28 results about "Genetic operator" patented technology

This application discloses a method, apparatus, device, and medium for determining simulation scenarios, relating to the field of autonomous driving technology. The method for determining simulation scenarios includes: acquiring at least one initial scenario corresponding to a target vehicle; performing scenario testing on each initial scenario to obtain test boundary indicators for each initial scenario; using the test boundary indicators as fitness, employing genetic operators of a genetic algorithm to select, crossover, and / or mutate the scenario parameters of the initial scenarios to obtain at least one optimized scenario; updating the initial scenarios to optimized scenarios, and returning to perform scenario testing on each initial scenario to obtain test boundary indicators for each initial scenario, until a preset stopping condition is reached to obtain at least one candidate scenario; and identifying candidate scenarios with fitness greater than a preset threshold as target simulation scenarios that meet performance boundary testing requirements.

The invention discloses a vehicle path planning problem optimization method and system for landscape smoothing, and belongs to the technical field of combinatorial optimization algorithms, and the method comprises the steps: randomly generating an initial population for a vehicle path planning problem, constructing a toy problem, and carrying out the optimization of the vehicle path planning problem; the toy problem and the vehicle path planning problem are equal in scale, and the fitness landscape of the solution space has a single-peak characteristic; constructing a fitness function, and evaluating each individual in the population through the fitness function to obtain a fitness value of each individual; executing operation of a selection operator, a crossover operator and a mutation operator to generate a next generation population; and the fitness function design and evaluation and genetic operator operation are repeatedly executed until a preset termination condition is met, and an individual with the minimum original problem objective function value in the evolution process is output to serve as an optimal solution of the vehicle path planning problem. According to the method, local optimum can be jumped out, global exploration and local development are balanced, the solving quality and stability can be improved, and the method has applicability in vehicle path planning.

The invention relates to the technical field of ship navigation model tests, and discloses a giant estuary ship navigation capability prediction model test method fused with machine learning, which comprises the following steps: initializing and defining a scene vector, training an agent model, generating a population, and setting archives and parameters; executing dynamic fitness evaluation to obtain a fitness score; generating filial generations based on score selection and genetic operators and ensuring physical effectiveness; selecting a scene with the highest score to obtain a predicted risk; when the error is greater than a threshold value, storing the candidate scene; if yes, outputting the file; and if not, replacing the population and returning evaluation. According to the method, a feedback closed loop is constructed through failure judgment and updating of a failure mode file, after a failure scene verified by a high-fidelity simulator is stored in the file, the areas are avoided when the novel fitness is calculated through dynamic fitness evaluation, the search direction is automatically adjusted, and automatic and self-adaptive exploration of a high-risk scene is achieved.

The invention provides a power distribution area optimization operation method giving consideration to distributed energy characteristics and electric energy quality management requirements, and the method comprehensively considers the cooperative relationship among a photovoltaic power generation system, an energy storage system and power distribution network nodes, builds an electric energy quality evaluation index according to the voltage deviation and fluctuation of the nodes, proposes a light-storage cooperative operation mechanism, and improves the power quality management efficiency. A power distribution area double-layer optimization model including constraint conditions such as node power balance, energy storage operation, voltage and capacity limitation and power flow constraint is formed; a solving method based on an improved particle swarm algorithm is proposed, the search diversity is enhanced by introducing chaotic Tent mapping and genetic operators, the convergence speed is increased, local optimum is avoided, and the practicability of the proposed method is verified through an actual example. According to the method, the influence of distributed energy access on voltage deviation and fluctuation is fully considered, and the method has important significance in the aspects of optimized operation and electric energy quality control of a power distribution area.

