14 results about "Gray wolf" patented technology

The invention discloses a rescue path planning method for a search and rescue robot, and the method comprises the steps: firstly, carrying out the nonlinear self-adaptive adjustment of a path optimization strategy through introducing a dual-perception system based on population information entropy and optimization progress; secondly, carrying out orthogonal cooperation on social level guidance of a grey wolf algorithm and a speed disturbance mechanism of a simplified meteorological optimization model, and realizing adaptive fusion of the two strategies through evolvable individual weights; and then, performing iterative optimization based on combination of a grey wolf optimization mechanism and a simplified meteorological optimization model on the candidate path scheme to generate an optimal rescue path. According to the method, the path search strategy can be dynamically adjusted through the optimization state sensed in real time, intelligent balance is achieved between global exploration and local development, the convergence speed, the path quality and the robustness of path planning of the search and rescue robot in a complex environment are remarkably improved, and the reliability and the safety of path planning are improved.

The invention relates to the technical field of unmanned aerial vehicle logistics and path planning, in particular to an improved grey wolf optimization method for unmanned aerial vehicle logistics distribution planning. The method comprises the following steps: firstly, constructing a three-dimensional simulation environment containing an obstacle and setting a path constraint condition; a comprehensive objective function is established, and four sub-objectives of path length, collision penalty, smooth penalty and height violation penalty are integrated through weighted summation; in the algorithm level, Kent chaotic mapping is adopted to generate an initial population to improve diversity, control parameters are changed into a nonlinear adaptive updating strategy to balance global exploration and local development, and a particle swarm optimization algorithm idea is fused to introduce an adaptive inertia coefficient and a dynamic weight coefficient to improve a grey wolf position updating rule; and finally, performing iterative optimization through an improved hybrid algorithm, and outputting an optimal path. The method can effectively improve the quality and efficiency of path planning, and enables the generated path to be economical, safe and smooth.

The application discloses a kind of fusion multi-objective grey wolf optimization aquatic product freshness multi-task detection method, including data foundation construction, enhancement, CNN-LSTM-SE feature extraction, multi-task model formation, super parameter optimization, model training and evaluation deployment etc. steps.This method expands sample scale by data enhancement, extracts features using CNN and LSTM in parallel combined with SE attention mechanism, constructs multi-task learning framework to realize multi-index synchronous prediction, and optimizes super parameters in stages using MOGWO algorithm.The beneficial effects are to improve small sample robustness, feature extraction accuracy, multi-task detection efficiency and model generalization ability, and modular design facilitates popularization and application.

An attitude maneuvering path planning method based on an improved gray wolf algorithm is proposed. The improved algorithm calculates the fitness value of each gray wolf. As the wolf's position is updated, individuals within the wolf pack continuously mutate, leading to constant population renewal and effectively solving the problem of entering local optima. Simultaneously, the high efficiency of the gray wolf algorithm allows for controlled implementation time. By comparing the fitness values ​​of individual gray wolves and applying a penalty function to wolves with poor fitness values, the global search capability is improved. The optimal parameters output by the improved gray wolf algorithm are used to simulate satellite paths and determine the changes in satellite control attitude data.

The invention provides a black box confrontation attack method based on a grey wolf optimization algorithm, belongs to the technical field of artificial intelligence security, and aims at solving the problems that an existing black box confrontation attack method is insufficient in mobility and low in search efficiency. The black box confrontation attack method comprises the steps that wolf pack initialization is carried out according to an input image, and an initial wolf pack is generated; a multi-level grey wolf optimization architecture is adopted, a wolf pack responsible for global guidance, local optimization and auxiliary exploration is constructed in combination with an initial wolf pack, all the wolf packs cooperatively run a grey wolf optimization algorithm, the positions of the wolf packs are updated through a dynamic hunting strategy, information exchange between the wolf packs is periodically executed, and group updating of the wolf packs is completed; after group updating of the wolf pack is completed, self-adaptive disturbance optimization of the wolf pack is carried out; and selecting a global optimal solution from the wolf pack responsible for global guidance after adaptive disturbance optimization, generating a final adversarial sample, and evaluating the attack success rate and average pixel disturbance of the adversarial sample on the target model according to the final adversarial sample.

