197 results about "Fast optimization" patented technology

The invention provides an evaluation method, system and device for a large language model and a storage medium, and the method comprises the steps: constructing a sample data set which comprises input data and corresponding reference answers, inputting the input data into an evaluated model, and generating an output result; a multi-dimensional evaluation index system is designed according to task requirements, a dynamic weight is allocated to each evaluation index, each evaluation index is provided with scoring standard description, and the evaluation indexes comprise at least two items of context correlation, term consistency, language fluency and expression accuracy; combining the input data, the reference answer, the output result and the scoring standard description into a standardized input instruction, and calling an evaluation model to perform multi-dimensional scoring on the standardized input instruction to generate an evaluation result; and analyzing the evaluation result according to a preset threshold value, and generating a structured feedback suggestion containing an improvement direction. According to the method, rapid optimization and iteration of the large language model can be effectively supported.

The invention discloses a multi-modal feature joint constraint whole-brain dynamics modeling and parameter optimization method which is applied to the field of computational neuroscience and brain science. Aiming at the defects of a current whole brain dynamic model in the aspect of synchronously fitting multi-modal scalp EEG signals and cortical and subcortical BOLD signals across spatial and temporal scales, the capability of the model for comprehensively and accurately representing real dynamic characteristics of brain activities on different scales is limited; in order to overcome the short plate, the core element of multi-modal feature combined constraint is added on the basis of a whole-brain dynamic model and a parameter space rapid optimization method, and a modeling method and a parameter rapid optimization strategy for the multi-modal feature combined constraint whole-brain dynamic model are designed; by applying a high-dimensional parameter rapid optimization algorithm, the capability of the whole-brain dynamic model in synchronous fitting of brain multi-modal spatial-temporal characteristic signals is further improved; the method has important significance and reference value for developing a whole-brain dynamic model modeling method and a parameter optimization system which are more biological rationality and interpretability.

The invention provides an airfoil profile adaptive design and optimization method based on reinforcement learning and a conditional generative adversarial network, and belongs to the field of airfoil profile optimization design. The method aims at achieving the two core functions of self-adaptive optimal airfoil generation and rapid optimization. According to the first function, through interaction of a reinforcement learning agent and a conditional generative adversarial network generator, generation strategies for generating optimal airfoil profile potential variables under different design conditions are continuously learned, so that the airfoil profile with the optimal aerodynamic performance is adaptively generated, intelligent and adaptive design of the airfoil profile is achieved, and the optimization period is remarkably shortened. Secondly, a self-adaptive agent model based on meta-learning is introduced, the model not only establishes a mapping relation among design conditions, aerodynamic performance and potential variables and directly predicts the optimal potential variables of a specific airfoil profile, but also can quickly adapt to an updated database through simple adjustment; the agent model predicts the optimal potential variable corresponding to the specific airfoil profile for the self-adaptive generation framework, so that the agent can quickly optimize the specific airfoil profile under the condition of fine adjustment.

The invention discloses an intelligent optimization method for a multi-type well seam joint control fine injection-production mode, and relates to the technical field of oil-gas field development. The method comprises the following steps: setting a well seam joint control fine injection-production mode, establishing an oil reservoir numerical simulation model in oil reservoir numerical simulation software, obtaining multiple groups of oil reservoir injection-production schemes based on a Latin hypercube sampling method, performing simulation according to each group of oil reservoir injection-production schemes by utilizing the oil reservoir numerical simulation model, generating multiple pieces of sample data, and establishing a sample database; a deep learning agent model is established, after the sample database is utilized to train and train the deep learning agent model, a particle swarm optimization algorithm is adopted to carry out single-target pre-search global optimization to obtain a preferred reference strategy, a reinforcement learning dynamic decision model is established, and a reinforcement learning agent is obtained through training based on a PPO near-end strategy optimization algorithm; and the optimal injection-production development scheme of the oil reservoir is obtained by utilizing the reinforcement learning agent, so that rapid optimization and decision support of the oil reservoir injection-production scheme in a new multi-type well seam joint control mode are realized.

