147 results about "Evolutionary algorithm" patented technology

The application provides a star group autonomous cooperative search and tracking method based on dynamic information gain, which comprises the following steps: acquiring satellite constellation parameters and target initial state information; long-time sequence tasks are decomposed into continuous short-time sequence decisions based on a closed-loop rolling horizon planning framework, the target state distribution is predicted to obtain the uncertainty degree and generate an atomic task candidate set in each rolling period; the tracking or search mode is adaptively selected according to the uncertainty degree and the value is evaluated; the optimal task scheme is generated based on a multi-objective evolutionary algorithm; and the observation is performed and the target state is updated according to the observation result to enter the next period. The application dynamically adapts to the target uncertainty evolution through the closed-loop rolling horizon planning, avoids tracking rupture, uniformly measures the value of search and tracking through adaptive dual-mode cooperative decision, solves the value conflict problem, realizes the space-time optimal configuration of star group resources through the multi-objective evolutionary algorithm, and improves the continuous tracking capability and resource utilization efficiency.

This invention discloses a multi-parameter coupling-based deep peak-shaving full-condition optimization control method for coal-fired power units. Addressing the problems of existing technologies, this invention constructs a multi-parameter coupling index system, calculates the comprehensive coupling strength between each parameter and the load, and selects core parameters. Dynamic clustering is used to finely divide the full-condition operation from 20% to 100% rated load into multiple operating zones. An LSTM-GRU dual-channel deep learning prediction model is established for each operating zone. A three-layer hierarchical control architecture is designed, comprising load optimization scheduling, boiler-turbine coordination control, and combustion optimization execution. The boiler-turbine coordination layer uses multivariate model predictive control and online self-tuning of multi-objective weights through fuzzy inference. The combustion execution layer uses the NSGA-III multi-objective evolutionary algorithm to optimize air distribution. A constraint-adaptive tuning mechanism based on safety margin is established to form a closed-loop iterative optimization across all operating conditions. This invention can significantly improve the load response speed, main parameter stability, and combustion economy of the unit during wide-load operation.

Embodiments of the present application disclose a method, device, equipment, medium and program product for generating an adversarial sample. The method comprises: obtaining a preset code file; determining a non-conditional jump instruction in the preset code file; modifying the non-conditional jump instruction into a reconstructed jump instruction according to a preset instruction modification rule; performing control flow flattening processing on a basic block of the preset code file; based on a result of the control flow flattening processing, inserting a preset application programming interface (API) sequence into the preset code file according to a preset rule to obtain a target code file containing a malicious code feature; inserting noise into the target code file to obtain an initial sample; and performing iteration on the initial sample according to a preset evolution algorithm to obtain an adversarial sample. The embodiments of the present application can generate a large number of malicious code adversarial samples that can evade antivirus software and sandboxes.

This invention belongs to the field of power grid fault diagnosis technology and discloses a power grid fault diagnosis method and system based on differential evolutionary logic operations. The method includes: collecting the action information of protection devices and circuit breakers in the power system; filtering out a set of candidate elements that have experienced faults based on preset association rules between faulty elements and protection / circuit breaker actions; establishing a 0-1 integer programming model based on the determined protection action information, circuit breaker action information, and candidate element set; and solving the 0-1 integer programming model using an improved differential evolutionary algorithm based on logic operations to determine the fault state of the candidate elements. This invention directly encodes individuals using binary, eliminating the need for floating-point to binary conversion. It also constructs a binary mutation operator and an improved crossover operator based on logic operations and employs adaptive parameter adjustment to balance the population diversity and search efficiency of the algorithm.

This invention relates to the field of water resource optimization scheduling technology in water conservancy and hydropower engineering, specifically to a reservoir optimization scheduling method and related products based on intertemporal sensitivity sparse repair. The method involves constructing and solving implicit equations for power generation and water volume, establishing an evaluation system and calculating intertemporal sensitivity to obtain a set of key time periods. Then, sparse repair is implemented on the initial power generation plan within this set, and the repair operation is embedded in an iterative optimization framework. This invention avoids the use of empirical estimation or coarse trial calculations in traditional reservoir scheduling, enhances the water resource safety net of reservoirs, suppresses the oscillation phenomenon of traditional intelligent evolutionary algorithms, and maintains the scheduling intent and temporal consistency of the initial predetermined plan on the power grid side.

