217 results about "Constrained optimization problem" patented technology

The disclosure relates to the technical field of power system management, and in particular to a large-scale electric vehicle cluster coordinated charging optimization method based on greedy repair genetic algorithm. the method includes the following steps: Si, establishing a task model: establishing a task model of electric vehicle cluster coordinated charging in large charging stations; S2, constructing constraint conditions: constructing constraint conditions for the coordinated charging optimization problem; S3, constructing a constrained optimization problem: designing an objective function with the minimum charging cost, performing linear mathematical transformation on non-linear constraint conditions, and constructing a 0 / 1 integer linear programming problem; and S4, solving the constrained optimization problem with a custom genetic algorithm based method: carrying out genetic coding on a decision model of the charging station, and specifically designing a genetic algorithm with greedy repair operators to solve the constructed constrained optimization problem.

The invention provides an unmanned aerial vehicle (UAV)-based flux-inductance integrated beam forming and trajectory optimization method, on one hand, an FP-AROS method is proposed based on fractional programming and semi-definite relaxation technologies in combination with approximate rank-one solution construction to carry out optimization design on ISAC beams, the Cramer-Rao bound of AoD is adopted as a sensing performance evaluation index, and the FP-AROS method is adopted to carry out optimization design on the ISAC beams; target estimation performance can be reflected more directly, so that a collaborative optimization effect of communication and sensing performance is realized; on the other hand, the invention provides an improved continuous convex approximation method based on a trust domain, and the method is used for optimization design of a UAV track. The ITRSCA method obtains a local trend approximate solution of an original objective function by calculating a second derivative, and introduces a dynamic trust domain mechanism to control approximate precision, so as to ensure global convergence of an output solution; in addition, the method is combined with a penalty function method to process non-convex constraints, and the limitation of a trust domain method in a constrained optimization problem is effectively overcome.

The invention provides a pulse passing rate improving method based on accumulation signal demodulation, and belongs to the technical field of nuclear electronics digital pulse processing. Comprising the following steps: performing baseline deduction and monopulse screening on a radiation field pulse signal; constructing a nuclear pulse signal model, fitting to obtain model parameters, and constructing a system response matrix; modeling a nuclear detection physical process into a mathematical model; solving a constraint optimization problem by using a non-negative least square method; and demodulating and calculating all the pulse signals including the stacked pulses which are discarded by stacking, so as to obtain delta pulses containing energy information. According to the invention, the accumulation signal which is abandoned by a traditional accumulation abandoning algorithm is demodulated, so that the technical problem of increasing the pulse passing rate is solved, and meanwhile, heavy peaks and high-energy trailing in an energy spectrum are inhibited. Compared with an existing accumulation discarding method, the pulse passing rate effect is remarkably improved on the premise that the energy spectrum accuracy is guaranteed.

The invention discloses a high-low cycle composite fatigue life prediction method based on a particle swarm optimization physical information neural network, which aims at high-low cycle composite fatigue life prediction of turbine blades under small sample data characteristics, and comprises the following steps of: introducing a classical physical formula of a linear cumulative damage theory into a loss function as a constraint term; training fatigue test data by adopting a neural network model; deducing a constraint space of model parameters according to domain knowledge of high-low cycle composite fatigue life prediction, and converting training of a neural network into a constrained optimization problem; a particle swarm optimization algorithm is adopted to optimize the weight and the threshold value of the neural network, and the learning ability of the neural network under a small sample data set is further improved. The method provided by the invention can predict the fatigue life of the turbine component under the high-low cycle combined load, has high prediction precision under a small sample data set, and provides theoretical support for fatigue life prediction and reliability analysis of the engine turbine component.

The invention relates to the technical field of wireless communication, and discloses an unloading method in an ultra-dense millimeter wave MEC network. Aiming at the problems of insufficient terminal capability, high energy consumption of an ultra-dense base station, millimeter wave coverage limitation, communication security risk and the like under the condition of sharp increase of a calculation-intensive task, the method comprises the following steps: firstly, obtaining network basic information, constructing a network architecture containing communication, calculation unloading and a security model, and establishing a multi-constraint optimization problem; based on the optimization problem, initializing a multi-strategy black-wing plinuary optimization algorithm population through logic mapping and elite selection; updating individual positions through global and local search in algorithm iteration, and screening historical optimal individuals; and finally, configuring and unloading resources according to the configuration. According to the method, the NOMA technology, a multi-step unloading framework and a hybrid communication mode are combined, the optimization efficiency is improved through a multi-strategy black-wing optimization algorithm, the local energy consumption is reduced, the communication rate is improved, the safety is guaranteed, the optimal scheme meeting time delay and safety constraints is rapidly converged, and the user experience is improved.

