57 results about "Artificial bee colony optimization" patented technology

The present invention discloses an improved extreme learning machine method based on artificial bee colony optimization, which includes the following steps: Step 1, generating an initial solution for SN individuals: Step 2, globally optimizing a connection weight ω and a threshold b for the extreme learning machine; Step 3, locally optimizing the connection weight ω and threshold b of the extreme learning machine; Step 4, if food source information is not updated within a certain time, transforming employed bees into scout bees, and reinitializing the individuals after returning to Step 1; and Step 5, extracting the connection weight ω and threshold b of the extreme learning machine from the best individuals, and verifying by using a test set. With the method provided by the present invention, the defect of worse results of the traditional extreme learning machine in classification and regression is overcomed, and effectively improves the results of classification and regression.

The invention discloses an aquaculture water quality short-time combination forecast method on the basis of multi-scale analysis. The method includes the steps that water quality time sequence data are acquired online and repaired; through empirical mode decomposition, the selected water quality time sequence sample set data are decomposed into IMF components and residual rn components, wherein the IMF components and the residual rn components are different in frequency scale; the IMF components and the rn components are classified, a manual bee colony optimization least square support vector regression machine, a BP neural network and an autoregressive sliding average model are respectively selected for forecast according to classifying features, and finally, all results are weighed and summed to obtain a water quality time sequence forecast result. According to the method, the original water quality time sequence data are decomposed into the components different in time frequency through the empirical mode decomposition, and change conditions in original water quality sequences can be mastered more accurately; advantages of the manual bee colony optimization least square support vector regression machine, advantages of the BP neural network and advantages of the autoregressive sliding average model are complemented and combined, and thus performance of a combined forecast model is effectively improved.

The invention discloses a diagnostic method for gas path faults of an aero-engine based on the sliding mode theory. The method comprises following steps: intelligently correcting a non-linear part-stage model of the aero-engine based on an artificial bee colony algorithm; obtaining a self-adaptive liner model of the aero-engine based on error feedback sliding mode control; designing an expansion interface estimation sliding mode observer, and achieving fault diagnosis, separation and reconstruction of an engine sensor; achieving fault diagnoses of gas path faults of the engine. The diagnostic method for gas path faults of the aero-engine based on the sliding mode theory has following beneficial effects: as for the gas path faults of aero-engine, a correction method for a high-precision part-stage mode for the engine is put forward with the purpose of improving reliability and safety of a system; a new way of thinking for a self-adaptive linear model for the engine is put forward for providing a new strategy of gas path faults of the aero-engine; difficulties of gas path faults of the aero-engine and sensor faults are solved so that an effective and highly-reliable fault diagnoses method is put forward.

The invention discloses an artificial bee colony refine edge potential field function-based unmanned plane target identification method. The method comprises the following steps of: 1, pre-processing an image; 2, calculating edge potential field distribution of an image to be matched; 3, initializing algorithm parameters of an artificial bee colony; 4, calculating a function value of the adaptability degree of each employed bee individual according to the edge potential field distribution of the image to be matched; 5, selecting the employed bee with more preferential value of the adaptability degree as a leading bee according to probability and searching a nectar source in surrounding space by a worker bee, and calculating the corresponding adaptability degree value; 6, updating the position of the employed bee according to the greed selection method; 7, searching a new nectar source around again by the employed bee; and 8, adding 1 to iterations, continuing the steps from the step 4 until the set maximum iterations is reached to finish the calculation, and outputting the optimal result. The method effectively improves the reliability and precision of the unmanned plane target identification.

The invention brings forward a constraint multi-target optimization method based on an improved artificial bee colony algorithm, for solving the defect of solving a multi-target optimization problem by use of a basic artificial bee colony optimization method. The constraint multi-target optimization method based on the improved artificial bee colony algorithm takes the material scheduling problem at a first emergency rescue phase as an application background and brings forward a food source initialization process integrated with reverse learning on the basis of the definition of a reverse solution for improving the quality of 50% of an initial solution. At the same time, for the purse of balancing the development capability and the exploration capability of an optimization process, the method integrates a reverse learning strategy and an extensive learning strategy into a honeybee search process so as to improve the search efficiency. The method constructs a non-linear deletion loss based many-to-many disposable consumption emergency material scheduling constraint multi-target optimization model, and embodiments are formed. Numerous test results indicate that compared to a conventional artificial bee colony optimization method, the method provided by the invention has the following advantages: more non-dominated leading-edge solutions are solved, the distribution on a solution space is wider and more uniform, and the solution are closer to a Pareto optimal solution.