The invention belongs to the technical field of multi-source solid waste cyclic utilization, and discloses a multi-source solid waste cyclic utilization path optimization method and system, and the method comprises the steps: solving a multi-target mathematical optimization model of a multi-source solid waste cyclic utilization system through employing a dual-stage multi-target optimization method, and obtaining a multi-source solid waste cyclic utilization optimization scheme set; selecting a preference-based solid waste optimal processing path from the optimization scheme set; in the first stage, an extreme weight vector and a center weight vector are constructed to search a heuristic solution, and in the second stage, further evolution is carried out on the basis of a heuristic solution population, so that the population has better convergence and diversity; in the second stage, deep reinforcement learning is introduced to adaptively optimize parameters of a genetic operator. Meanwhile, the invention also provides a dynamic switching method based on population fitness to avoid unnecessary waste of computing resources. According to the method, the problems of low algorithm convergence speed, difficulty in determining important parameters of genetic operators and the like in practical application are solved, and efficient optimal configuration of path optimization is realized.

The invention discloses a cancer genome variation prediction system based on multiple genetic operators, relates to the technical field of genetic variation detection, and aims to solve the problems that the traditional cancer genome variation prediction is mainly based on a static mutation rate and a single mutation type, the dynamic balance and nonlinear evolution of a tumor in mutation, selection and clone expansion processes are difficult to accurately reproduce, and the accuracy is poor. According to the cancer genome variation prediction method and the cancer genome variation prediction system, high-fidelity simulation and evolution reconstruction of the cancer genome are realized by fusing a multi-type mutation mechanism, fitness feedback and a polyclonal competition strategy. According to the method, through the synergistic effect of multiple genetic operators and dynamic parameter adjustment, the dynamic balance and nonlinear evolution of the tumor in the mutation, selection and clone expansion process can be accurately reproduced, and then the accuracy of cancer genome variation prediction is improved. The application provides a brand new technical approach for early screening of cancers, variation detection and precise medical treatment.

The invention discloses a multi-objective optimization lithium battery thermal runaway escape gas spectral wavelength selection method, which comprises the following steps: collecting near infrared spectral signals in sample gas to obtain spectral absorbance and component concentration, defining the wavelength as a decision variable, and carrying out hybrid coding based on a SparseEA double-layer coding mechanism. Selecting an optimal wavelength combination subset by taking the weight threshold as a reference standard of a forward heuristic search strategy of the sequence; then, a SparseEA genetic operator based on a weight value serving as a fitness value is used for guiding population evolution to generate filial generations, and a Pareto optimal solution is output after environment selection; and finally, determining an optimal compromise solution of gas spectral quantitative analysis in the Pareto optimal solution by using GEMONN, thereby obtaining an optimal spectral wavelength variable combination scheme. More key wavelength variables are intelligently screened out through the SFS strategy, a high-performance spectrum quantitative analysis model is constructed in a multi-objective mode, and the accuracy of multi-component spectrum feature selection is greatly improved.

The invention relates to an energy internet system scheduling model solving method based on a niche genetic algorithm, and the method comprises the following steps: S1, constructing a multi-objective optimization mathematical model containing an economic cost objective function and an environmental protection cost objective function at the same time, and determining constraint conditions in the operation of a scheduling system; s2, representing candidate solution individuals by adopting a real number coding mode; s3, calculating fitness values of all individuals in the current population, and entering a main optimization iteration process until a termination condition is met; s4, in each round of iteration, according to the current individual fitness value, executing genetic operator operation to generate a next generation of filial generation population; s5, by calculating the normalized Euclidean distance between individuals, only the individual with the optimal local fitness is reserved; and S6, outputting the individual with the highest fitness value as the optimal solution of the final scheduling model parameters of the multi-energy complementary system. According to the method, the overall economy, the environmental protection property and the operation reliability of the scheduling scheme can be remarkably improved, and the method has good engineering applicability and popularization value.

The application provides a sparse antenna array optimization method and device fusing a neural network and a genetic algorithm, and the method comprises the following steps: initializing a genetic algorithm model; determining the fitness values of different individuals in an initial population based on a neural network model; judging whether a set termination condition is reached, and outputting an array element position distribution optimization result of the sparse antenna array in the case where the set termination condition is reached; otherwise, updating the population based on genetic operators and catastrophe operators, and determining the fitness values of different individuals in the updated population based on the neural network model, and repeatedly updating the population and determining the fitness values of different individuals in the updated population until the set termination condition is reached, and outputting the array element position distribution optimization result of the sparse antenna array. Therefore, the efficiency of genetic algorithm calculation can be improved, and the algorithm performance of the sparse antenna array is improved.

