96 results about "Constrained optimization problem" patented technology

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 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.

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 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.

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.

The application provides a hybrid control method for a six-bar tensegrity robot for Mars surface exploration, which firstly models the six-bar tensegrity robot used, defines the coordinates and adjacency relationship of each face, and pre-establishes an energy landscape graph. Then in the process of controlling the six-bar tensegrity robot, the environment data is synchronously collected, and the rolling or jumping mode is switched according to the robot state and the environment data. In the rolling mode, the A algorithm is used to search for an energy-optimal face conversion sequence, and based on the adaptive thrust of the quadratic programming, the optimal motor thrust distribution is obtained by solving a constrained optimization problem for each face conversion in the face conversion sequence. In the jumping mode, a safe margin height is reserved, the size and angle of the motor thrust are adjusted to overcome the Mars gravity, and the terrain adaptability of the six-bar tensegrity robot is improved. The combination of the two solves the problems of low marching efficiency, high energy loss and weak maneuverability of the prior art on the complex surface of Mars.

The application discloses a preset time control multi-electron gun cooperative scanning optimization method and system, relates to the technical field of industrial process control and optimization, and comprises the following steps: S1, a dynamic coupling model is constructed, and multi-electron gun system parameters and a communication topology are initialized; S2, a leadership comprehensive score is dynamically calculated, and a main leader and an auxiliary leader are elected; S3, the main leader solves a local constraint optimization problem, and a global energy reference trajectory is generated; S4, each electron gun independently generates a control instruction through a respective preset time distributed optimal controller based on the global energy reference trajectory, and converges a tracking error; S5, the control instruction of each electron gun is subjected to constraint processing, and is executed; and S6, system performance is monitored, and adaptive adjustment is performed; the preset time control multi-electron gun cooperative scanning optimization method and system provided by the application solve the problems of serious interference between electron guns and poor anti-disturbance robustness in the prior art.

The invention discloses a low-altitude intelligent network-oriented multi-stage security task unloading optimization method, and belongs to the technical field of low-altitude intelligent network edge computing. The method aims at solving the problems that exploration and development of a traditional task unloading method in a low-altitude dynamic heterogeneous network environment are unbalanced, optimization precision is insufficient, and safety guarantee is insufficient. The method comprises the following steps: constructing a network architecture containing communication, calculation and security models based on basic information of low-altitude intelligent network equipment; configuring a multi-constraint optimization problem; after the initial population is generated, searching a globally optimal solution by adopting an improved escape optimization algorithm; and executing calculation energy consumption optimization configuration according to the optimal solution. According to the method, algorithm innovation and security constraint design are combined, task unloading efficiency, resource allocation rationality and data transmission security are cooperatively improved, and dynamic heterogeneous features of the low-altitude intelligent network are adapted.

The application discloses a signal strength and sidelobe suppression combined optimization elastic Doppler complementary waveform design method, first proposes a Pareto effective framework for Golay complementary waveform design, and the framework jointly optimizes a Doppler elastic transceiving sequence pair. The framework considers a constrained optimization problem in terms of SMR and SNR, constructs a loss function of a Pareto multi-objective optimization problem by using an epsilon-constraint method, and obtains all possible Pareto optimal solutions. Secondly, in order to solve the optimization problem, a model-driven machine learning algorithm is innovatively designed to process the multi-objective optimization problem. The application effectively reduces the interference of sidelobes on weak target detection, improves the accuracy and reliability of target detection of a radar system, can more clearly detect and distinguish multiple targets, and enhances the overall performance of the radar system.

The application provides a delay compensation control method for a double pendulum wave energy conversion device, which comprises the following steps: numerical modeling and hydrodynamic analysis are performed on the double pendulum wave energy conversion device, time domain motion equations are established, a time delay function is introduced into the time domain motion equations to simulate control signal transmission delay, and partial differential equations are used to simulate the execution delay of the brake; state space equations of the device are established, the convolution items in the time domain motion equations are replaced by the state space equations, and the motion state of the device is calculated; a Hamilton function is defined to convert a constrained optimization problem into an unconstrained optimization problem, the optimal control criterion considering the control delay is obtained by solving the Hamilton function, and the maximization of energy capture under the optimal control criterion is realized. The application truly reflects the operation characteristics of the physical system by introducing the control delay, so that the closed-loop control method is effectively implemented in the actual physical device, and the energy capture efficiency of the wave energy conversion device and the reliability of the system operation are effectively improved.