The invention discloses a graph calculation-driven complex publication derivative resource content layout planning system, which belongs to the technical field of digital publication and intelligent media, and comprises a multi-modal data analysis module for analyzing heterogeneous data such as texts, images, tables and the like in publications into structured node features; a feature fusion module; a content relationship modeling module; a graph calculation driving layout optimization module; performing layout constraint modeling: converting a layout problem into a graph constraint optimization problem; a modeling optimization module; a multi-terminal adaptive rendering module; the layout generation engine is used for converting a graph calculation result into a visual layout scheme; and the interactive feedback module is used for feeding back user operation to the graph in real time to trigger layout reconstruction. According to the method, advanced technologies such as graph calculation, multi-modal fusion and constraint solution are deeply fused, the limitation that a traditional typesetting tool depends on a fixed template is broken through, and a content-driven one-stop solution is provided for scenes such as academic periodicals, electronic textbooks and digital reports.

A vehicle is controlled for travel on a road having a geometric design defined by one or a combination of a docking line, a cross-section, and a cross-section of the road such that different values of parameters of the geometric design of the road, traffic on the road, traffic rules for traffic flow on the road define different traffic scenarios. By relaxing configuration parameters of a real-world scene and tightening corresponding limiting parameters, a mixed integer non-convex constraint optimization problem for a current real-world traffic scene is converted into a mixed integer convex optimization problem for approximate representation of the real-world traffic scene, so that a carrier is controlled. A mixed integer convex optimization problem for the converted approximate representation of the real-world traffic scene is solved to generate current control commands for controlling one or more actuators of the vehicle.

The invention discloses an electric vehicle lifting torque control method and system. The method comprises the steps that driving operation, vehicle state and environment data are collected in real time and preprocessed into multi-dimensional real-time feature vectors; based on the vector, extracting and updating a driving style feature vector by using an incremental clustering and lightweight time sequence convolutional network, and further mapping to generate a personalized cost function weight coefficient in model prediction control; synchronously operating the vehicle dynamics and thermodynamics simplified model, predicting a dynamic physical constraint boundary of a driving system in a future time domain, and constructing a dynamic feasible domain curve; taking a driver instant demand as a tracking target, fusing a personalized weight and a dynamic feasible region curve, constructing and solving a finite time domain constraint optimization problem, and generating a current period target torque instruction; and filter parameters are adaptively adjusted through inverse dynamic model feedforward compensation in combination with a dynamic feasible region, a final torque control instruction is generated, and unification of safety, smoothness and individuation of torque control is achieved.

The invention provides a delay compensation control method for a double-vibrator wave energy conversion device, which comprises the following steps of: performing numerical modeling and hydrodynamic analysis on the double-vibrator wave energy conversion device, establishing a time domain motion equation of the double-vibrator wave energy conversion device, and introducing a time delay function into the time domain motion equation to simulate control signal transmission delay; simulating the execution delay of the brake by using a partial differential equation; establishing a state-space equation of the device, replacing a convolution term of the time-domain motion equation with the state-space equation, and calculating to obtain a motion state of the device; a Hamiltonian function is defined to convert a constrained optimization problem into an unconstrained optimization problem, and the Hamiltonian function is solved to obtain an optimal control criterion considering control delay so as to realize maximization of energy capture under the optimal control criterion. According to the method, the operation characteristics of a physical system are truly reflected by introducing control delay, so that the locking control method is effectively implemented in an actual physical device, and the energy capture efficiency of a wave energy conversion device and the reliability of system operation are effectively improved.

The invention relates to an unmanned aerial vehicle cluster target monitoring path planning method based on space-time constraint multi-agent reinforcement learning, and belongs to the field of unmanned aerial vehicle path planning. Comprising the following steps: acquiring an unmanned aerial vehicle cluster and environment characteristic information, establishing time and space constraint conditions, and constructing a constraint optimization problem taking minimization of total energy consumption as a target; based on the obtained feature information, mapping a constraint optimization problem into an Actor network and a Critic network corresponding to each unmanned aerial vehicle and a Critic network corresponding to an unmanned aerial vehicle cluster; carrying out reverse training on the Actor network and the Critic network based on a simulation result of the unmanned aerial vehicle cluster; and inputting the feature information at the current moment into the Actor network corresponding to each unmanned aerial vehicle for path planning. According to the method, the situation that actions of unmanned aerial vehicles at the current moment meet time / space constraint conditions is evaluated based on time / space feature information, energy consumption of an unmanned aerial vehicle cluster at the current moment is evaluated based on global state information, and therefore it is ensured that optimal path planning meeting various constraint conditions and achieving a target is found.