The application provides a feature selection method, device, equipment, medium and product. In an initialization stage, a multi-objective optimization problem is decomposed into multiple single-objective optimization sub-problems, initial solutions are randomly sampled in a search space to generate initial solutions, the initial solutions are actually evaluated, and the initial solutions are stored in an archive. In a cyclic processing stage, a multi-objective kernel ridge regression model is trained by using the archive, solutions suitable for each sub-problem are selected from the archive as initial parent populations of the sub-problems, genetic operators are used to obtain child populations, the multi-objective kernel ridge regression model is used to predict the prediction accuracy and reasoning time of the child feature solutions, the parent populations are updated according to the prediction results, and new solutions for actual evaluation are selected from the evolved populations by using a sparsity-driven filling sampling criterion to update the archive. Through sub-problem collaborative optimization assisted by a proxy model, non-dominated solutions of the original multi-objective optimization problem are obtained, and the prediction accuracy, calculation efficiency, convergence and diversity are improved.

The invention relates to the technical field of asphalt mixture modulus prediction, and discloses an asphalt mixture multi-scale modulus prediction method and system, and the method comprises the steps: obtaining a three-scale joint data set, determining the design stage weight of each scale data, and outputting a design stage dynamic modulus prediction value through a trained first prediction model. The system corresponds to the method. According to the method, the prediction precision of the design stage is remarkably improved by constructing the three-scale data system, comprehensively capturing the modulus influence factors of the asphalt mixture and combining the dynamic weight distribution mechanism adaptive to the stage; through a cross-stage constraint function driven by data quality, the difference from design to service data quality is quantified, and the problem of performance disjunction of the two is solved. A transfer learning mechanism is constructed through a genetic operator, the dynamic change of data quality distribution is adapted, and the prediction stability and adaptability in the service stage are improved; therefore, full-life-cycle modulus accurate prediction is realized, and scientific support is provided for mixture ratio optimization and pavement maintenance decision making.

This invention relates to the field of scheduling optimization combining local tabu search strategy and genetic algorithm, specifically a workshop flexible operation scheduling method based on tabu search genetic algorithm. It aims to generate flexible operation scheduling schemes and improve production efficiency by optimizing the maximum completion time. The invention consists of two parts: a global search phase, which divides the basic elements of the genetic algorithm into chromosome encoding / decoding, population initialization, setting iterative genetic operators, and population constraint rules, and performs a fast parallel search in the population space to obtain uniformly distributed feasible solutions; and a local tabu search phase, which establishes a tabu list, sets the tabu search length, and performs iterative search for locally optimized solutions, eliminating repetitive work and avoiding premature entrapment in neighborhood optima. This method combines the advantages of genetic algorithm and tabu search, improving the search efficiency in the population space and optimizing the maximum completion time within a specified number of iterations, making it suitable for guiding flexible operation workshop production.

This invention belongs to the field of software refactoring and serverless computing technology, specifically relating to a fine-grained service decomposition and dynamic deployment method for object-oriented applications in edge environments. By constructing a method call graph through static analysis, and for the first time applying a hybrid heuristic algorithm combining particle swarm optimization and genetic operators to this graph partitioning problem, it can intelligently find a decomposition scheme that minimizes cross-function communication overhead while satisfying the limited resource constraints of edge nodes. This fundamentally solves the contradiction between computing resources and communication latency that traditional decomposition methods struggle to balance. Furthermore, by introducing Function-as-a-Service (FaaS) proxies and wrappers, it achieves automated refactoring of the original object-oriented code, enabling seamless migration of tightly coupled local calls to a distributed FaaS environment. This significantly reduces the technical threshold and operational complexity of application migration, achieving fine-grained resource utilization and elastic scaling.