The invention discloses a flood regulation and control two-stage reverse deduction method under a large-scale watershed complex engineering system, which comprises the following steps: S1, constructing a watershed flood deduction model: constructing a flood control engineering system joint scheduling and flood routing coupling model according to watershed water system and flood control engineering characteristics; evaluating flood influence and disaster loss based on the coupling model; s2, two-stage reverse deduction based on the flood deduction model and the intelligent optimization algorithm: constructing a calculation framework based on the flood deduction model and the intelligent optimization algorithm, iteratively optimizing and dynamically adjusting scheduling parameters such as water volume, opening degree and the like by taking the minimum overall disaster as a reverse deduction target and taking the minimum watercourse overflow and collapse and the minimum flood diversion water volume as hard constraint conditions, and calculating the overall disaster of the flood deduction model and the intelligent optimization algorithm; and reverse deduction is carried out through a reverse deduction mechanism, and an optimal scheduling scheme considering flood control safety and disaster damage control is generated. The method has the advantages that fast and refined reverse deduction can be achieved, and the actual requirements of basin flood routing refined rehearsal, engineering regulation and control scheme fast optimization generation and the like are met.

The invention relates to the technical field of power system scheduling, and discloses an intra-day look-ahead scheduling rapid solving method considering large-scale new energy cluster power generation volatility. Comprising the following steps of S1, new energy cluster space-time fluctuation scene generation based on a neuron cellular automaton, S2, power grid dynamic security domain definition and simplification based on a physical information neural network, S3, scheduling rapid optimization solution based on model prediction path integration, and S4, scheduling scheme dynamic elasticity and stability evaluation based on a Kupman operator theory. The new energy cluster space-time fluctuation scene generation method based on the neuron cell automaton can effectively generate a space-time scene reflecting large-scale new energy cluster power generation volatility, supports uncertainty analysis, has the advantages of being high in calculation efficiency and scene authenticity, and is suitable for large-scale new energy cluster power generation. The problem that scene generation is inaccurate due to the fact that a traditional statistical model ignores space-time coupling is solved.

The invention discloses a fault co-seismic sliding surface inversion method and system based on Beidou and artificial intelligence. The method comprises the steps of collecting Beidou data and seismic waveform data of a target area in real time, and performing preprocessing to extract co-seismic displacement and waveform features; fusing the Beidou deformation features and the seismic wave features to form a multi-modal feature vector; a fault prediction model is constructed based on a graph neural network, forward modeling and historical earthquake example data training are utilized, and multi-modal features are input to obtain preliminary fault sliding distribution prediction; rapid optimization under physical constraint is carried out through the elastic dislocation model, and a final fault sliding model conforming to the geophysical law is obtained; finally, uncertainty quantification is carried out, and a visual product of the dynamic process including sliding distribution, seismic moments and fault deformation is generated. According to the system, full-process automatic processing is achieved through cooperation of all the units, the inversion speed, precision and physical credibility are improved, and reliable support is provided for earthquake emergency response and disaster assessment.

The invention discloses a data processing method and system for a precise automatic multi-axis numerical control electric spark forming machine, and the method comprises the steps: mapping preprocessed data into a digital twin model for fusion analysis, and extracting fusion features including instantaneous energy density, an effective machining angle, a thermal error compensation amount and a chip removal efficiency evaluation value; based on the fused feature data, performing real-time state evaluation and future discharge trend prediction by using a LightGBM neural network model, and outputting a processing prediction result; and a multi-target optimization function is established, an improved NSGA-II algorithm is utilized to carry out rapid optimization search by taking the current processing parameters as an initial population, a multi-axis cooperative motion instruction is generated, and the multi-axis cooperative motion instruction is transmitted to a multi-axis motion control module and a pulse power supply module to be executed. And the processing stability and reliability are improved.