The invention discloses a quadruped robot landing planning method based on scene decoupling and risk avoidance. The quadruped robot landing planning method aims at solving the problems that in the prior art, perception is not precise in a complex environment, and the obstacle avoidance capacity is insufficient. The method comprises the following steps: acquiring and fusing multi-source sensor data, and segmenting an original point cloud; generating a scene decoupling elevation map of a multi-layer structure by using the segmentation point cloud; constructing a multi-objective optimization problem taking risk avoidance as a core based on the elevation map, wherein the multi-objective optimization problem comprises a comprehensive cost function and a strict obstacle avoidance constraint; utilizing a reaction formula adjusting module of a capturable region theory to cope with a dynamic instability risk; and solving the multi-target and multi-constraint optimization problem in real time by adopting a hierarchical solving strategy, and finally generating an optimal foot end drop point considering safety and stability. According to the method, the terrain adaptability, the motion stability and the decision intelligence of the quadruped robot in an unstructured environment are remarkably improved through fine decoupling of a scene and quantitative avoidance of multi-source risks.

The invention discloses a tractor trailer robot track generation method based on a polynomial and a symbolic distance field, and the method comprises the steps: dividing grids through a point cloud map, estimating the curvature and normal direction of each grid, extracting obstacle information, and calculating a symbolic distance value corresponding to each grid; taking the initial pose of the tractor trailer robot as a root node, and based on the obstacle information, utilizing a multi-end-point heuristic graph search algorithm for spatial search in SE (2) to generate an initial path for arriving at the target area; and in combination with the symbol distance value corresponding to each grid, modeling the trajectory generation of the tractor trailer robot as a constraint optimization problem, and solving the constraint optimization problem by taking the initial path as an initial value to obtain a final trajectory. The problem that track generation is difficult due to complex kinematics, high state space dimension and deformable structure of the traction trailer robot is solved, the problems that an existing algorithm is low in efficiency and loses solution space are solved, and efficient track generation of the traction trailer robot is achieved.

A method of performing computation in a hybrid quantum-classical computing system includes computing an approximate cost function of an optimization problem with variables constrained by an inequality, wherein the inequality constraint is included using a polynomial approximation of a Heaviside step function, mapping the approximate cost function of the optimization problem to a model Hamiltonian, setting a quantum processor in an initial state, executing one or more iterations, each iteration including applying a parametrized quantum circuit to the quantum processor based on a set of variational parameters and the model Hamiltonian, measuring an expectation value of the model Hamiltonian, and replacing the set of the variational parameters with another set of variational parameters, and outputting the set of the variational parameters after executing the one or more iterations.

The invention relates to the technical field of enterprise management, in particular to an enterprise process custom configuration method and system based on a code-free technology, and the method comprises the steps: constructing a multi-layer directed hypergraph containing data, control and condition dependence, calculating an edge weight, and converting the dependence into a first-order logic constraint set; performing incremental satisfiability detection, and outputting a conflict core set containing conflict types and severity in combination with minimum unsatisfiability subset mining and strong connected component analysis; mapping the conflicts into modifiable elements, and generating parameterized candidate repair schemes based on a repair operation library and historical cases; converting the candidate scheme into a weighted constraint optimization problem, and solving an optimal scheme by adopting a branch and bound algorithm; influence propagation verification is completed in the copy, a differential patch and a rollback identifier are generated, transactional submission is carried out after passing, and the weight and the case library are updated. According to the method, automatic diagnosis and low-risk repair of implicit dependency conflicts are realized.

The application discloses a target direction finding method based on a short baseline unified model, which is realized by a computer and comprises the following steps: firstly, an improved polar coordinate representation (MPR) unified model is constructed; secondly, a closed-form initial solution of direction finding under the MPR model is obtained; then, an improved successive unconstrained minimization method based on the MPR is used to solve a quadratic constraint optimization problem to obtain the angle of the target source; thirdly, a direction finding initial solution is used to guide source deployment; fourthly, time-space reference system error calculation is performed; finally, target source TDOA measurement and time-space reference system error elimination and final solution solving are performed; the application effectively avoids the threshold effect of the short baseline positioning system based on the far-field direction finding unified model of the modified polar coordinates, so that the far-field target direction finding method of the unmanned aerial vehicle group based on the time difference of arrival measurement becomes possible; the calibration source and the target source are jointly monitored, the time-space reference system error is eliminated through the time difference measurement of the calibration source, and thus the direction finding precision is improved.

