1144results about "Non-linear system models" patented technology

The invention provides an intelligent inspection path planning method and system based on reinforcement learning, and relates to the technical field of path planning, comprising the steps of constructing a scene understanding model and a dynamic interaction model, and forming a cognitive enhanced digital twin system. Metaheuristic search and social learning mechanisms are fused, and the inspection efficiency, energy consumption and risk are balanced through a multi-target conflict coordinator. The group behavior emergence prediction model achieves dynamic prediction of a collaborative mode, the federated learning framework achieves aggregation of experience knowledge, a group evolution strategy with high adaptability is generated, and closed-loop optimization from task execution to cognitive enhancement is achieved through a robustness evaluation mechanism fed back in real time and a task allocation optimizer based on an attention mechanism. And the inspection efficiency and adaptability are improved.

The invention relates to the technical field of quartz sand detection, in particular to a quartz sand impurity detection method which comprises the following steps: acquiring spectral data of a quartz sand sample, and obtaining spectral line interference degrees of characteristic peaks according to energy distribution conditions of the characteristic peaks of the spectral data; based on the peak shape variation condition of each characteristic peak and the offset condition of each characteristic peak and a matrix characteristic peak, the matrix interference degree of each characteristic peak is obtained, and the interference intensity of each characteristic peak is obtained by combining the anti-interference degree of each characteristic peak so as to optimize the chaos regulation factor when the chaos mapping population is initialized; and finally, combining with an optimization algorithm to obtain a characteristic wave band with minimum interference, and detecting the purity and impurity content of the quartz sand based on the characteristic wave band. The quartz sand impurity detection precision can be improved.

The invention relates to the technical field of agricultural intelligent irrigation, and discloses a high-standard farmland intelligent irrigation system based on Internet of Things and data analysis, and the system comprises a soil moisture content sensing module which collects data through a multi-source sensor to construct a three-dimensional soil moisture content distribution model, and generates a soil moisture content characteristic spectrum; the soil moisture content prediction module generates water demand prediction data based on the soil moisture content characteristic spectrum, the meteorological data and the crop growth stage; the irrigation strategy module fuses terrain elevation and pipe network pressure parameters to generate an irrigation control map; the equipment state monitoring module collects water pump current waveform and other data to generate equipment health degree parameters; the pipe network optimization module optimizes pipe network topology and generates an adjusting instruction; the multi-source data fusion module generates fusion evaluation indexes by using an evidence theory, an entropy weight method and the like; and the intelligent execution module generates an execution control instruction accordingly. All the modules cooperate to achieve precise irrigation, and the utilization efficiency of water resources and the intelligent level of farmland management are improved.

The present application relates to the technical field of gallium nitride drivers, and discloses an efficient heat dissipation management method and system for gallium nitride drivers, the method comprising: performing three-dimensional modeling and meshing on the gallium nitride driver to construct an initial heat dissipation model; performing feature extraction and transformation analysis on temperature field distribution data in the initial heat dissipation model to generate a temperature sensor layout scheme; collecting multi-operating condition temperature data according to the temperature sensor layout scheme, and performing local weighted regression analysis and thermal characteristic parameter extraction on the multi-operating condition temperature data to obtain characteristic parameters of the heat dissipation system; performing difference calculation on the characteristic parameters of the heat dissipation system and the initial heat dissipation model, and iteratively optimizing the heat dissipation parameters through a polynomial temperature regression method to obtain a target heat dissipation model; performing heat dissipation performance calculation and scheme evaluation based on the target heat dissipation model, and outputting a heat dissipation system optimization scheme, thereby realizing accurate updating of the heat dissipation model and improving the model prediction accuracy.