The embodiment of the invention provides a 5G-Advanced sensing communication collaborative resource scheduling method and system for a smart substation, and belongs to the technical field of power system operation and maintenance. The method comprises the steps that a 5G-Advanced cellular-free unmanned aerial vehicle communication and sensing integrated network architecture is constructed, the network architecture comprises a passive sensing layer, an air-ground cooperative access layer and a centralized processing layer, and deep integration of sensing, energy supply and communication functions is achieved; establishing a system model based on the network architecture, wherein the system model comprises a network topology model, a communication and inductance integrated transmission model and a constraint optimization problem model; a cross-CPU dynamic resource scheduling algorithm based on MADDPG is designed, and joint optimization of user association, power distribution and load balancing is achieved through a centralized training-distributed execution mechanism; based on a scheduling algorithm and a system model, substation global real-time sensing, data transmission and operation and maintenance performance collaborative guarantee are achieved. The sensing fusion and coverage capability is greatly improved, the dynamic scene adaptability is better, and the sensing-operation and maintenance collaboration value is higher.

This application provides a proxy-assisted method, apparatus, and computer equipment for reservoir production optimization, relating to the field of reservoir production optimization technology. The method includes: constructing a proxy optimization model under preset constraints, using production parameters of a target work area as decision variables and a first function as the objective function. The first function quantifies the oil and water production state of the target work area under the production parameters; obtaining the optimal solution to the optimization problem of the proxy optimization model to output the optimal production parameters. The proxy optimization model is a model that uses a machine learning-based proxy model to accelerate the optimization of a reservoir numerical simulator. Potential candidate points used to update the proxy model during the optimal solution search are generated based on a random probability perturbation method. Generating potential candidate points based on a random probability perturbation method simplifies the optimization process and improves the efficiency of solving time-consuming bounded constraint optimization problems when performing reservoir production optimization based on a proxy optimization model, thereby improving the efficiency of reservoir production optimization.

The application discloses an LSTM-MPC collaborative closed-loop multivariable constraint optimal control method based on a green ammonia production system, and relates to the related technical field of multivariable collaborative optimal control of a green ammonia preparation process. The scheme adopts LSTM to perform multistep prediction on an ammonia generation rate, and utilizes process feedback to perform online correction on prediction errors. Accordingly, an MPC solves a constrained optimization problem in a rolling time domain, and generates optimal control instructions acting on a feed valve opening degree and electric heater power. Model parameters are recursively updated based on MPC execution results and process measurement data, a closed-loop collaborative mechanism of prediction optimization-correction is constructed, accurate tracking of a set value by the ammonia generation rate and suppression of overshoot are realized, and the control quality of the green ammonia preparation process is improved.

The application discloses a kind of accurate target positioning method and system under non-line-of-sight environment.Method includes: the coordinate of N anchor nodes in positioning scene and a target node coordinate are substituted into distance formula, obtain the distance expression for representing target node and each anchor node between, carry out square expansion, obtain overdetermined equation and the intermediate variable matrix b that represents target node and the position information and positioning error information of each anchor node, overdetermined equation solving problem is converted into fractional programming constraint optimization problem, by introducing expected value to the optimization problem, obtain fractional programming constraint optimization problem variant;The optimization problem variant is converted into parameter programming constraint optimization problem, and matrix b is used to solve parameter programming constraint optimization problem using Lagrange multiplier method, and the optimal position coordinate is obtained.The application has low computational complexity and can effectively eliminate the influence of non-line-of-sight error in TOA measurement value.

The invention relates to the technical field of wireless communication secure transmission, and provides a dynamic secure transmission strategy optimization method and device based on RIS-UAV. According to the method, the gain change caused by the directivity of the RIS antenna is dynamically compensated by introducing the three-dimensional attitude angle of the RIS, and accurate control on a wireless channel is realized, so that the leakage of privacy signals to an eavesdropper is inhibited while the shielding obstacle is overcome. A constrained optimization problem is converted into an unconstrained double-layer optimization problem through a Lagrange duality method, so that the learning complexity of the intelligent agent is reduced; a post-decision state mechanism is introduced into problem solving, so that the intelligent agent can quickly learn and adapt to the change of user channel state information in the dynamic wireless communication system; the fuzzy logic controller can increase exploration when the environment uncertainty is high and enhance the utilization when the strategy is stable, so that more intelligent and efficient exploration-utilization balance is realized, the convergence speed is remarkably accelerated, and the final performance is improved.