The invention relates to the technical field of regional sustainability, and discloses a regional sustainable development mode display system based on virtual simulation, and the system comprises a data collection module which is used for collecting related resource data of regional sustainability in real time; the multi-dimensional modeling module is used for constructing an agricultural production model, a social economic system model and an ecological environment model based on the resource data; the optimization decision-making module is used for generating a multi-objective and multi-constraint optimization problem based on the agricultural production model, the social economic system model and the ecological environment model and performing resource allocation optimization, and the optimization problem comprises multiple objectives of economic benefit, ecological protection and social benefit; the invention also provides a regional sustainable development mode display method based on virtual simulation. The method comprises the following steps: collecting regional sustainable resource data in real time; by adopting a real-time data acquisition and feedback control scheme based on the Internet of Things technology, precise monitoring and dynamic adjustment of regional sustainable resources are achieved.

The embodiment of the invention relates to the technical field of radars, and discloses an MIMO radar emission sequence set design optimization method, which comprises the following steps: constructing a signal model designed based on an AF emission sequence set for an MIMO radar system, selecting ISL of minimizing local AAF and CAF as an optimization criterion, selecting an energy constraint and a dynamic range constraint as constraint conditions, and selecting an energy constraint and a dynamic range constraint as constraint conditions; establishing a four-order non-convex constraint optimization problem; decomposing the fourth-order non-convex constraint optimization problem into a plurality of sub-problems which can be iteratively solved by utilizing an MBI algorithm; sequentially solving the sub-problems by means of CVX to obtain a preliminary solution; and continuously adjusting a search direction and a step length by using a BLS algorithm, and carrying out iterative optimization on the approximate solution until an Armijo condition is met, thereby obtaining a global optimal solution meeting a design requirement, and further designing a radar emission sequence set with good related characteristics, thereby improving the target detection capability, the anti-interference capability and the resolution of the MIMO radar system.

The invention discloses a zero-order primal dual method and device for a black-box constraint optimization problem, and the method comprises the steps: constructing a corresponding black-box constraint optimization problem based on a preset constraint optimization scene, and determining a tightly convex feasible region containing an original variable; converting the black box constraint optimization problem into an unconstrained optimization sub-problem which can be solved iteratively; on the basis of a function value query result of the black box constraint optimization problem, gradient correlation information of the unconstrained optimization sub-problem about an original variable is estimated, iterative solution is carried out on the unconstrained optimization sub-problem on the basis of the gradient correlation information, and the original variable after iterative updating is limited in the tightly convex feasible region; and when a preset convergence condition is met, obtaining a final original variable as an optimization result. A black box constraint optimization problem is converted into an unconstrained sub-problem which can be solved iteratively by introducing a near-end Lagrangian fusion optimization framework, so that efficient optimization is realized under the black box condition that only function values can be obtained.

The invention relates to the field of intelligent scheduling allocation, in particular to an edge cloud computing resource allocation optimization method based on deep learning, which adopts a space-time prediction algorithm based on multi-head attention and gating mechanism optimization to design time coding and space coding. The spatial relationship and interaction between time sequence characteristics of the computing power load and edge server nodes are captured, and meanwhile, a multi-head attention mechanism and expansion causal convolution are combined, so that instantaneous computing power load fluctuation can be captured, and the long-term trend of the computing power load can be mined; therefore, a reliable basis is provided for subsequent computing power scheduling by predicting an accurate computing power load. The invention designs an alternating direction multiplier method based on genetic algorithm optimization, which is not only suitable for a nonlinear and multi-constraint optimization problem, but also can be expanded to a larger-scale distributed edge node cloud computing system, and meanwhile, a global optimal solution is quickly approached through the genetic algorithm, so that the quality of an initial solution is improved, and model convergence is accelerated; and the distributed collaborative allocation scheduling efficiency is improved.

A vehicle includes an electric powered propulsion system. An electric energy storage system is electrically connected to the electric propulsion system and is configured to have an electrical energy storage component and a controller. A charging port is connected to the electric energy storage system and configured to connect to an external power source. The controller includes a memory and a processor. The memory stores instructions for causing the processor to optimize a charging profile based on a plurality of received parameters using a multi-objective constrained optimization problem. The received parameters include a power type of a connected external power source, and at least one of a requested ready to depart time, a targeted state of charge, and an effective range.