The invention relates to the technical field of safety monitoring, and particularly discloses an embankment danger rapid identification method and system, and the method comprises the steps: collecting multi-modal data in real time through arranging a multi-source sensor network; according to the phase space trajectory, extracting a Lyapunov exponent spectrum, correlating the dimension and the Kolmogorov entropy, and forming a structure response chaos degree index; calculating a hydrogeological coupling coefficient in combination with multi-scale decomposition and mutual information analysis; and fusing the two into a three-dimensional dangerous case feature tensor, inputting the three-dimensional dangerous case feature tensor into a pre-training model based on a deep convolutional neural network and a long-short-term memory network, realizing intelligent discrimination of high, medium and low risk levels, generating an adaptive monitoring instruction for a low-risk working condition, outputting a risk evolution trend map, and supporting closed-loop management and control.

The invention discloses a charging and discharging control system and electronic equipment. The method comprises the following steps: S1, constructing a charging pile multi-target optimization model; s2, initializing a chaotic particle swarm optimization algorithm; s3, each particle in the particle swarm is optimized according to an objective function in the charging pile multi-objective optimization model; s4, carrying out loop iteration until the chaotic particle swarm optimization algorithm reaches a convergence condition; s5, generating a prediction result of the operating environment of the charging pile; s6, updating the current optimal particle and the global optimal particle; and S7, generating a multi-target charging and discharging optimization control strategy of the charging pile according to the optimized global optimal particle, applying the optimization control strategy to a charging pile charging and discharging control execution module, and executing multi-target charging and discharging control of the charging pile. According to the method, the globally optimal solution in a complex scene can be quickly solved, and the efficient operation requirement of the electric vehicle charging pile in a large-scale deployment scene is met.

The invention discloses a VMD-PIML bearing fault prediction method based on physical information constraint, and the method comprises the steps: firstly improving an FNGO algorithm, initializing eagle group distribution through introducing chaotic mapping, and introducing a sine function and a refraction reverse learning strategy, thereby improving the calculation precision and search capability of the algorithm; iterating VMD important parameter composition [K, alpha] by using an improved FNGO algorithm, and extracting signal features; and finally, using an improved FNGO algorithm to optimize hyper-parameters in the PIML prediction model. By using the PIML prediction model, physical constraint and a machine learning model can be combined, bearing signal features can be synthesized, and the accuracy and robustness of model prediction can be improved. Compared with the prior art, the method has the advantages that the PIML prediction model is constrained through physical knowledge, the improved NGO algorithm is used for optimizing the VMD denoising model and the PIML prediction model, and generalization and robustness of bearing fault prediction are improved.

The invention discloses an artificial intelligence-based bid invitation file error content optimization method and system, and relates to the technical field of artificial intelligence, and the method comprises the steps: carrying out the causal inference through employing a CIRCA frame, constructing a causal Bayesian network, introducing the Chen's three-dimensional chaotic disturbance, optimizing the edge probability, recognizing potential error nodes through a GNN model, generating a semantic embedding vector through employing SBERT, and carrying out the calculation of the error content of a bid invitation file. The method comprises the following steps of: mapping to a Riemannian manifold, realizing embedded dimension reduction through Riemannian principal component analysis, determining an optimal path leading to an ideal state by using a geodesic path algorithm, and calculating a candidate fitness score by constraining an Abbe structure and conflict strength. According to the method, causal reasoning and chaotic disturbance are combined, so that the accuracy of error positioning and the reliability of classification judgment in the bid invitation file are improved, and the semantic error recognition capability and the global optimality of a correction path are improved through semantic manifold modeling and geodesic path optimization.

The invention discloses a reservoir group scheduling interval multi-target optimization method under an uncertain condition, and belongs to the technical field of reservoir optimization scheduling. The method comprises the following steps: firstly, establishing an innovative cascade reservoir group interval multi-target scheduling model; secondly, providing a double-population dynamic coordination interval optimization method for solving the interval multi-target scheduling model, and further obtaining an optimal scheduling scheme of the reservoir group; and finally, outputting an optimal scheduling scheme set of the reservoir group. According to the optimization method, through organic combination of a chaos-order mapping initialization strategy, a dynamic adaptive mutual promotion updating strategy and an elite guidance optimization strategy, optimization is rapidly and efficiently carried out to obtain a group of reliable and accurate scheduling scheme sets considering runoff uncertainty; an effective decision-making technical means for solving the cascade reservoir group optimization multi-target complex engineering problem is provided for a scheduling decision maker, the intelligent level and comprehensive benefits of water resource management are improved, and the capacity of a water resource system for coping with complex environments and uncertain factors is further enhanced.