The invention relates to an under-actuated UUV (unmanned underwater vehicle) depth control method and provides an under-actuated UUV depth control method based on artificial bee colony optimized model predication to solve the problem of UUV depth control under complex horizontal angle constraint conditions. The method includes: firstly, acquiring an incremental vertical plane predication model of an under-actuated UUV; secondly, unifying horizontal angle control input constraint conditions as H delta U (k) < / =gamma; thirdly, by model predication control, turning the UUV depth control problem into a secondary planning problem under the constraint conditions; fourthly, obtaining a global optimum position of a nectar source; fifthly, obtaining control input at a moment k; sixthly, guaranteeing that the UUV reaches an appointed UUV diving depth R (k+1) to complete diving. The under-actuated UUV depth control method is applicable to the field of UUV depth control.

The invention discloses an extreme learning machine method for improving artificial bee colony optimization. The method is characterized by comprising the following steps: 1, generating initial solutions for SN individuals; 2, carrying out global optimization on a connection weight Omega and a threshold b of an extreme learning machine; 3, carrying out local optimization on the connection weight Omega and the threshold b of the extreme learning machine; 4, if food source information is not updated in a certain time, converting employee bees into investigation bees and returning to the step 1 to re-initiate the individuals; and 5, extracting the connection weight Omega and the threshold b of the extreme learning machine from the optimum individual, and carrying out verification by using a test set. The method disclosed in the invention can be used for better overcoming the defect that the traditional extreme learning machines are relatively bad in results when being applied to classification and regression; and compared with the traditional extreme learning machines and SaE-ELM algorithms, the method has relatively strong robustness and can be used for effectively improving the results of classification and regression.

The invention discloses a method for predicting the pH value of stream in a rare earth mining area based on a cloud model and artificial bee colony optimization. A support vector machine (SVM) is utilized to predict the pH value of the stream in the rare earth mining area; a penalty factor C of the SVM, a radial basis kernel parameter g and a parameter epsilon in an insensitive loss function are encoded into individuals of an artificial bee colony algorithm; a cloud model sampling based observation bee search operator is designed; and a chaotic elite backward learning based scout bee search operator is designed for improving the search capability of the algorithm; and compared with a similar method, the method provided by the invention can improve the precision of predict the pH value of the stream in the rare earth mining area.

The invention relates to a planetary gear box fault diagnosis method. The method comprises the following steps: firstly, decomposing and reconstructing a signal by utilizing salp swarm group optimization-variational mode decomposition (SSO-VMD); then, extracting fault features from multiple domains, and carrying out dimension reduction processing by adopting improved supervised self-organizing incremental learning neural network landmark point isometric mapping (ISSL-Isomap); and finally, using an artificial bee colony optimization support vector machine (ABC-SVM) classifier to carry out diagnosis and identification. According to the method, the problem of parameter selection in the VMD algorithm is solved, and the problem of information redundancy of multi-domain features is solved. A planetary gear box fault diagnosis experiment result shows that the method can effectively identify each fault type and has a great practical value.

The invention discloses an intrusion detection method and system of a BP algorithm based on artificial swarm optimization. The method comprises the following steps: forming a packet through a collected host log file and network data, preprocessing the packet to obtain feature vectors of the host log file and the network data, and converting the feature vectors into input values that can be identified by a BP neural network algorithm; initializing the BP neural network algorithm, using a weight Wij connecting an input layer and a hidden layer and a weight Wjk connecting the hidden layer and anoutput layer as optimization targets of the artificial swarm algorithm to initialize the parameters of the artificial swarm algorithm, and transmitting an optimal honey source to the BP neural networkalgorithm to replace the weight Wij connecting the input layer and the hidden layer and the weight Wjk connecting the hidden layer and the output layer; and making a corresponding response operationon the behavior of a user according to the output value of the BP neural network algorithm. By adoption of the intrusion detection method and system, the problems of low convergence rate, vulnerability to local minimum point and large calculation amount of the existing BP neural network.