The invention provides a new energy power system optimization regulation and control system and method based on a large language model, and belongs to the technical field of power system operation control. In the system, a new energy power system scheduling model knowledge base stores an objective function and a constraint condition of a new energy power system scheduling problem model; the information extraction agent generates a structured scheduling demand according to an input scheduling instruction based on a natural language and sends the structured scheduling demand to the problem modeling agent; the problem modeling agent calls an objective function and a constraint condition corresponding to the scheduling demand from the new energy power system scheduling model knowledge base to construct a corresponding constraint optimization problem; and the code writing agent converts the constraint optimization problem into an executable code. According to the method, the scheduling instruction based on the natural language can be automatically converted into the optimization code to be output, the optimization result can be obtained after the code is operated, automatic modeling and solving of the new energy power system scheduling problem are achieved, and therefore the threshold of new energy power system scheduling is greatly reduced.

The invention provides an active power distribution network fault self-healing cooperative control method, and belongs to the technical field of power grids, and the method comprises the steps: collecting the multi-source operation data of a power distribution network in real time, and enabling a model to output a prediction sequence of a power grid state in a period of time in the future through the fusion of a power grid topological relation and spatial-temporal dynamic characteristics; inputting the prediction sequence into a collaborative optimization controller; the collaborative optimization controller takes the space-time diagram convolutional network model as an internal dynamic prediction model, and generates an optimal switch action control sequence by solving a constrained optimization problem; and executing a first-step control instruction in the control sequence. The method has the advantages that the space-time correlation between the power grid topology and the dynamic operation data is captured by using the graph structure model, the accuracy of system state prediction under the new energy output fluctuation condition is improved, and more reliable look-ahead information is provided for subsequent optimization decision making.

The invention discloses a path planning method based on a quadratic unconstrained binary optimization model, which belongs to the technical field of path planning, is used for unmanned system navigation, and comprises the following steps: rasterizing an environment area, determining a starting point, an ending point and the positions of barrier grids, determining a set of the grids where barriers are located, and defining grid binary variables; defining a target function, and establishing an optimization problem of the target function under the conditions of adjacent grid constraints, in-out constraints, obstacle avoidance constraints and starting point and terminal point position constraints; converting the constrained optimization problem into a quadratic unconstrained binary optimization model, solving the quadratic unconstrained binary optimization model, determining the value of each grid variable according to the value of the independent variable when the target function takes the minimum value, and finally obtaining a planned path. According to the method, multiple complex constraint conditions are directly embedded into the target function through the penalty term, the complex constraint conditions do not need to be independently processed, and the logic of the solving process is simplified.

The invention discloses a safety reinforcement learning control method and device for a maglev train suspension system and a medium, and relates to the technical field of maglev control, and the method comprises the following steps: S1, constructing a suspension system dynamic model as a training environment; s2, modeling a control problem of the suspension system as a constraint optimization problem; s3, converting a constrained optimization problem into an unconstrained optimization problem, and performing iterative solution on the target function; s4, constructing a cost function, accelerating solution of an unconstrained optimization problem, and designing a penalty term to guide a system state to be away from a security boundary; and S5, in a reinforcement learning environment, combining a cost function, and carrying out iterative training solution on the unconstrained optimization problem. The method does not depend on an accurate suspension system model, an independent control strategy does not need to be designed for each specific challenge, the optimal control strategy can be adaptively learned through interaction with the suspension system, and the method has high robustness and flexibility.

The invention relates to the technical field of water treatment, and discloses an electrochemical descaling automatic control system based on autonomous prediction of a pole reversal time sequence and an operation mode, the system comprises a sensing layer used for collecting raw water quality parameters and system state parameters in real time, and the system state parameters comprise real-time scale thickness and real-time scale thickness; the prediction layer is connected with the sensing layer, the prediction layer is connected with the sensing layer, the acquired data are utilized, the future scale thickness and the descaling efficiency under different control strategies are predicted through an AI prediction model, and the decision-making layer is connected with the prediction layer and used for solving an optimization problem with constraints according to a prediction result, and an optimal reverse-pole power supply control strategy is generated. The system can sense the dynamic change of the water quality of industrial raw water and the equipment state in real time, dynamically predict the scaling trend through the AI model and generate an optimally matched inverse pole strategy, so that the system can automatically adjust the equipment descaling state to the optimal working point according to the water quality fluctuation, efficient descaling is realized, and the effect is ensured to be continuous and stable.