The invention relates to the technical field of typhoon prediction, and discloses a typhoon rapid enhancement prediction method based on time-space sequence and multi-modal feature fusion, and the method comprises the steps: constructing a multi-modal time-space sequence data set and an auxiliary data set based on typhoon optimal path data and multi-source satellite observation data; a unified manifold approximation and projection method is adopted to carry out dimension reduction preprocessing on the high-dimensional multi-modal space-time sequence data, and one-dimensional time sequence embedding representation of the typhoon observation sequence is generated; taking the one-dimensional time sequence embedded representation and the auxiliary data as independent input channels, and inputting a trained typhoon observation network model to predict a typhoon rapid enhancement probability; wherein the typhoon observation network model is a multi-mode time-space fusion deep learning architecture, the core of the typhoon observation network model is composed of a variational attention recurrent neural network, and hyper-parameter optimization is carried out through an improved Harris eagle optimization algorithm. According to the invention, accurate and robust identification of the typhoon rapid enhancement process is realized.

A system and method for automatic generation and automatic conduct of a chaos experiment in distributed applications includes a software application having a layer and a chaos configuration, a chaos engine having a chaos orchestrator, a chaos prompt generator having a prompt template composer, and a chaos engineering large language model. The chaos engineering large language model is configured to recommend chaos experiments for the chaos engine to process and to inject into the software application. The chaos engineering large language model is also configured to patch the software application in accordance with objectives and parameters developed in the system and method. The system and method may be automated through a trigger. The trigger may be the chaos engineering large language model recommending one or more chaos experiments to the chaos engine. The trigger may also be the implementation of one or more patches to the software application.

The invention discloses a water turbine wear fault diagnosis method based on a deep learning model, and belongs to the technical field of water turbine fault signal processing. Comprising the following steps: acquiring a water turbine wear fault sound signal; a Tent chaotic mapping function is adopted to optimize an initial population of a white shark optimization algorithm WSO; introducing a long and short term memory (LSTM) network model to solve the limitation of a convolutional neural network (CNN) in processing a long sequence problem, and constructing a CNN-LSTM water turbine wear fault diagnosis model; an improved white shark optimization algorithm IWSO is introduced into the CNN-LSTM model, and hyper-parameters of the model are optimized, so that the diagnosis precision is improved, and the calculation amount is reduced; and importing the collected fault signal as an input into the model for training and testing to obtain a final diagnosis result. Diagnosis and identification of the water turbine wear fault signal are completed through the steps, the deep learning model has good robust performance, the diagnosis precision of the water turbine wear fault can be improved, and the deep learning model serves as a beneficial supplement of an existing water turbine state monitoring and fault diagnosis system.

The invention discloses a charging pile direct current power distribution method based on chaos topology, and the method comprises the steps: building a charging pile power initial chaos topology distribution network based on a matrix topology network according to user demands; expanding the dimension of the charging pile power initial chaotic topological distribution network to a high-dimensional curved surface through a Gaussian kernel function to obtain a charging pile power chaotic topological distribution network with a plurality of charging nodes; on the basis of the charging pile power chaos topology distribution network, the required power of each charging gun and a power distribution strategy, controlling on and off of a contactor switch unit so as to control guiding of power output by a power conversion unit; therefore, efficient power flexible scheduling can be realized, the degree of freedom of all nodes in the power topology can be customized at will, and it is ensured that each node can reach the efficiency expected by a user.

