded control systems and the like.

The application provides a salt mist deposition rate model construction method, device and equipment of aerospace materials, relates to the artificial intelligence technical field, and the model construction method comprises the following steps: inputting the obtained salt mist deposition parameter into a preset initial model to obtain an initial deposition rate prediction result; obtaining a kinetic constraint result through kinetic constraint according to a growth rate coefficient, an inhibition coefficient, a maximum deposition rate and the initial deposition rate prediction result; obtaining a stability constraint result through stability constraint according to the kinetic constraint result and the number of sampling points; obtaining a deposition constraint matrix through automatic differentiation according to the kinetic constraint result and the stability constraint result; generating an update parameter according to the deposition constraint matrix and the current parameter of the initial model, updating the initial model according to the update parameter, and obtaining a salt mist deposition rate model. The application improves the accuracy of the salt mist corrosion evaluation of aerospace materials.