The invention discloses an optimization method, system and device combining antenna selection and precoding, and a storage medium, and belongs to the technical field of wireless communication. The method comprises the following steps: acquiring communication data from a base station to a user in a target XL-MIMO communication system; calculating a received signal of the user based on the communication data; based on the received signal, constructing a joint optimization model which takes the maximum sum rate as an optimization target and takes an antenna selection matrix and a hybrid precoding matrix at a base station as optimization variables; decomposing a solving problem of the joint optimization model into sub-problems by adopting an alternating direction method; alternately solving the sub-problems to obtain an optimization result; and obtaining an antenna selection matrix and a hybrid precoding matrix at the base station according to an optimization result. According to the method, the antenna selection problem is converted from integer programming to matrix sparsity optimization by adopting the alternating direction method, the constraint optimization problem of key variables is effectively solved, and the spectrum efficiency of a communication system is improved.

The application relates to the field of intelligent scheduling and allocation, in particular to an edge cloud computing resource allocation optimization method based on deep learning. The application adopts a time-space prediction algorithm based on multi-head attention and gate mechanism optimization, designs time encoding and space encoding, captures the time sequence characteristics of computing power load and the spatial relationship and interaction between edge server nodes, simultaneously combines the multi-head attention mechanism with the expansion causal convolution, can capture the instantaneous computing power load fluctuation, can mine the long-term trend of the computing power load, and thus predicts the accurate computing power load to provide a reliable basis for subsequent computing power scheduling. The application designs an alternating direction multiplier method based on genetic algorithm optimization, is suitable for nonlinear and multi-constraint optimization problems, can be extended to a larger-scale distributed edge node cloud computing system, simultaneously rapidly approaches the global optimal solution through the genetic algorithm, improves the initial solution quality, accelerates model convergence, and improves the distributed collaborative allocation and scheduling efficiency.

The invention relates to the technical field of power distribution network intelligent control, in particular to a power distribution network fault self-healing distributed resource collaborative reconstruction method, system and device and a storage medium. The method comprises the following steps: collecting multi-source fault information of a power distribution network, constructing a fault information set, determining a fault section based on the fault information set through confidence-weighted information consistency verification, and generating a hierarchical isolation strategy to minimize a power failure range of a non-fault area; generating a fault isolation operation sequence according to the position of the fault section, and evaluating the photovoltaic output capacity, the energy storage charge and discharge capacity and the connection line transmission capacity; establishing a coupling relationship between network topology adjustment and distributed resource output adjustment, and determining a switching operation sequence and a resource scheduling scheme by solving a multi-constraint optimization problem; and calculating a dynamic importance score according to the load type, the current power utilization state and the social influence factor, sorting the loads according to the dynamic importance score, sequentially accessing the loads, and verifying the voltage and the line load rate.

The invention discloses an elastic cooperative control method and system for multiple air-ground cross-domain robots under a replay attack. The method comprises the following steps: firstly, establishing a Newton-Euler kinetic model of the air-ground cross-domain robots; based on the pilot following model, respectively constructing local collaborative error dynamic models for the ground mode and the air mode; in order to resist a replay attack, designing a timestamp composite error vector fused with a historical state, and establishing an incremental state iteration prediction relation of the timestamp composite error vector; based on the prediction model, constructing and solving a constraint optimization problem taking minimization of a collaborative error and control input change as targets, and generating an optimal control increment in real time; and finally, synthesizing a final lift and torque control instruction by solving a lift optimization problem meeting the ground contact force constraint and combining attitude kinematics inverse solution and robust tracking control. Replay attacks are effectively identified and suppressed, formation stability under cross-domain mode switching is controlled and guaranteed, and cooperative elasticity and task reliability of the unmanned cluster in a antagonistic environment are remarkably improved.