The invention relates to a ship port resource coordinated scheduling method and device, and belongs to the technical field of port resource optimized scheduling, and the method comprises the steps: building a target function with the maximization of the anchor ground utilization rate, the maximization of the berth utilization rate and the minimization of the ship waiting time as targets; solving the objective function by using MILP and a particle swarm algorithm to obtain an initial scheduling scheme meeting constraint conditions; the initial scheduling scheme comprises anchor ground distribution, berth distribution, berthing time and leaving time; clustering the initial scheduling scheme based on the multi-core weighted distance of the initial scheduling scheme to obtain a clustered scheduling scheme; and optimizing the clustered scheduling scheme based on a game theory model to obtain a target scheduling scheme. According to the ship and port resource coordinated scheduling method provided by the invention, global optimization of ship berth allocation, channel use and anchor ground scheduling is realized, and high efficiency and safety of ship operation are ensured while channel resource conflicts are reduced.

The invention discloses a memristor neural network circuit for generating a multi-direction multi-double-scroll chaotic attractor, which comprises a first integral channel circuit, a second integral channel circuit, a third integral channel circuit, a first memristor, a second memristor, a third memristor, a first switch, a second switch and a third switch, and is characterized in that the first integral channel circuit comprises three input ends and one output end; the second integral channel circuit comprises three input ends and one output end; the third integral channel circuit comprises two input ends and one output end. The memristive neural network circuit can generate a multi-direction multi-scroll chaotic attractor by controlling the on-off state of the memristive channels connected with the input ends of the integral channel circuits, and the circuit has a discharge characteristic closer to an interactive neuron in a real nervous system; the method can provide new possibility for physically constructing an intelligent neuromorphic system, and has important value in the fields of chaotic secret communication, image encryption, complex signal processing and the like.

The invention provides an image enhancement optimization method based on multi-strategy improved sparrow search. The method comprises the following steps: performing normalization processing on a gray value of each pixel point of an original image; initializing a sparrow population and parameters according to an elite chaos reverse learning strategy; the method comprises the following steps: preprocessing an input image by adopting a dual gamma function BiGA; calculating the value of the evaluation function; updating the optimal fitness value of the sparrow population; when an iteration stop condition is not met, updating the position of the sparrow population by adopting a probabilistic following strategy, a self-adaptive local search strategy and a Cauchy Gaussian mutation strategy, and updating parameters of a dual gamma function BiGA; otherwise, ending iteration, and calculating an optimal parameter; processing the image, and outputting an enhanced image; according to the method, the visual effect of the image can be improved, edge details and textures can be richer, and the enhanced image can have higher contrast and better definition. The method is best in overall enhancement effect, and the obtained image is better in quality.

The invention relates to a virtual power plant bidding strategy optimization method based on a chaotic game optimization algorithm. The method comprises the following steps: generating an initial bidding strategy of a virtual power plant through chaotic mapping; each virtual power plant receives feedback information of the market, games by considering factors of cost, income, fluctuation of photovoltaic power generation, scheduling cost and attenuation of an energy storage battery and service life of equipment, and adjusts a bidding strategy of the virtual power plant; punishment income is applied to bidding behaviors violating market ethics or benefit distribution constraints; calculating the income of the power plant according to the adjusted strategy, and updating the bidding strategy of each virtual power plant; and repeating the game process until a convergence condition is reached, and outputting a virtual power plant bidding strategy. Compared with the prior art, the bidding strategy of the virtual power plant is more flexible and adaptive by using the randomness and unpredictability of the chaotic system, and meanwhile, the fairness, sustainability and healthy market development of the system are ensured by introducing ethical constraint and equipment life management.