The invention discloses a deep-sea exploration air gun array arrangement optimization method and system for improving wavelet quality, and relates to the field of deep-sea resource exploration. The problem of rapid optimization design of the air gun array in the existing engineering is solved. The method comprises the following steps: inputting initial information parameters of an air gun array, and obtaining air gun bubble pressure wavelet data and sound pressure level spectrum data before optimization based on an air gun bubble dynamics mathematical physical model; establishing upper and lower limits of the air gun sinking depth and constraint conditions of an air gun volume change range; setting an optimization objective function and optimization control parameters; performing iterative optimization on the array in combination with a spherical bubble dynamics theory and a particle swarm optimization algorithm; calculating an air gun array pressure wavelet and a sound pressure level frequency spectrum, if a calculation result meets a requirement, obtaining air gun array information optimized based on a particle swarm algorithm, calculating the pressure wavelet and the sound pressure level frequency spectrum, outputting the optimized air gun array information, and realizing air gun array inversion optimization; and if the requirements are not met, continuously carrying out iterative optimization on the array.

The invention belongs to the technical field of GNSS-R SAR (Global Navigation Satellite System-Radar Synthetic Aperture Radar) super-resolution imaging, and discloses a fast distance super-resolution imaging method based on a GNSS-R SAR. According to the method, matrix compression, regularization modeling and a rapid optimization algorithm are creatively combined, and a set of efficient and stable distance super-resolution processing flow is formed. Specifically, after a preliminary imaging result of a GNSS echo signal is obtained and a signal convolution model form of the GNSS echo signal is given, firstly, dimension reduction processing is performed on an observation matrix in a signal convolution model through singular value decomposition, and on the basis of a weighted singular value maintenance strategy processing result, the signal convolution model is reconstructed by using an inverse matrix of a truncated measurement matrix; then, starting from a regularization strategy, introducing an L1 norm constraint to construct a target function by utilizing the sparse characteristic of a target; and finally, solving the target function by adopting a rapid iterative optimization algorithm. According to the method provided by the invention, the range resolution of the GNSS echo data is remarkably improved.

The invention relates to a rapid optimization design method for a deep and far sea floating type offshore integrated platform mooring system, which comprises the following steps of: preliminarily designing a mooring scheme according to environmental conditions such as wind waves and flows and requirements such as an integrated platform design foundation, and then performing corresponding optimization; the specific optimization process comprises the steps that the non-dominated genetic algorithm and numerical calculation software are combined, the optimization process is divided into static analysis and dynamic analysis, static analysis is conducted firstly to reduce the initial range of parameters, then dynamic analysis is conducted to determine the optimal design scheme, the genetic algorithm and mooring analysis software are combined on the basis of the Python programming language, and the optimal design scheme is obtained. In the whole analysis process, only an initial parameter range needs to be set, the optimal design scheme in the parameter range can be obtained, and intelligent optimization of the design scheme of the mooring system is achieved.

The invention provides a horizontal subdivision productivity optimization method based on adaptive entropy deep reinforcement learning, and belongs to the field of oil and gas field development engineering, and the method comprises the following steps: constructing an adaptive entropy deep reinforcement learning strategy model, interacting with a horizontal well subdivision oil production reservoir shaft numerical simulator, storing horizontal well subdivision experience, and optimizing the horizontal well subdivision productivity. Training and updating the adaptive entropy deep reinforcement learning strategy model by using historical subdivision experience; the strategy model is applied, the oil reservoir state is input, and a subdivision optimization scheme at the current moment is rapidly output. Compared with an existing horizontal well subdivision oil extraction optimization method, the method has the advantages that a sequence subdivision optimization scheme can be quickly output, strategy coefficients are automatically adjusted, and the capacity of strategy exploration for global optimum is effectively improved. Meanwhile, the method is combined with an oil reservoir shaft coupling model, it is ensured that an optimization strategy is tightly attached to the actual physical process, the accuracy of yield evaluation is improved, and the purpose of rapid optimization of the horizontal well sequence subdivision scheme is achieved.