This invention discloses a flexible scheduling method, system, and application for UPMS (Upright Manufacturing System) workshops based on an improved differential evolution algorithm. The method includes: establishing an independent parallel machine scheduling model with sequentially dependent mold-changing time as the objective of minimizing the maximum completion time; using real-valued vector encoding and decoding it into a scheduling scheme through the maximum position value rule; performing mutation and crossover operations of the differential evolution algorithm on the generated test vectors, with optimizations including load balancing optimization based on destruction and reconstruction and mold-changing optimization based on sequential smoothing; finally, updating the population and outputting the optimal scheduling scheme. This invention significantly reduces sequentially dependent mold-changing time, achieves dual optimization of load balancing and mold-changing cost, and improves solution accuracy and efficiency, making it particularly suitable for discrete manufacturing workshops such as textile printing and dyeing, and injection molding.

This invention discloses a lightweight adaptive optimization method for high-dimensional hyperparameters of convolutional neural networks (CNNs) for non-stationary time-series signal classification. It aims to address the technical challenges of performance degradation in time-series signal classification models and the reliance on expensive real-world evaluations for hyperparameter configuration under non-stationary perturbation scenarios. This method uses a deep convolutional neural network as the core classification carrier, treating the hyperparameter combinations within the deep convolutional neural network as decision variables to be optimized. With robust classification error rate, computational complexity, and training time as core optimization objectives, it constructs a closed-loop collaborative optimization mechanism of "perception-evaluation-decision" and utilizes a meta-learning dual-branch convolutional polynomial surrogate-assisted evolutionary algorithm (MetaDCP-SAEA) to achieve efficient configuration. This method requires no manual intervention; the convolutional neural network used for classifying non-stationary time-series signals can automatically search for the optimal hyperparameter combination. In simulated non-stationary noise environments, the reduction in classification accuracy can be controlled within 9.17%. It is suitable for robust classification scenarios of non-stationary time-series signals such as industrial IoT monitoring and medical signal diagnosis. It helps to lower the engineering threshold of artificial intelligence technology, promotes the large-scale application of automatic machine learning in complex environments, and has broad market prospects and application value.

The application discloses a kind of carbon dioxide foam fracturing fluid gas injection parameter multi-objective optimization method and device, wherein, method includes: obtaining target formation geologic parameter and engineering constraint parameter and carrying out normalization processing;Normalize geologic parameter is converted into dynamic constraint condition associated with equipment pressure resistance capacity, to dynamically adjust gas injection pressure upper limit;Multi-objective function is constructed with fracture complexity, flowback rate and carbon emission as optimization target;With dynamic constraint and engineering constraint as boundary, improved multi-objective evolutionary algorithm is used to solve function, which generates uniformly distributed reference points in target space and uses reference point correlation strategy for population evolution, to generate Pareto optimal solution set;According to engineering constraint, the solution set is filtered to obtain the feasible optimal solution set, and the gas injection pressure, discharge capacity and carbon dioxide phase proportion are real-time closed-loop controlled according to the optimal solution set.The application realizes collaborative optimization, and significantly improves fracturing effect and low-carbon performance.

The present application relates to a kind of multi-band adaptation PDN cross-level optimization method, system, equipment and medium, method is by to the joint parameterization processing of the designable parameter of chip, package and board level, generates uniform design vector;A multi-objective optimization function set containing wide-band impedance performance, cost and area is constructed;Through sampling and simulation, training data is generated, and deep neural network proxy model is trained to quickly predict design performance;The proxy model is used to drive multi-objective evolutionary algorithm, and the best trade-off solution set, i.e. Pareto frontier, of performance, cost and area is automatically searched;High-fidelity verification is carried out on the frontier solution, and the final design scheme is selected according to the actual constraint.The method realizes the change from manual trial and error to automatic optimization, effectively solves the problems of insufficient wide-band impedance control, difficulty in automatic trade-off of multi-objective and cross-level design fragmentation in the prior art, and significantly improves the efficiency and effect of PDN design.