The invention discloses a hydroelectric generating set fault diagnosis method based on a chaos Kepler optimization algorithm, and belongs to the technical field of hydroelectric generating set fault diagnosis. A Tent chaos function is introduced to replace an initial population of a Kepler optimization algorithm (KOA), and by improving the uniformity and uncertainty of the population, the convergence speed of the KOA algorithm is increased, and the global search capability is improved. An improved Kepler optimization algorithm is adopted to optimize hyper-parameters of the BiLSTM, a BiLSTM prediction model CKOA-BiLSTM of the hydroelectric generating set is established, the robustness and generalization ability of an existing model are improved, and the model is made to be suitable for processing complex nonlinear fault signals of the hydroelectric generating set. The model shows the optimal diagnosis effect on recognition of complex fault types, and the research can be used as a beneficial supplement of an existing hydroelectric generating set state monitoring and fault diagnosis system.

The invention discloses a hierarchical encryption method, system, device and medium for a privacy image, and relates to the technical field of image processing and information security, and the method comprises the steps: segmenting a to-be-encrypted image divided according to a region threshold, and obtaining images with privacy of different importance degrees; according to the privacy importance degree of an image region, selecting encryption algorithms of different levels to encrypt a high-privacy-level image, a middle-privacy-level image and a low-privacy-level image, and performing main element product operation by using a chaos matrix and the encrypted high-privacy-level image, middle-privacy-level image and low-privacy-level image to obtain a high-privacy-level image, a middle-privacy-level image and a low-privacy-level image; and a final encrypted image is obtained. According to the image encryption method, selective algorithm encryption is carried out on each part of the to-be-encrypted image, high-strength encryption processing does not need to be carried out on the whole image, and the encryption method is numerous, so that the encryption strength and security can be remarkably improved while the encryption efficiency is improved.

The invention provides a pelican optimization strategy-based mobile edge computing unloading method, and aims to solve the problems of low computing task unloading efficiency and poor system performance in application scenes such as Internet of Things and smart cities. By introducing chaotic mapping and a flight strategy, a traditional pelican optimization algorithm is improved, and the global search ability and the ability of jumping out of a local optimal solution of the algorithm are enhanced. A calculation unloading model of a multi-user multi-edge server is constructed, and delay and energy consumption in the data transmission and calculation process are comprehensively considered. Experimental results show that the method can effectively reduce the time delay and energy consumption of the system, optimize resource allocation and improve the overall performance of the system. Especially in a multi-user and multi-task complex scene, the method disclosed by the invention is excellent in performance, and a new solution is provided for the calculation unloading problem in the mobile edge calculation.

The invention discloses a big language model-based RAG enterprise data hierarchical decentralized management method, which relates to the technical field of data security and access control, and comprises the following steps of: acquiring a voice audio stream and a lip-shaped three-dimensional track point set, loading permission baseline parameters, and generating a permission token; the permission token comprises a quantum watermark seed, a role identifier, a permission token aging stamp and a permission level; based on the permission level and the role identifier, obtaining permission data blocks through a retrieval enhancement generation framework, dividing levels according to permission baseline parameters, generating retrieval keywords based on the permission token, establishing a hierarchical index, matching the permission data blocks, performing role identifier matching on the permission data blocks, then reordering, and generating a response text; converting the quantum watermark seed into a binary stream, and embedding the binary stream into the response text to obtain a watermark response text; according to the method, through multi-modal biological feature fusion and Shannon entropy calculation, the problem that the authority decision is lack of dynamic risk perception is solved.

The invention belongs to the technical field of motor fault diagnosis, and particularly relates to an asynchronous induction motor fault intelligent detection method based on a multi-modal time sequence. The method comprises the following steps: acquiring various signals of a motor, including three-phase current, vibration, equipment sound, environment sound and the like, and constructing a multi-modal data set; thirdly, constructing a Transform network classified by a single time sequence, and on the basis of the Transform network, connecting a plurality of the networks in parallel to obtain a lumped Transform network; and then, features of different modal data are dynamically integrated by using a self-attention mechanism, and accurate recognition of a motor fault is realized. In a network training process, a hyper-parameter of the lumped Transform network is optimized by adopting a chaos WOA algorithm; finally, different types of faults of the asynchronous motor can be accurately detected and judged, the situation that different fault features of the motor cannot be accurately represented by a single time sequence is avoided, and the accuracy and reliability of fault detection of the asynchronous motor are guaranteed.