The present application relates to the technical field of data processing, and particularly relates to a natural reserve unmanned aerial vehicle inspection optimization method, which solves the technical problem of unreasonable inspection route planning of an unmanned aerial vehicle in inspecting a natural reserve in the prior art. The method comprises the following steps: dividing the natural reserve into a plurality of grids, calculating the static obstacle density of the non-obstacle grids in the plurality of grids based on the point cloud data of the static obstacles of the natural reserve, and calculating the dynamic obstacle density of the non-obstacle grids based on the image data of the dynamic obstacles of the natural reserve; determining the comprehensive obstacle factor of each non-obstacle grid based on the static obstacle density and the dynamic obstacle density; selecting the initial point of the grey wolf algorithm based on the static obstacle density, and performing iterative calculation of the grey wolf algorithm with the minimum weighted value of the inspection path length and the comprehensive obstacle factor of the inspection path as the optimization target to determine the inspection path.

The invention belongs to the technical field of mechanical part design, and discloses a bridge crane main beam lightweight design method, which comprises the following steps of: acquiring cross section data of a crane main beam as a design variable, and taking weight reduction of the crane main beam as an optimization target; generating constraint conditions based on the vertical static stiffness, the normal stress, the shear stress, the fatigue strength and the overall stability to construct a crane girder section model; carrying out design variable coding on the crane main beam section model based on an improved grey wolf optimization algorithm, carrying out initialization setting, selecting a first wolf by constructing a fitness function, introducing adaptive update control parameters, updating the position of a grey wolf in a population, and when a termination condition is met, carrying out self-adaptive optimization on the grey wolf; and outputting the potential solution corresponding to the gray wolf position as an optimal solution. Based on the crane main beam section model and the improved grey wolf optimization algorithm, the intelligent lightweight design of the crane main beam is realized, the scheme design efficiency is improved, and the crane manufacturing cost is effectively reduced.

The invention provides a hydrological model optimization method suitable for an arid inland river basin, and the method comprises the steps: collecting hydrological meteorological data, determining the noise number K in an algorithm through employing a K-Means clustering algorithm, enabling the processed hydrological meteorological data to exchange K clusters, enabling each cluster to represent a specific hydrological feature, calculating the center point of each cluster, and carrying out the calculation of the center point of each cluster, randomly initializing a group of grey wolf optimization algorithms in each cluster; instantiating an LSTM model network structure in real time according to the position vector of the current population, setting a fitness function, and evaluating the fitness of grey wolf individuals; and training the hydrological model by using an optimal parameter combination obtained through optimization of the grey wolf algorithm, and evaluating the prediction capability of the model by using a cross validation method. According to the method, the historical hydrological data are clustered to identify different hydrological modes, then the grey wolf algorithm is used to carry out specific parameter optimization on each mode, and through adaptive parameter adjustment and data grouping analysis, accurate prediction of future hydrological conditions can be realized.

The application provides a structure adjustment method, medium and system of a monolithic integrated inverse-conducting GaN-HEMT power device, and belongs to the technical field of GaN-HEMT power devices, and comprises the following steps: a heat flow model and a heat dissipation path topology of the internal structure of a GaN-HEMT chip are established, and a key path analysis algorithm is used to calculate and identify the main bottleneck area of heat dissipation; an upper model with the optimal heat dissipation as the target and a lower model with the lowest drain current of the GaN-HEMT as the target are established, and the constraint conditions of the upper model and the lower model are determined; the micron-level metal heat sink parameters, the micro-channel cooling structure parameters and the vertical groove parameters formed by the source and drain regions of the GaN-HEMT are used as Alpha wolves, Beta wolves and Delta wolves of a grey wolf hunting algorithm, the grey wolf hunting algorithm is used to solve the game model, and the optimal solution obtained is used as an adjustment parameter.