The invention discloses a multi-unmanned aerial vehicle formation flight fault assessment method and system based on fuzzy testing, and belongs to the technical field of unmanned aerial vehicles. The method comprises the following steps: constructing an initial population formed by an unmanned aerial vehicle state-action sequence; performing crossover mutation operation on the population by adopting a genetic operator to generate a new generation of population; performing clustering mapping on the original state in the population to form an abstract population of an abstract state-action pair; calculating the multi-target fitness of the abstract individuals, wherein the multi-target fitness comprises round reward, prediction fault probability and action certainty level; performing non-dominated sorting on the fitness by adopting a multi-objective optimization algorithm, screening elite individuals, and updating the elite individuals to an archive set; and iteratively executing the steps until a termination condition is met, and outputting a high fault probability individual set. Through state space compression and multi-objective optimization screening, the problems of state space explosion and low fault case generation efficiency of an existing test method are effectively solved, and the reliability test efficiency of the unmanned aerial vehicle formation control system is remarkably improved.

The invention provides a feature selection method and device, equipment, a medium and a product. In an initialization stage, a multi-target optimization problem is decomposed into a plurality of single-target optimization sub-problems, random sampling is carried out in a search space to generate an initial solution, the initial solution is really evaluated, and the initial solution is stored in an archiving and circular processing stage; training a multi-target kernel ridge regression model by using the archive, selecting a solution adaptive to each sub-problem from the archive as an initial parent population, obtaining a filial generation population by using a genetic operator, performing prediction of prediction precision and reasoning time on a filial generation feature solution by using the multi-target kernel ridge regression model, and updating the parent population according to a prediction result to obtain a new filial generation feature solution; and selecting a new solution for real evaluation from the evolved population by adopting a sparsity-driven filling sampling criterion, and updating and archiving the new solution. Through collaborative optimization of the sub-problems assisted by the agent model, a non-dominated solution of an original multi-objective optimization problem is obtained, and the prediction precision, the calculation efficiency, the convergence and the diversity are improved.

The application provides an operation and maintenance management method and platform of a photovoltaic power generation system, relates to the technical field of photovoltaic power station operation and maintenance, and comprises the following steps: obtaining a set of operation and maintenance tasks to be processed and a set of idle operation and maintenance personnel; scheduling and optimizing the operation and maintenance tasks by using an improved NSGA-II algorithm, establishing a multi-objective optimization problem of minimizing total operation and maintenance cost, minimizing total task completion time and minimizing response time, using a clustering enhanced coding structure, a hybrid population initialization strategy and a probability weighted genetic operator to solve the multi-objective optimization, and outputting a Pareto optimal solution set; selecting a final scheme from the Pareto optimal solution set and converting the final scheme into an executable format, and outputting a complete operation and maintenance scheduling scheme including a personnel scheduling list, path planning instructions, a material and tool list and a general time schedule. The application can realize multi-objective collaborative optimization of operation and maintenance cost, response time and execution efficiency.

Computer-automated tools to assist machine learning model development may be suitable for users without domain expertise or specialized expertise in particular model types. An initial grid search via gradient boosting is applied to a source data set and target to generate a reduced search space including prioritized set of features ranked by importance value. A genetic algorithm searches the supervised learning classification problem space with various model choices and feature engineering. AI model training techniques may include machine learning test models, test hyperparameters, and network architectures. Feature engineering generates new features, and model development promotes features with fitness scores satisfying a defined threshold. Feature engineering may employ feature combinations, scaling, normalization, and feature convolutions. ML model development may apply a genetic operator including one or more of crossover, mutation, and selection. Iterative test ML model development generates pre-trained models that can yield code for deployment of production models.

The invention provides a dynamic compilation method, system and equipment for multiple tasks of an unmanned aerial vehicle and a medium. The method comprises the following steps: constructing a killing chain element path model through task logic of the unmanned aerial vehicle to obtain killing chain element path expressions, and grouping high-correlation killing chain element path expressions through feature vector construction and cosine similarity calculation to obtain task groups; based on the degree of adaptation of the unmanned aerial vehicle and the task group, generating an initial population, constructing a fitness function, and evaluating the individual fitness of the initial population, thereby improving the proportion of an initial effective solution, reducing the elimination cost of an invalid solution in the earlier stage of iteration, and accelerating the convergence efficiency; iterative optimization is carried out by adopting a genetic operator method and reinforcement learning based on individual fitness, the problem of poor adaptability in different scenes caused by manual parameter adjustment of a traditional genetic algorithm is solved, the applicability of an optimization result is improved, and multi-task compilation of the unmanned aerial vehicle can be dynamically provided in real time.