The application relates to a low-friction design method of a planetary roller screw, and belongs to the technical field of mechanical design. Step 1, a contact point position solving model and a kinematics model of the planetary roller screw are established, and the sliding speed and the entrainment speed of a screw-roller and a roller-nut contact pair at a contact point are solved; step 2, according to the definition of a slide-roller ratio, a slide-roller ratio calculation model of the screw-roller and the roller-nut contact pair of the planetary roller screw is established; step 3, according to the correct movement and meshing requirements of the planetary roller screw, a structure constraint condition is established; step 4, taking the minimization of the slide-roller ratio of the screw-roller and the roller-nut contact pair as an optimization target, and combining the structure constraint condition, a low-friction design model of the planetary roller screw is established; step 5, a multi-objective evolutionary algorithm is adopted to calculate the low-friction design model of the planetary roller screw, and optimal solutions and corresponding structure design parameters are obtained.

The application discloses an unmanned aerial vehicle countermeasure interception system based on flight path prediction, relates to the technical field of low-altitude safety defense, acquires flight parameters of a target unmanned aerial vehicle, constructs a prediction model and generates a predicted path in combination with environment and terrain data, identifies a path segment most likely to enter a forbidden area, and calculates an expected intrusion time; candidate interception areas are screened based on a Hamilton-Jacobian inverse reachable set, and a multi-objective evolutionary algorithm is used to select an optimal interception node; in combination with a device response radius and preparation time, a starting time is determined, and an interception device is deployed; when the target enters an interception threshold range, a physical countermeasure action is triggered; the method realizes early identification, intelligent deployment and high-precision interception of the target unmanned aerial vehicle, improves the real-time performance and effectiveness of a low-altitude safety system, and has good engineering adaptability.

The application discloses an edge intelligent body flexible scheduling method based on Lyapunov optimization and Stackelberg game. The application constructs a cloud-edge-end three-layer edge intelligent body federated learning network architecture to complete distributed collaborative training under the premise of protecting data privacy. In view of the problems of heterogeneous edge intelligent body network nodes, non-independent and identically distributed data and limited communication resources, a long-term joint optimization model of client selection and bandwidth allocation is established, and model accuracy, time delay and energy consumption are considered. Lyapunov optimization is used to decouple long-term random optimization into real-time decision-making per time slot, and to convert constraints into virtual queue stability control. Dynamic client selection is realized based on Stackelberg game, and high-quality nodes are adaptively selected. Adaptive bandwidth allocation is completed through an evolutionary algorithm, and communication requirements of key clients are preferentially guaranteed. The application can significantly improve model accuracy, reduce time delay and energy consumption, guarantee long-term stability of the system, and is suitable for high-dynamic edge intelligent body federated learning scenes.

This invention belongs to the field of multi-objective evolutionary neural architecture search, proposing a multi-objective evolutionary neural architecture search method that integrates convolution and self-attention. By constructing an improved multi-objective evolutionary algorithm, the efficiency of neural architecture search is increased, allowing for faster discovery and preservation of individual neural architectures during the search process, while balancing convergence and diversity. By constructing a search space that integrates convolution and attention operations, the processing capability of global and local information in image processing tasks is improved, enhancing the plasticity of the search space and the generalization performance of individuals across different task sets. Decision variables such as convolution kernel size and number of layers in the search space are encoded as individuals in the population of the improved multi-objective evolutionary algorithm. The optimal individual is searched through the operation of the improved multi-objective evolutionary algorithm and decoded into the optimal network architecture. This invention effectively improves search efficiency and obtains the best-performing neural architecture.