The invention discloses a quick optimization design method and device for anti-flutter structural parameters of a wind turbine blade, and relates to the technical field of wind turbine blades, and the method comprises the following steps: constructing a blade aeroelastic characteristic equation to respectively determine the blade critical flutter speed of a plurality of to-be-optimized structural parameter combinations of a to-be-designed blade; carrying out standard orthogonal polynomial approximate expansion on the initial parameter-critical flutter velocity high-dimensional expansion model, converting coefficient solution into a 1-norm minimization problem, and carrying out solution based on each to-be-optimized structure parameter combination and each blade critical flutter velocity to determine a target parameter-critical flutter velocity high-dimensional expansion model; calculating the sensitivity index of each to-be-optimized structure parameter to determine a key to-be-optimized structure parameter; taking maximization of the critical flutter speed of the blade as a target, adjusting the parameter gradient of each key to-be-optimized structure to solve a target parameter-critical flutter speed high-dimensional expansion model, and determining a target optimization structure parameter combination. According to the scheme, the optimal anti-flutter structure parameter design can be quickly determined.

The invention provides a rapid optimization method suitable for a radio frequency heterogeneous integrated microsystem interconnection structure, and the method comprises the steps: constructing an equivalent circuit model of a complex interconnection structure required by a system based on an equivalent circuit model of a basic interconnection structure unit or an equivalent circuit model library of the basic interconnection structure unit; parameter information such as the physical size of the interconnection structure is obtained by rapidly optimizing the constructed equivalent circuit model, then corresponding physical size parameters serve as initial values of the electromagnetic simulation model of the interconnection structure, and the electromagnetic simulation model of the interconnection structure is obtained by combining the trend and rule of the constructed equivalent circuit model changing along with the physical size of the interconnection structure. The optimization and design process of three-dimensional electromagnetic simulation of the interconnection structure is guided and accelerated; the design method effectively solves the problems that the efficiency is low and the resource overhead is high when a traditional three-dimensional electromagnetic simulation optimization design is adopted for the interconnection structure of the integrated microsystem, and rapid deployment of the radio frequency integrated microsystem is achieved.

The invention belongs to the technical field of ball valve machining process routes, and particularly relates to a multi-target flexible process planning system and method for a ball valve assembly, and the system comprises a data collection module, a constraint condition construction module, a target weight determination module, a target function design module, an optimization solution module, and an optimization result analysis and display module. And setting an optimal processing sequence of each process, distributing processing equipment, comparing and analyzing optimization results of different objective functions and system performance index change conditions before and after process route optimization, and visually displaying the optimization results. The method comprises the steps of S1, performing feature representation of the ball valve assembly, S2, determining constraint conditions in the machining process, S3, determining objective functions, S4, determining weight coefficients of different objective functions based on a fuzzy analytic hierarchy process, and S5, determining a process planning decision of the ball valve assembly under multiple constraints. According to the method, rapid optimization of a complex process route is realized, and the development trend of green manufacturing is met.

The present invention belongs to the technical field of GNSS-R SAR super-resolution imaging, and discloses a fast range super-resolution imaging method based on GNSS-R SAR. The present invention innovatively combines matrix compression, regularization modeling and fast optimization algorithm to form a set of efficient and stable range super-resolution processing procedures. Specifically, after obtaining the preliminary imaging results of the GNSS echo signal and giving its signal convolution model form, the observation matrix in the signal convolution model is first reduced in dimension by singular value decomposition, and on the basis of the processing results of the weighted singular value preservation strategy, the signal convolution model is reconstructed using the inverse matrix of the truncated measurement matrix; then, starting from the regularization strategy, the L1 norm constraint is introduced to construct the objective function using the sparse characteristics of the target; finally, a fast iterative optimization algorithm is used to solve the objective function. The method of the present invention significantly improves the range resolution of GNSS echo data.

The invention relates to the technical field of power distribution networks, and discloses a power distribution network and micro-grid collaborative reliability improvement method, which specifically comprises the following three steps: constructing an SOP-based power distribution network and micro-grid operation model, constructing an SOP-containing power distribution network and micro-grid operation model, and constructing an SOP-containing power distribution network and micro-grid operation model. It is ensured that the micro-grid can operate independently or cooperatively in different fault scenes, and the reliability and flexibility of the system are improved; a multi-stage power distribution network and micro-grid collaborative optimization model is constructed, a multi-stage optimization strategy is designed for different operation stages of a power distribution network and a micro-grid, full-process collaborative optimization from prevention before a fault, scheduling during the fault to recovery after the fault is realized, and efficient distribution and management of resources are ensured; according to the rapid optimization method of the collaborative model, a rapid optimization algorithm is provided for improving the real-time performance and the application efficiency of the model, and rapid response and implementation of collaborative optimization of a large-scale power distribution network and a micro-grid are achieved by reducing the calculation complexity and improving the solving speed.