The application relates to a shared bicycle parking point location method and system based on multi-point joint distribution, which comprises the following steps: acquiring information of multiple demand points and multiple alternative parking points, and determining optimization targets and / or constraint conditions of a non-dominated sorting genetic model of multi-point joint distribution according to the information; generating an initial population of the non-dominated sorting genetic model based on a binary and integer combined chromosome coding mode; calculating non-dominated levels and crowded distances of each chromosome; screening the initial population by using a tournament selection method, and performing crossover and variation on the screened population based on a position vector crossover operator and a mutation operator to obtain a child population; and iteratively processing the parent population based on an elite strategy and the non-dominated levels and the crowded distances until an optimal solution meeting one or more constraint conditions is found. The application establishes an NSGA-II model based on multi-point joint distribution, and optimizes genetic operators in the model, so that the practicability of the model is improved.

The application discloses a data processing method and device, equipment and storage medium, and relates to the technical field of computers, and comprises the following steps: initializing a population to obtain a current population with a function to be evaluated as an individual; a gene expression programming is improved by using a CUDA architecture and a linear table-based coding mode in advance to obtain a target population evolution algorithm; an ADF gene for enhancing individual expression is used in the algorithm; all individuals are subjected to parallel evaluation processing based on the CUDA architecture in the algorithm and a predefined kernel function to determine the current fitness of each individual, and it is judged whether the preset evolution termination condition is met at present; if not, the individual whose current fitness meets the preset elimination condition is eliminated, the remaining individuals are processed based on the preset genetic operator in the algorithm to generate a new current population, and the step of parallel evaluation processing is rejumped; and if yes, the individual solution is output. The application improves the population evolution speed and reduces the error of the found function.

The invention relates to the technical field of artificial intelligence, in particular to a method and system for predicting the aging trend and residual life of a proton exchange membrane fuel cell and a storage medium. Adaptively sliding along a time dimension by using a variable-scale convolution kernel, sampling through adaptive pooling operation, and capturing a discriminative short-term aging time pattern; an aging feature attention mask is introduced, attention calculation of a redundant stationary section is shielded, meanwhile, a multi-head self-attention head number distribution strategy is optimized, and aging dependency features of the proton exchange membrane fuel cell at different time scales are adapted; taking prediction error minimization and feature redundancy minimization as a common target to construct a fitness function to optimize the aging trend features of the hyper-parameter and the proton exchange membrane fuel cell; and proposing an adaptive genetic operator strategy to select aging trend characteristics related to the aging trend in the time sequence data of the proton exchange membrane fuel cell. The invention aims to solve the problem of how to improve the prediction precision of the proton exchange membrane fuel cell.

The application discloses a multi-heterogeneous unmanned aerial vehicle system cooperative task allocation method based on a multi-objective evolutionary algorithm considering conditional probability, which comprises the following steps: firstly, according to the enemy target information, the current battlefield situation, the combat risk and the combat resources of the self side, a target function is set based on conditional probability, constraint conditions are given, and a multi-objective optimization model of the heterogeneous unmanned aerial vehicle cooperative task allocation is established; secondly, corresponding genetic operators are selected based on the actual combat situation, and an improved multi-objective optimization algorithm is used to solve the multi-objective optimization model of the cooperative task allocation; and thirdly, a decision maker selects a solution from a Pareto solution set according to the preference for the target function, and the corresponding task allocation scheme is taken as an execution scheme. The application is suitable for the multi-heterogeneous unmanned aerial vehicle cooperative task allocation under the condition of multiple combat resources, can provide a task allocation scheme which is suitable for the actual combat situation for the pre-war task allocation, and has a certain significance for the research on the multi-objective optimization problem of the multi-heterogeneous unmanned aerial vehicle cooperative task allocation.