The invention discloses a low-altitude intelligent car networking energy efficiency and safety rate joint optimization method and system, and belongs to the technical field of car networking communication. The method comprises the following steps: constructing an air-ground integrated cooperation model comprising a far-end base station, an unmanned aerial vehicle relay and multiple vehicles, depicting fast time-varying characteristics based on a geometric random channel model, and constructing a secure communication model according to the Shannon confidentiality theorem, a multi-constraint optimization problem with the sum of the minimum total energy consumption of the unmanned aerial vehicle and the maximum vehicle safety rate as the target is formed; designing an alternative optimization solution framework, and decoupling an original problem into three sub-problems of base station transmitting power control, unmanned aerial vehicle transmitting power control and unmanned aerial vehicle track optimization; and in each iteration, other variables are fixed, the three sub-problems are solved by adopting a Newton iteration method, a model prediction control method, a targeting method and a gradient descent method in sequence, and the corresponding variables are updated until convergence. According to the method, the flight path and the communication power of the unmanned aerial vehicle are jointly optimized, so that the problem of cooperative improvement of safety communication and system energy efficiency under a fast time-varying channel is effectively solved.

This invention discloses a convex clustering analysis method and system for assembly error propagation networks. The method includes the following steps: S1, acquiring the assembly error propagation network data to be clustered; S2, constructing a convex optimization problem based on the network data acquired in step S1, and transforming it into an equality-constrained optimization problem; S3, solving the equality-constrained optimization problem using the alternating direction multiplier method, iterating repeatedly until a preset iteration termination condition is met, and obtaining the cluster centers corresponding to each network data; S4, completing the clustering of the network data according to the distance between the cluster centers, and outputting the clustering results. This invention applies the convex clustering framework to the clustering problem of network sets, which can effectively handle graph-structured data and overcome the shortcomings of traditional clustering methods, such as being prone to getting trapped in local optima, being sensitive to initial values, and requiring a preset number of clusters, thus achieving efficient and robust network set clustering.

The application belongs to the field of non-stationary signal processing, health monitoring and fault diagnosis, and discloses an online instantaneous frequency ridge extraction method based on an adaptive cost function. The method innovatively proposes an online instantaneous frequency extraction algorithm framework, and through the construction of an adaptive cost function, introduces a search frequency band dynamic adjustment strategy, converts the ridge line search process into a constrained optimization problem, so that it can accurately identify and extract non-continuous or intermittent frequency components in a low signal-to-noise ratio environment, and effectively solve the feature extraction problem caused by local signal distortion or transients. The experimental results show that the method has good instantaneous frequency extraction accuracy and robustness, provides an efficient and reliable technical means for real-time operation state monitoring and early fault diagnosis of the compressor, and has important engineering application value.

This invention discloses an automatic parallel optimization method for large-scale model training in hybrid heterogeneous clusters. This invention requires only real-time collection of a small number of system performance parameters to achieve efficient prediction of training time costs in hybrid heterogeneous cluster environments, while comprehensively considering factors such as GPU performance differences, communication overhead, and resource allocation. This invention formalizes the parallel strategy and parameter optimization problem for large-scale model training in hybrid heterogeneous clusters into a constrained optimization problem and establishes a bandwidth-aware training cost estimation model based on a theoretical model. It decomposes and estimates the time cost during training item by item, achieving rapid and accurate prediction of training costs under different heterogeneous configurations. This invention also designs a parallel strategy search and parameter optimization method based on intelligent optimization algorithms, which significantly reduces the traditional search space size while ensuring near-optimal solutions, improving search efficiency and reducing additional overhead.

The application discloses a meshless frequency estimation method for layout unconstrained blade tip timing signal and belongs to the technical field of array signal processing. The method comprises the following steps: setting a sampling model; setting the sampling model of the irregular layout of the blade tip timing signal according to the basic principle of the blade tip timing and the continuous compressive sensing theory; obtaining the rank constraint optimization problem description based on the atomic norm from the sampling model; obtaining the characteristic matrix through the irregular Vandermonde matrix decomposition; solving the rank constraint optimization problem based on the atomic norm through the alternating projection algorithm; and obtaining the frequency estimation value of the signal from the irregular Vandermonde matrix based on the polynomial root estimation method of the root multiple signal classification. The application is based on the basic principle of the blade tip timing and the continuous compressive sensing theory, and the frequency grid does not need to be divided in advance, so that the sensor layout is not restricted by the sparse linear array layout, and the blade tip timing signal obtained is uniform.