The invention discloses a hierarchical reinforcement learning-based multi-target reservoir gate optimal scheduling method, which comprises the following steps of: constructing a reservoir drainage facility gate optimal scheduling model, and determining a target function and a constraint condition; establishing a hierarchical reinforcement learning framework, determining options, an option network and an option internal strategy, determining a reservoir water level by the option network according to information such as reservoir inflow to minimize the maximum reservoir outflow of the reservoir, and determining a gate operation scheme by the option internal strategy according to the reservoir water level determined by the option network to minimize gate adjustment times; and solving the layered reinforcement learning framework by using a competitive double-depth Q network algorithm to obtain a reasonable gate operation scheme. According to the method, gate scheduling is decomposed into two levels of decisions, the problems that the number of gate optimization scheduling decision variables is too large and the solution dimension is too large are effectively solved, a reasonable gate operation scheme can be obtained through rapid optimization, and the maximum output flow of a reservoir and the gate adjustment frequency are effectively reduced.

The invention provides a multi-target deployment rapid optimization method based on deep reinforcement learning. The invention innovatively provides a thought of processing a complex optimization problem by using the strong environment understanding ability and generalization ability of the deep reinforcement learning algorithm, and the model constructed based on the TD3 algorithm enables the intelligent agent to optimize the decision strategy through interaction learning with the environment. Preliminary experiments show that although the model can better understand the environment, the early-stage learning efficiency is low, and strategy convergence is slow. Therefore, the invention further provides a reinforcement learning model based on pre-training, a pre-training target is provided through a traditional optimization algorithm, the learning process is accelerated, and the convergence quality is improved. Experiments prove that the pre-trained agent is superior to an unpre-trained agent in convergence speed and quality.

The invention discloses a heuristic log board cutting optimization method driven by three-dimensional reconstruction, and relates to the field of wood processing and manufacturing. The invention provides a heuristic log board cutting optimization method based on a three-dimensional point cloud reconstruction model. The method specifically comprises the following steps: accurately reconstructing a log three-dimensional shape by using point cloud data acquired by a sensor, and converting the log three-dimensional shape into a two-dimensional plane projection by adopting dimension reduction processing; further, a heuristic search algorithm is constructed based on a graph theory to carry out contour extraction, so that the adverse effect of the complex geometric shape of the log on the outturn percentage is eliminated; on the basis, a simulated annealing algorithm improved on the basis of a skyline constraint rule is provided, rapid optimization of a cutting scheme is achieved through an intelligent search strategy, the calculation speed is greatly increased while the solving quality is guaranteed, and the aging requirement of large-scale production is practically met.

The invention discloses an anti-swing predictive control method for an offshore crane, and belongs to the technical field of ocean engineering equipment control. The problem that in the prior art, due to control lag and an inaccurate model, the load swing restraining effect of an offshore crane is poor is solved. According to the scheme, the method is characterized in that the state of a crane and prediction information of future waves are obtained in real time through a state sensor set and a multi-source environment sensing system; mixing the prediction model to predict a crane system state sequence under different control instructions in a future time domain in a rolling manner; based on the prediction sequence, solving a reference control track aiming at suppressing swing and reducing structural fatigue in upper-layer optimization, and solving and outputting an instant control instruction meeting the constraint of an execution mechanism in lower-layer rapid optimization; meanwhile, system health management is independently executed, and a control mode is dynamically adjusted according to evaluation. The method is mainly used for precise anti-swing control of the offshore crane under the complex sea condition, load swing can be effectively inhibited in advance, and operation precision and equipment safety are improved.