This invention discloses a cross-scenario collaborative optimization method for a two-sided dismantling line for waste household appliances, belonging to the fields of intelligent optimization algorithms and dismantling line balancing. The invention establishes a balanced optimization model for a two-sided dismantling line for waste household appliances, obtains the optimization objective function of minimizing the cycle time and minimizing the total task adjustment amount during the dismantling process, and constructs a multi-objective cross-scenario collaborative evolutionary algorithm to solve it. This significantly improves the allocation efficiency of dismantling tasks under multi-objective optimization and generates an optimized scheduling scheme that can balance cycle time and total task adjustment amount.

The application provides a pressure dynamic scheduling method for a pneumatic conveying pipeline with scalability, which combines evolutionary multi-objective optimization and machine learning to solve the dynamic scheduling problem of the pneumatic conveying system. The application regards the pressure dynamic adjustment problem of the pneumatic conveying pipeline as a dynamic (static) multi-objective optimization problem, reuses the high-quality solution found in the solving process to generate a better initial population, and thus can significantly improve the performance of the evolutionary algorithm when the demand of the pneumatic conveying system changes, so that the control system has better pressure adjustment capability.

The application provides a ground source heat pump air conditioner intelligent energy-saving control method and system, belongs to the technical field of building energy saving and intelligent control, collects historical load data and environmental parameters of building operation, and constructs a time sequence feature set; an initial quantum population is generated based on Tent chaotic mapping, population observation and updating are carried out in combination with an improved quantum genetic algorithm, and differential variation and crossover operation of individual fitness are carried out by using an improved self-adaptive differential evolution algorithm, so that the hyperparameters of the IADE-IQGA-BiLSTM model are optimized, multiple rounds of iterative training are carried out by using a training set, and an early stopping strategy is implemented by using a verification set; the trained model is used for load prediction of a test set, the prediction result is inversely normalized into an actual load value, and a performance evaluation index is calculated; the IADE-IQGA-BiLSTM model is packaged as a RESTful API interface and deployed on an intelligent energy-saving control software to receive real-time data flow and regularly generate future load prediction; based on the load prediction result, the water pump frequency and the heat exchanger shape of the ground source heat pump system are dynamically adjusted.

The embodiment of the present disclosure provides an information adaptive fusion industrial process energy consumption optimization method, which belongs to the technical field of computing, and specifically comprises the following steps: step 1, obtaining key process parameters related to energy consumption and data related to energy consumption cost in an industrial process to form industrial process characteristics; step 2, taking minimum energy consumption and minimum energy consumption cost as targets, constructing a constraint condition with the industrial process characteristics, and establishing an industrial process energy consumption optimization model; and step 3, solving the industrial process energy consumption optimization model by using an information adaptive fusion multi-objective evolutionary algorithm to obtain a plurality of groups of optimal key process parameters with the lowest energy consumption and energy consumption cost. Through the scheme of the present disclosure, the optimization efficiency and environmental protection are improved.

The application provides a kind of multi-dimensional sound scene ecological element space layout optimization method and system, belongs to mechanical vibration measurement technical field field, this method includes integration three-dimensional point cloud and environmental parameter construction digital basement, establishes the native plant model containing growth and acoustic characteristics, generates the space sound effect contribution degree pressure field representing the noise reduction potential by reverse sound wave ray tracing, constructs the plant growth agent driven by the pressure field, runs the symbiotic simulation engine to generate the morphological dynamic evolution sequence, calculates the acoustic performance, ecological stability and landscape robustness index, and uses it as the fitness evaluation basis of evolution algorithm, iteratively optimizes the initial planting layout, the application realizes the coupling of acoustic physical simulation and plant growth competition logic, is used to guide the long-term planning and species configuration of ecological garden, improves the sound environment quality and succession stability of landscape.