The invention discloses a runner plate assembly layout and pipeline optimization method, and relates to the field of new energy automobiles. The invention provides a runner plate assembly layout and pipeline optimization method based on optimization modeling and an intelligent search algorithm. A decision framework based on Laplacian matrix spectrum characteristics is adopted, and a center layout generation algorithm or a force-oriented layout generation algorithm is adaptively selected. According to the method, a detailed mathematical model is established to define a layout space, components, interfaces and a connection relationship, a runner plate design problem is converted into a constraint optimization problem, and an efficient optimization algorithm is used for solving, so that a layout scheme with minimum flow resistance is automatically generated.

The application discloses a kind of intelligent vehicle motion planning and control method for traffic light intersection, belongs to intelligent automobile automatic driving trajectory planning and motion control technical field.The purpose of the present application is around the scene with traffic light intersection, according to vehicle, intersection relative position relationship and signal light phase, design self-vehicle speed time-varying constraint condition for intelligent vehicle motion planning and control method of traffic light signal intersection.The present application is around the scene with traffic light intersection, according to vehicle, intersection relative position relationship and signal light phase, design self-vehicle speed time-varying constraint condition, utilize model predictive control to solve constrained optimization problem, propose corresponding motion planner scheme, give the model predictive control trajectory tracking controller design connected with upper layer, realize stable tracking to the local reference information planned.The present application considers the state of preceding vehicle and real-time traffic light phase, adopts speed constraint mode, brings more optimization indexes for local trajectory planning problem, guarantees driving safety, and also improves road traffic efficiency.

The application discloses a hydraulic arm safety control method and device based on ASMO and CBF, adopts a D-H method to establish a kinematics model of a hydraulic heavy-load mechanical arm, converts a kinematics uncertainty problem into an equivalent disturbance estimation problem, designs an adaptive sliding mode observer for the uncertain kinematics system to realize estimation on the kinematics disturbance, and automatically approaches a real value of an upper limit of the disturbance through adaptive adjustment, constructs an ASMO-CBF with an observer correction term based on a disturbance estimation result, and realizes safety correction on a desired joint speed by solving a constraint optimization problem QP.

Abstract In some aspects, a computing system can generate and optimize a neural network for risk assessment. The neural network can be trained to enforce a monotonic relationship between each of the input predictor variables and an output risk indicator. The training of the neural network can involve solving an optimization problem under a monotonic constraint. This constrained optimization problem can be converted to an unconstrained problem by introducing a Lagrangian expression and by introducing a term approximating the monotonic constraint. Additional regularization terms can also be introduced into the optimization problem. The optimized neural network can be used both for accurately determining risk indicators for target entities using predictor variables and determining explanation codes for the predictor variables. Further, the risk indicators can be utilized to control the access by a target entity to an interactive computing environment for accessing services provided by one or more institutions. Abstract 20 26 20 53 15 06 J ul 2 02 6 A b s t r a c t 2 0 2 6 2 0 5 3 1 5 0 6 J u l 2 0 2 6

The invention relates to a customer life cycle value prediction and intervention method based on dynamic time sequence analysis, and the method comprises the steps: obtaining a multi-source data set corresponding to a target customer, and generating a customer behavior data set; performing time sequence segmentation on the customer behavior data set to obtain a corresponding behavior feature vector sequence; inputting the behavior feature vector sequence into a value prediction model to generate predicted value trajectory information; candidate intervention actions are determined, and simulation prediction trajectory information is obtained through simulation; calculating expected value increment information corresponding to the candidate intervention actions, constructing and solving a constraint optimization problem, and generating a target intervention strategy; in conclusion, the customer life cycle value trajectory is analyzed and predicted through the dynamic time sequence, the effect of the intervention action is simulated to generate the optimization strategy, customer behavior changes are dynamically captured, continuous value trajectory prediction is provided, and the expected effect of the intervention action is quantified; and the accuracy of customer life cycle value prediction and the practicability of the intervention strategy are improved.