The invention relates to a multi-objective collaborative optimization scheduling method for an integrated energy system, and aims at a multi-energy unit optimization problem, namely a multi-parameter optimization problem, in an optimization model, an optimization model parameter initialization method based on improved chaotic mapping is adopted. An initial parameter with a high random average distribution probability is generated to improve the convergence speed of a system optimization process and reduce the probability of falling into a local optimal solution. Meanwhile, aiming at a multi-objective optimization problem, an improved grey wolf algorithm based on a Pareto optimal solution theory is adopted to realize optimal scheduling of the system for an economical efficiency optimal objective and a carbon emission minimum objective. According to the system, a comprehensive energy system architecture including water electrolysis hydrogen production using photovoltaic and wind power generation and hydrogen energy storage, power generation, heating and refrigeration is comprehensively considered, comprehensive utilization and control of various new energy power generation and utilization units are achieved, and carbon emission can be effectively reduced.

The invention relates to the technical field of hardware fault early warning, in particular to a server hardware fault early warning and recovery method and system based on an intelligent optimization algorithm, and the method comprises the steps: obtaining hardware self-inspection information; acquiring load state information; acquiring historical operation data; constructing a preliminary mapping matrix based on the acquired hardware self-inspection information, load state information and historical operation data; according to the preliminary mapping matrix, through a topological space optimization algorithm, key modes are extracted, and an optimized mode matrix is generated; based on the optimized mode matrix, combining a fault-tolerant strategy group, and screening a fault-tolerant or backup strategy suitable for the current state; according to a fault-tolerant or backup strategy, evaluating the benefit, the cost and the success probability of the strategy, and generating an optimal decision scheme; outputting fault early warning information including early warning time, an early warning level and a trigger strategy; and outputting an optimal decision scheme, including a fault-tolerant strategy or a backup strategy, so that a more complex hardware abnormal mode can be captured, and the accuracy of early warning is greatly improved.

The invention discloses a mine intelligent ventilation combined optimal adjustment and optimal control method and system. The method comprises the steps that S1, multi-source mine ventilation data are collected through a distributed multi-mode sonic sensor by adopting a gridding deployment strategy; s2, performing fusion optimization on the multi-source mine ventilation data by using a trust function theory to obtain comprehensive state data; s3, on the basis of the comprehensive state data, obtaining optimal preliminary regulation and control actions of different mine equipment through a self-adaptive reinforcement learning model; and S4, based on the optimal preliminary regulation and control action, introducing a roadway structure safety degree as a dynamic constraint factor, combining with an improved pigeon inspired optimization algorithm to obtain a final regulation and control action, and realizing underground ventilation control based on the final regulation and control action. The safety degree of the roadway structure is incorporated into a fitness function, so that the safety of the roadway structure is fully considered while the ventilation effect and the energy consumption are optimized by the algorithm.

The invention belongs to the field of uncertainty quantification and robustness design optimization of aero-engine air-cooled turbines, and particularly discloses a one-dimensional uncertainty design optimization method and system for a reverse cooling turbine based on a one-dimensional aerodynamic analysis method for the reverse cooling turbine. Forming an augmented space by the optimization variables and the uncertainty parameters, and generating a sample set; establishing a Kriging global agent model, and generating an uncertainty parameter sample set; an ASPC model is established, and one-dimensional uncertainty quantification of the reverse cooling turbine is completed; an NSGA-II multi-objective optimization algorithm is coupled, and one-dimensional robustness design optimization of the reverse cooling turbine is completed. The one-dimensional uncertainty design optimization method and system framework of the reverse cooling turbine are provided and established, aerodynamic performance analysis of the reverse cooling turbine can be completed through simple parameters, a geometric entity is not needed, and tasks such as one-dimensional uncertainty quantification and robustness design optimization of the reverse cooling turbine can be achieved.