The invention discloses a service dynamic matching space-time-frequency resource scheduling method based on adaptive control, which belongs to the technical field of space-time-frequency resource scheduling, and comprises the following steps of: constructing a resource scheduling evaluation function, acquiring a service queuing queue, and coding by adopting a space-time-frequency resource to be scheduled on the basis of the service queuing queue; according to the method, multiple resource scheduling parameter codes are obtained, local cosine optimization, local joint optimization, local rapid optimization and global jump optimization are carried out on the resource scheduling parameter codes for multiple times to obtain the optimal resource scheduling parameter code, and space-time-frequency resource scheduling can be finally carried out according to the optimal resource scheduling parameter code. The resource arrangement utilization rate is improved, and network congestion can be effectively avoided.

The invention discloses an unmanned aerial vehicle trajectory optimization and bandwidth allocation method based on hierarchical deep reinforcement learning. The method comprises the following steps: (1) constructing an unmanned aerial vehicle assisted IoT node data collection system model; (2) based on the constructed system model, establishing an optimization problem of maximizing the data collection amount by adjusting the position and bandwidth allocation proportion of the unmanned aerial vehicle; and (3) modeling the optimization problem as a semi-Markov decision process, and training an intelligent agent capable of autonomously adjusting the position and bandwidth allocation proportion of the unmanned aerial vehicle by adopting a hierarchical deep reinforcement learning method. According to the method, a hierarchical deep reinforcement learning technology is utilized, an unmanned aerial vehicle position and bandwidth allocation rapid optimization algorithm is designed for an unmanned aerial vehicle assisted IoT node data collection scene, maximized data collection from environment ground IoT nodes is realized under the constraints of energy, speed and bandwidth of the unmanned aerial vehicle, and the scale is smaller while a training model can converge rapidly.

The invention relates to the technical field of computational fluid mechanics and biological microfluidic design, in particular to a model order reduction method for rapid optimization of a circulating tumor cell (CTC) sorting structure, which comprises the following steps of: constructing a three-dimensional full-order computational fluid mechanics model by collecting channel geometric structure parameters and fluid working condition parameters, and generating a training data set; organizing a flow field snapshot into a matrix form, extracting a dominant mode, constructing a low-dimensional modal space by taking the dominant mode as a base vector, establishing a low-dimensional ordinary differential model through Galerkin projection, and training a parameter-modal coefficient mapping relation by adopting a deep neural network to form a complete reduced-order model; and finally, constructing a multi-objective optimization problem based on the reduced-order model, carrying out optimization iteration by adopting an evolutionary algorithm, and returning an optimization result to the full-order model for verification, so that rapid optimization design of the CTC sorting structure is realized, and an effective solution is provided for intelligent design of a biomedical microfluidic device.

The invention discloses a decoupling cell-hybrid equivalent model and a permanent magnet flat wire motor multi-physical field cooperation rapid optimization method, a rotor and a stator are modeled separately, and complete separation of a motor topological structure and an analysis grid is realized at the rotor part through decoupling cells, so that the motor topological structure can be completely separated from the analysis grid when facing various different rotor topological structures. And a new grid structure does not need to be repeatedly constructed, so that the modeling efficiency, the model universality and the modeling flexibility of a complex structure are greatly improved. And the stator adopts hybrid equivalent modeling, so that the solving complexity is remarkably reduced while the high calculation precision is kept. The rotor side adopts boundary encryption subdivision to ensure stress precision, and the stator side adopts a parameterized equivalent magnetocaloric network model to reduce analysis time consumption, and unification of high precision and high efficiency is realized. By means of an NSGA-II algorithm, a bidirectional coupling calculation framework among an electromagnetic field, a temperature field and a stress field is constructed, efficient coupling iteration among physical quantities of the three fields is achieved through coupling calculation, and coupling prediction precision and design reliability are effectively improved.