The invention provides an intelligent identification method for the current situation of a deep and far sea based on a multi-temporal remote sensing image, and belongs to the technical field of identification of the current situation of the deep and far sea. According to the method, sub-pixel-level image registration is realized by adopting a layered registration strategy based on coastline stable ground features, geometric, texture and spectral features of deep and far sea structures are extracted to establish a feature vector library, and the cosine similarity of adjacent inter-phase feature vectors is calculated to carry out change detection; a coevolution network structure is combined with a sea area utilization state identification model of a multi-head grey wolf hunting model to carry out sea use type classification, illegal behaviors are judged, and an intelligent identification report including spatial distribution, change trend and compliance evaluation is generated. The technical problem that the sea area structure change detection precision is insufficient due to the fact that the multi-temporal remote sensing image is affected by cloud shielding in the deep sea environment is solved.

This application discloses a method for predicting harmful algal blooms based on Tent-GWO-GRU, comprising: acquiring and preprocessing historical water quality data of the target watershed; constructing a gated recurrent unit neural network (GRU) model based on the historical water quality data; optimizing the Grey Wolf Algorithm (GWO) based on the Tent chaotic mapping algorithm; and optimizing the GRU model parameters and making predictions. This application applies a time delay to the index data, fully considering the lag in algal bloom growth and evolution, thus improving the feasibility of the method. The introduction of the Tent chaotic mapping algorithm improves the method for generating the initial wolf pack in the GWO algorithm, resulting in a more uniform distribution of the initial wolf pack, significantly improving the algorithm's fitness value, and making it easier to find the global optimum. This application uses the Tent-GWO optimization algorithm to optimize the hyperparameters and network structure of the GRU model, which not only improves the model's stability but also achieves higher prediction accuracy.

This invention relates to the field of power plant planning technology and discloses a method for optimizing the distributed power generation capacity of a virtual power plant based on an improved gray wolf algorithm. This method, based on the operating strategy of the virtual power plant, simulates the configuration scheme represented by each individual gray wolf in the initialized gray wolf population to obtain operating data. Then, based on the operating data and the distributed power generation capacity optimization configuration model, the initial fitness value of each gray wolf is obtained. Based on the distance between gray wolf individuals and the radius of their niche, a corresponding niche is constructed with each gray wolf as the center. Through a sharing mechanism, based on the number and distribution of gray wolves within each niche, the initial fitness value of each gray wolf is adjusted to obtain a target fitness value. Based on the target fitness value, the gray wolf individuals in the population are iteratively updated until the convergence condition is met, resulting in the optimal configuration strategy.

This invention belongs to the field of data recognition technology and provides a method and device for fish recognition based on an optimized convolutional neural network. It utilizes acquired historical fish image data to train an improved YOLOvx target recognition model. The model construction process includes first performing a global optimization using the Tornado algorithm to obtain the initial tornado position, then employing a strategy combining the Tornado algorithm and the Grey Wolf algorithm for local optimization. The global optimization process in the Tornado algorithm is improved by incorporating the fourth defense mechanism of the crested porcupine, resulting in a Tornado combined intelligent algorithm. This algorithm is then used to improve the upsampling process of the YOLOvx target recognition model, resulting in an improved YOLOvx target recognition model. The trained model is then used to recognize the target image to obtain the recognition result of mixed fish. This invention solves the problem of low recognition accuracy in existing technologies and can improve the accuracy of fish recognition.

The invention discloses a natural gas consumption prediction method based on an adaptive gray seasonal model, and the method comprises the steps: collecting natural gas consumption data to form an original seasonal sequence, and calculating a cumulative sequence; building a natural gas consumption prediction model according to the cumulative sequence; then, a grey wolf optimizer is used for searching adaptive parameters of the natural gas consumption prediction model; then, linear parameters of the natural gas consumption prediction model are obtained through least square estimation; generating a time response equation of the natural gas consumption prediction model according to the linear parameters; then, calculating a predicted value of the cumulative sequence; subsequently, a predicted value is obtained by a restoration operation. According to the invention, a natural gas consumption prediction task is executed based on a novel adaptive gray season model; according to the model, the nonlinear trend of data can be effectively captured through adjustment of structural parameters; by introducing seasonal correction, the seasonal change of the data can be accurately identified.