The application discloses a JPEG image multi-carrier steganography method based on a monkey king evolution algorithm, which is used for embedding secret information into images and comprises the following steps: processing N images according to a user input password and an error correction matrix H based on a (7, 4) Hamming code to obtain N carrier images, wherein a matrix of the i-th carrier image is H i ′, i = 1, 2,..., N; converting secret information to be embedded Secret into a binary form, and encrypting to obtain a binary sequence M = (m1, m2,..., m L L is the length of the binary sequence M; taking the minimum difference between a carrier image feature set and a stego image feature set as an objective, using a monkey king evolution algorithm to perform non-uniform dynamic allocation on the load size of each carrier image, and according to the allocation result, cutting and writing the secret information into the corresponding carrier image to obtain a stego image set. The application can solve the problem of efficient allocation of secret information among multiple carrier images, so that the stego image is least detected.

The application discloses a kind of based on data-driven small modular prismatic high temperature gas cooled reactor core optimization design method and system, comprising: constructing training data set;Training a data-driven proxy model is used to replace the execution high-cost simulation calculation;With a multi-objective evolutionary algorithm, and with the proxy model as its fitness function evaluator, to explore design space and generate a set of Pareto optimal candidate solution.The key features of the application are an iterative validation and model updating closed loop: the candidate solution is verified by high-fidelity simulation, and if the prediction accuracy does not meet the convergence criteria, the new data points after verification are expanded to the training data set, and the proxy model is retrained.The application can significantly reduce the calculation cost, efficiently explore the complex design space, and obtain a physically reliable and more optimal reactor design scheme.

The application discloses a supply chain network resilience prediction optimization method fusing machine learning and an evolutionary algorithm, and comprises the following steps: constructing a multi-layer dependent supply chain network model and a hybrid cascading failure model; secondly, multi-scene simulation is carried out on the model, the resilience evaluation index and the resilience value sequence based on a full-role subgraph are obtained, and a training sample is formed; then, a graph neural network model fusing a GraphSAGE framework and a GAT multi-head attention mechanism is constructed, and the model is trained by using the training sample; finally, the supply chain network structure data to be evaluated and an attack scene are input into the trained graph neural network model, and a node-level failure probability is quickly predicted. The application greatly improves the efficiency of supply chain network resilience optimization under the premise of ensuring accuracy by fusing the rapid prediction capability of machine learning and the optimization capability of an evolutionary algorithm, and provides effective support for risk management and resilience improvement of a complex supply chain network.

The application discloses a kind of water safety evaluation methods for cascade hydropower development basin, comprising the following steps: S1, first data is defined as time-varying ecological stress vector field;S2, second data is defined as initial ecological resilience potential distribution;S3, time-varying ecological stress vector field and initial ecological resilience potential distribution are coupled dynamics solved using resilience dissipation gradient field, generate the indication signal of ecological resilience dissipation gradient field sequence;S4, the target node set in ecological resilience dissipation gradient field sequence that exceeds safety threshold is extracted, and ecological discharge flow optimization instruction is generated;S5, ecological discharge flow optimization instruction is used as constraint condition to carry out iteration simulation, and the final water environment and water ecological control strategy are output;The application calculates ecological resilience dissipation gradient field to realize fine-grained dynamic early warning;Optimal release instruction is deduced in reverse relying on evolutionary algorithm, and ecology and economy are considered.

The disclosure relates to methods and systems of automatically generating rules based on an evolutionary algorithm. A system may generate an initial population of rule candidates bound by a respective constraint. The system may perform a crossover of the initial population of rule candidates with one another to generate a plurality of crossover child rules and may apply a probabilistic mutation to at least some of the plurality of crossover child rules. The system may evaluate performance of the plurality of crossover child rules based on one or more fitness evaluation metrics that measure performance of a given crossover child rule and select a subset of the plurality of crossover child rules based on the evaluated performance. The system may repeat the process until a specified number of evolutionary generations is reached.