The invention provides a three-dimensional flow field acquisition method and system based on two-dimensional through-flow and a neural network, belongs to the field of fluid mechanics calculation, and can at least partially solve the problem that in the prior art, calculation efficiency is low, and two-dimensional through-flow cannot reflect three-dimensional features. Geometric parameters and working condition parameters of fluid machinery are input, and flow parameters on a two-dimensional flow surface are rapidly output; constructing and training a neural network model, wherein the neural network learns a mapping relation between a two-dimensional flow surface calculation result and a three-dimensional flow field; for the fluid machinery under the target working condition, a result is obtained through two-dimensional through-flow calculation, the result is preprocessed and then input into the trained neural network, the low-dimensional representation of the three-dimensional flow field is output, and complete three-dimensional flow field parameters are obtained through reconstruction. The calculation period is remarkably shortened, and rapid optimization of multiple schemes is supported.

The present invention relates to the technical field of artificial intelligence devices, and provides a system and method for compiling a convolutional neural network model for an embedded device. The system comprises a model computation graph representation unit, a convolutional neural network fixed optimization module, a convolutional neural network automatic optimization module, an automatic optimization process module, a model compilation optimization unit, and a model deployment and reasoning unit. A convolutional neural network operator having the highest computational load is optimized, and only the edge cropping size and the loop unrolling step size of the convolutional neural network operator are optimized, so that the optimization space is reduced and the optimization time is shortened while optimizing the operator to the greatest extent; in addition, the entire process from model training output to embedded device reasoning is established, and rapid optimization and deployment of convolutional neural network models are achieved. By means of the method, a high-performance convolutional neural network model can be rapidly deployed on an embedded device, so that the present invention has high practical value and innovative value.

The invention relates to a chaotic variation metamaterial parameter rapid optimization method based on a random forest model. The method comprises the following steps: S1, determining a structural parameter search space of a metamaterial and upper and lower bounds of optimization variables; s2, an initial population is generated, fitness values of individuals of the initial population are calculated, the individuals are metamaterial structure parameters, and the fitness values of the individuals are obtained through prediction of a random forest regression model; s3, variation and crossover operation is carried out on the current population, new individuals are obtained, fitness values of the new individuals are calculated, and the variation operation is hyperchaotic variation through hyperchaotic mapping; and S4, performing selection operation according to the fitness values of the individuals, obtaining a new population, returning to S3 until a termination condition is met, and obtaining the optimized metamaterial structure parameters. According to the method, the design and optimization efficiency of the metamaterial can be remarkably improved.

The invention relates to the technical field of structure fatigue evaluation, in particular to a pipe truss structure fatigue analysis parameter optimization method and system based on deep learning, and the method comprises the following steps: constructing a deep learning proxy model fusing a multi-scale structure diagram encoder, a working condition parameter encoder and a damage dimension curve decoder; training the model by using a high-fidelity training data set; for the target truss structure and the load working condition, predicting a damage dimension relation curve through the model; determining the optimal dimension of a rain flow counting matrix by analyzing the convergence characteristic of the curve; and predicting a multi-working-condition damage value through a model, and screening out a dominant working condition based on a damage contribution percentage. According to the method, the deep learning agent model is used for replacing traditional iterative simulation, rapid optimization of key parameters in fatigue analysis is achieved, the efficiency and accuracy of fatigue evaluation of the large truss structure are remarkably improved, and the generalization ability of the model is improved by fusing the physical effect of a multi-scale structure chart encoder to capture key hotspots.

The invention provides a power system black-start partition decision method based on lightweight reinforcement learning, and the method comprises the steps: constructing a black-start model of a power system containing multiple physical constraints, and formalizing the black-start model into a Markov decision process, the Markov decision process comprises a state space, an action space, a reward function and a transfer process; decomposing the power system into a plurality of sub-regions by adopting a dynamic partition recovery strategy; based on the Markov decision process, a depth deterministic strategy gradient algorithm is adopted to design and train a lightweight strategy network of deep reinforcement learning; enabling the intelligent agent to interact with the environment in each sub-region through the lightweight strategy network of deep reinforcement learning, and learning an optimal decision strategy; and according to the optimal decision strategy, performing partition power supply capability recovery on the power system. According to the invention, black-start rapid optimization decision in a complex power grid environment is realized.