The invention discloses an engine turbine shaft sequential optimization design method based on ensemble learning, and the method comprises the steps: carrying out the parametric modeling of an aero-engine turbine shaft through employing a finite element method, and determining a design variable; determining an optimized objective function and constraint conditions based on the turbine shaft structure; constructing a training point set by using the target function and the constraint condition, and establishing an ensemble learning agent model by using the training point set through an ensemble learning algorithm; executing a sequential optimization design process based on the ensemble learning agent model; searching a current optimal sequence design point by utilizing an intelligent algorithm and a sequential updating criterion, and judging whether the sequence design point meets a threshold value requirement or not; if the current optimal sequence design point meets the threshold requirement, outputting the current optimal sequence design point as a design result; otherwise, adding the sequence design points as new training points into the training point set to update the training point set, updating the ensemble learning agent model by using the updated training point set, and iteratively calculating the sequence design points.

The invention discloses an intelligent power prediction method considering dynamic load change, and relates to the technical field of power grid load prediction, and the method comprises the steps: collecting original load data, carrying out the preprocessing, constructing a VMD constraint optimization model, carrying out the four-stage improvement of optimization parameters through employing an improved dung beetle optimization algorithm, so as to generate IMF components, reconstructing the IMF component by calculating a sample entropy to obtain a low-frequency component and a high-frequency component; establishing a Kalman filtering state space model based on the low-frequency component, and decomposing the low-frequency component into a residual component and a pseudo trend component through a Kalman filtering recursive algorithm; external features are obtained, the high-frequency component, the residual component and the pseudo trend component are aligned and spliced with the external features, multi-component collaborative prediction is carried out through a local-global interactive attention mechanism, and a final load prediction result is obtained; and generating a power demand visualization chart based on the final load prediction result. And reliable decision support is provided for power dispatching and energy management.

The invention belongs to the technical field of radiation protection, and particularly relates to a neutron shielding material optimization method based on machine learning and chaotic mapping NSGA-III, and the method comprises the steps: 1, determining an optimization target, and building a multi-target neutron composite shielding material attribute optimization model; step 2, building a Monte Carlo model to simulate and calculate shielding doses of the composite shielding materials with different attributes to a neutron source, and obtaining a data set with the material component proportion and thickness as characteristic attributes and the shielding doses as labels; and step 3, optimizing initial parameters of the BP neural network by adopting an improved particle swarm optimization algorithm, improving calculation precision and operation efficiency of the neural network, and training a neural network regression model by taking the data set obtained in the step 2 as a sample to complete establishment of a shielding dose prediction model and the like. An improved particle swarm optimization algorithm for improving Levy flight mechanism strategy optimization is designed to optimize initial parameters of the BP neural network, and the precision and operation efficiency of a neural network shielding dose regression prediction model are improved.

The invention discloses an electric vehicle charging load prediction method, which comprises the following steps: firstly, screening variables with high correlation through a maximum information coefficient, selecting features with high correlation as model input, and taking a charging load as output; reconstructing the data training set by adopting cross validation; a charging load integration model is constructed, the integration model is divided into a base learning device and a meta learning device, the base learning device comprises a DCformer model, a BiTCN model and an artificial neural network ANN model, and a lightweight gradient elevator LightGBM is used as the meta learning device; the parameters of the LightGBM model are optimized by adopting an improved star graffiti optimization algorithm INOA; an error sequence is obtained after a preliminary prediction value is obtained through model prediction, and a final prediction value is obtained by taking LightGBM as an error prediction model. According to the method, data features are effectively extracted by integrating a plurality of single models, the prediction accuracy is improved, and compared with a traditional model, the stability is better, and the generalization ability is higher.