The application discloses an unmanned ship formation control optimization method based on a double-channel reinforcement learning framework, aiming at solving the problems of existing adaptive dynamic programming control parameter setting difficulty, traditional genetic algorithm easy to fall into local optimum and multi-dimensional error physical quantity dimension not unified. The application constructs an unmanned ship formation mathematical model and a bottom online control channel, and constructs an upper offline optimization channel; a dimensionless fitness function based on state energy ratio is designed; the upper channel utilizes a reinforcement learning intelligent agent to dynamically decide a genetic operator selection strategy according to a population evolution state to generate a control parameter, the bottom channel utilizes the parameter to perform simulation and feeds back a performance index to the upper channel to update a decision model, and the optimal control parameter is output through double-channel closed-loop iteration. The application improves control parameter optimization efficiency and precision, eliminates the interference of order of magnitude difference on optimization, and significantly enhances the dynamic response speed, steady-state tracking precision and adaptive capacity of the unmanned ship formation system.

The application discloses a high-speed maneuvering coherent accumulation method based on GA (genetic algorithm) and GRFT (generalized Radon-Fourier transform), which comprises the following steps: after the echo signal of a high-speed maneuvering target after pulse compression is subjected to symmetric windowing processing, the generalized Radon-Fourier transform is carried out after the blind speed sidelobe is suppressed, and the parameter searching process is optimized by using the genetic algorithm, and finally the target detection result is output. The GA comprises the following steps: the basic parameter range and step length are determined, the population is initialized and the fitness value is calculated, the population is updated through the genetic operator operation, and the maximum fitness value and the corresponding best individual are obtained. The application combines the GA with the GRFT, effectively reduces the algorithm complexity under the premise of guaranteeing the detection performance, and thus the coherent accumulation of the high-speed maneuvering target is completed.

The invention discloses a no-waiting flow shop bad scene set robust scheduling method based on a memetic algorithm, and the method comprises the steps: carrying out the processing time description of a to-be-processed workpiece set and a processing machine set through employing a discrete multi-scene method, defining a threshold value bad scene set according to a preset performance threshold value, and building a robust optimization model of the threshold value bad scene set. According to the method, a threshold value bad scene set robust scheduling model of a wait-free flow shop is established, and an optimization objective function is determined. An initial population is generated by using a scene-based NEH heuristic algorithm. And performing global search on the current population through a genetic operator and a selection mechanism, selecting a target scene by using a strategy related to the model, constructing a neighborhood structure by using knowledge related to problems in the target scene, and performing local search. The whole optimization process is circularly executed under iteration control until the preset maximum iteration number is reached, and finally the optimal scheduling solution in the optimized population is output. The method can effectively cope with the uncertainty of the processing time, improves the robustness and optimization efficiency of no-waiting flow shop scheduling, gives consideration to the global search capability and the local refinement capability, and achieves the quick solving of a high-quality scheduling scheme.

The invention discloses a discrete genetic operator NSGA-II-based power grid optimization scheduling method, which overcomes the problems of low power grid operation efficiency and low power grid scheduling efficiency in the prior art, and comprises the following steps: constructing an energy layering and partitioning framework, and establishing an intra-region equipment model; constructing an upper-layer global optimization model by taking the lowest network loss of the global power transmission network, the optimal economy of the regional power distribution network and the lowest voltage offset degree as objective functions; constructing a lower-layer local optimization model by taking maximization of the utilization efficiency of the distributed energy and minimization of the scheduling cost as objective functions; and solving the optimization model by using a multi-target genetic algorithm based on a discrete genetic operator, and carrying out power grid optimization scheduling according to a solving result. And the power grid operation efficiency and the resource configuration rationality are improved.

This invention discloses a CPGA-SMCMN method for identifying single-drive pathways in cancer, comprising the following steps: 1. Setting up an SMCMN model; 2. Setting up CPGA: 2.1) Clustering; 2.2) Chromosome coding and initial population; 2.3) Fitness function; 2.4) Genetic operators; 2.5) Specific CPGA process. This method can identify more biological information, has strong scalability and practicality, and can identify more genes enriched in important drive pathways.