The application discloses a sum and difference beam optimization method of a cover array antenna, and comprises the following steps: creating an array antenna model and a cover array antenna model; performing quadrilateral mesh division on the models respectively; based on the corresponding quadrilateral mesh division of the models, performing simulation calculation by using a high-order moment method, and extracting corresponding active unit directional diagrams; applying excitation to the antenna units in the models, combining the active unit directional diagrams, calculating the sum and difference beams of the models under different scanning angles, and comparing corresponding performance indexes; and using an improved hybrid differential evolution algorithm to optimize the sum and difference beams of the cover array antenna model which does not meet the index requirements; wherein the improved hybrid differential evolution algorithm performs global and local iterative searches, and adaptively calculates a mutation factor and a crossover factor in the global and local iterative search processes. The application realizes the optimization design of the sum and difference beams of the cover array antenna, and is more in line with engineering application requirements.

This invention discloses an underwater electric field source localization method based on a determinant orthogonal propagation operator and adaptive hybrid-driven differential evolution, belonging to the field of electromagnetic positioning and detection. This invention acquires and receives array signals and calculates the covariance matrix. Based on the propagation operator, it constructs a projection operator matrix orthogonal to the array manifold to replace the noise subspace, thereby constructing a determinant spatial spectrum function and defining a cost function. Then, based on this cost function, it constructs an adaptive hybrid-driven differential evolution framework, fusing the alpha evolution algorithm and an adaptive differential evolution algorithm based on successful historical records to solve for the target position coordinates. This invention eliminates the need for eigenvalue decomposition, reducing computational complexity, and accelerates convergence through an adaptive hybrid-driven search strategy, enabling fast and high-precision electric field source localization in complex underwater environments.

Fig 4 A configuration module 70 uses an evolutionary algorithm module 71 to determine an optimum hierarchy of perceptual control units (or negative feedback control units) 40 for multi-variable contro

The application relates to the technical field of artificial intelligence and control systems, and discloses a heavy gate dynamic torque regulation system based on reinforcement learning, which comprises the following: a neural evolution and meta-learning double-layer optimization framework for simultaneously optimizing a neural network structure and parameters; a neural gene coding and decoding system for coding a controller network structure into an evolvable neural gene; a layered neural architecture search space for providing a structured search space for an evolution algorithm; a neural modular evolution algorithm for realizing network structure evolution through function similarity evaluation module value; a super network meta-learning framework for generating specialized controller parameters for specific tasks; an evolutionary meta-learning collaborative optimization mechanism for promoting and complementing each other; and a deployment and online adaptation system for realizing continuous optimization of the system in actual operation; the application solves the problems of insufficient generality, low design efficiency and poor adaptability in the prior art.

The application relates to the field of intelligent meal matching, and discloses a modular precise nutrition intelligent meal box and a self-adaptive meal matching device. The self-adaptive meal matching device comprises a clustering module, a calculation module and a construction module. The clustering module is used for clustering a user feature vector based on a fuzzy C-means clustering algorithm. The calculation module is used for calculating a meal energy target and a macro-nutrient target according to a membership vector. The construction module is used for constructing a multi-objective optimization model comprising an energy matching degree, a macro-nutrient balance degree, a taste preference satisfaction degree and a dish diversity gain target function. The decision module is used for visualizing and displaying a Pareto optimal solution set for user selection. By constructing a multi-dimensional user feature vector, generating a personalized meal energy target and a macro-nutrient target according to a membership weight, establishing a multi-objective optimization model and outputting a Pareto optimal solution set by using a multi-objective evolutionary algorithm, the Pareto optimal trade-off and hierarchical constraint management of multiple conflict targets are realized.

A method for generating artificial intelligence (Al) models comprises receiving a plurality of Al models, encoding the Al models to generate a population of respective encoded hierarchical numerical representations that each encode computational units of the respective Al model as linear input-varying systems (LIVs) across a backbone, operator structure, and featurization hierarchical level, scoring each of the plurality of Al models to generate one or more respective score metrics, selecting a plurality of pairs of Al models from the received Al models based at least in part on the score metrics, for each pair, generating a respective offspring encoded hierarchical numerical representation, by combining numerical information from the encoded hierarchical numerical representations for each Al model in the respective pair, and for one or more of the offspring encoded hierarchical numerical representations, generating a respective offspring Al model based on the offspring encoded hierarchical numerical representation.