The invention relates to an inertial parameter error-oriented optimal balancing mass gradient prediction method, which comprises the following steps of: based on balancing mass minimization, establishing and solving a constraint optimization problem of optimal balancing of dynamic and static imbalance of a satellite, and obtaining optimal balancing mass and position meeting an existing counterweight mounting position set of the satellite; analyzing a gradient matrix of the optimal balancing quality relative to the dynamic and static unbalance of the satellite; establishing a transformation matrix from the local coordinate system of each part of the satellite to the overall coordinate system of the satellite, and further calculating the gradient value of the dynamic and static unbalance of the satellite relative to the inertial parameters of each equipment or part; and in combination with the balancing quality optimal solution, the gradient matrix, the transformation matrix and the gradient value, analyzing the influence of the inertial parameter error of each device or part on the optimal balancing quality.

The application discloses a backdoor attack method based on color frequency injection and adaptive local enhancement, and relates to the technical field of machine learning and artificial intelligence security. The method comprises the following steps: introducing low-frequency color offset and weak high-frequency signal into an image in a CIELAB color space to perform global color-frequency injection; using a pre-trained proxy model to locate a high-sensitive perception domain of the model through mixed evaluation of gradients and class activation maps, and generating a binary mask; in an HSV color space, respectively applying nonlinear stretching factors to saturation and brightness of the sensitive domain based on the mask to perform adaptive local enhancement; using Gaussian smoothing, adaptive noise and histogram matching to eliminate edges and statistical abnormalities caused by local enhancement, completing compensation color enhancement to generate a poisoned image; and modeling a trigger core parameter as a constrained optimization problem, and using a particle swarm optimization algorithm to jointly dynamically update the trigger core parameter to obtain an optimal strategy. The application anchors the trigger feature depth in the core semantic area of the model and lurks in the normal data manifold, guarantees a high attack success rate, realizes extreme visual and feature concealment, and has strong anti-defense robustness.

Methods and systems are disclosed for automatically integrating subsurface structural maps with strike and dip information measured in subsurface wells. The method includes obtaining a seismic image volume for a subsurface region of interest and a well log for each of a plurality of wellbores penetrating the subsurface region of interest. Further, the method includes determining a seismic map of a geological surface from the seismic image volume, wherein the seismic map comprises an estimated depth and an estimated vector normal to the seismic map at a plurality of horizontal locations and determining an intersection point for each of the plurality of wellbores with the geological surface. Additionally, the method includes forming a cost function based, at least in part, on the seismic map and the intersection points of the plurality of wellbores and constructing a subsurface map by solving a constrained optimization problem based on the cost function.

Traditional algorithms for solving constrained optimization problems are complicated to implement, difficult to interpret, and require significant computational resources. Disclosed embodiments convert constrained optimization problems into parametric optimization problems, in which at least a subset of the constraints are converted into parametric quadratic penalty (PQP) terms that each depends on a translational parameter. The parametric optimization problem may be used for optimization in a power system (e.g., for optimal power flow, economic dispatch, etc.). When solving the parametric optimization problem, the translational parameters are updated to ensure convergence. The parametric optimization problem can be solved with reduced computational expense, using only a linear equation solver to solve a sequence of primal variables only, thereby reducing computational complexity and expense. In addition, the disclosed embodiments provide a means to incorporate constraints into machine-learning algorithms. The disclosed algorithmic framework also provides interpretability and insights for analysis.

This application relates to the field of rail transit technology and provides a method and system for deploying metamaterial tags in a railway scenario. The method includes: constructing an observation equation with the Euclidean distance between the train and the metamaterial tag as the observation distance, and constructing observation vectors for N metamaterial tags based on the observation equation; constructing a Fisher information matrix about the motion state vector based on the observation vectors and the train's motion state vector, and using the trace of the inverse of the Fisher information matrix as the Cramer-Rhodes lower bound; using the spatial coordinates of the N metamaterial tags as optimization variables, minimizing the Cramer-Rhodes lower bound as the objective, and introducing dynamic environmental constraints to form a constrained optimization problem; solving the optimization problem using a particle swarm optimization algorithm to obtain a set of target tag position coordinates; and deploying metamaterial tags at corresponding positions along the track based on the target tag position coordinates. This improves motion compensation accuracy.