A ducted unmanned aerial vehicle anti-sway method based on optimized quadratic form control of artificial bee colony includes eight steps: 1, the mathematical model for pendulum oscillation is built; 2, control structure and control law are designed; 3, the parameters of the artificial bee colony algorithm and the employed bee colony are initialized; 4, the performance indicator function of the linear quadratic form is calculated according to individual parameters; 5, worker bees select bee individuals with better fitness as leading bees according to the fitness value of each employed bee, and each worker bee continues to seek for honey sources near the leading bee solution space and the fitness value is calculated; 6, if the number of searches Bas is greater than the set threshold, the employed bees seek for new honey sources again, namely parameter values are initialized again; 7, by the greedy selection method, the positions of the employed bees are updated with larger fitness values, and searches proceed near solution spaces; and 8, the Step 4 is carried out repeatedly until T > Tmax, and the optimum component weighted value parameter, the optimum feedback gain matrix and the optimum fitness value are output.

The invention relates to a fault diagnosis method for a photovoltaic array based on a semi-supervised extreme learning machine, which comprises the steps of firstly, acquiring an output voltage-current curve of the photovoltaic array through an acquisition device; then performing feature extraction on the current-voltage curve, and constructing a fitting feature output equation with an adjustmentcoefficient; secondly, solving the adjustment coefficient by adopting a nonlinear least square method based on particle swarm-trust region reflection optimization; obtaining a feature standardizationequation by performing item shifting and standardization on the feature output equation; furthermore, carrying out fault identification on the photovoltaic array in combination of a small number of label samples and a large number of label-free samples by adopting a semi-supervised extreme learning machine based on artificial bee colony optimization as a classifier; and finally, regularly measuring a current-voltage curve of normal operation of the photovoltaic array to update the standardized equation so as to adapt to natural aging of the photovoltaic array.

The invention discloses a crowd evacuation simulation method and a device based on optimizing positive emotion infection. For the trust relationship between individuals, an emotional contagion networkis constructed; For the emotional contagion network, a positive emotional contagion model is constructed. The process of emotional infection was analyzed in a parameterized manner, and the infectionprobability of each node in the emotional infection network was calculated by using continuous-time Markov chain. Construct the optimization model of maximizing positive emotional infection; Using artificial bee colony optimization emotional infection algorithm and the infection probability of each node, the optimal objective function value of a given groomer in the model of maximizing positive emotional infection optimization problem is calculated. Using the optimal objective function value of the given dredger, the optimal position of the safety dredger can be calculated. The present disclosure provides guidance for crowd evacuation by leveraging optimized positive emotional contagion, provides new solutions to public safety issues, and provides guidance for emergency management.

The invention discloses a multi-strategy artificial bee colony optimized concentration predication method for dissolved oxygen in water. A support vector machine serves as a concentration predication model for dissolved oxygen in water, and a penalty factor C, a radical basis kernel parameter g and a parameter epsilon in an insensitive loss function of the support vector machine are subjected to optimal design according to a multi-strategy artificial bee colony algorithm. In the multi-strategy artificial bee colony algorithm, various search strategies are fused, and search strategies are adaptively selected for optimization according to a current evolution state in an optimization process, so that algorithm optimization capacity is improved. By the multi-strategy artificial bee colony optimized concentration predication method for dissolved oxygen in water, precision in concentration predication of dissolved oxygen in water can be improved.

The invention discloses a method for identifying key protein using an artificial bee colony optimization algorithm of a foraging mechanism. The method comprises: converting a protein-protein interaction network to an undirected graph, obtaining ribonucleic acid genetic expression values corresponding to protein, preprocessing edges and nodes of the protein-protein interaction network, establishing a dynamic protein-protein interaction network, selecting known key protein as a honey source, honey bees searching neighbourhood of the honey source, following bees searching neighbourhood of the honey bees, updating the honey sources, investigating bees searching new honey sources in a global manner, updating the honey sources, and generating the key protein. The method can accurately identify the key protein. Results of simulation experiments show that sensitiveness, specificity, positive predictive values, negative predictive values and other performance indexes are relatively excellent. Compared with other key protein identification method, the identification process of key protein realized by combining optimizing characteristics of artificial bee colony with characteristics of the protein-protein interaction network improves identification accuracy rate of the key protein.

The invention relates to the technical field of bioinformatics and provides a method for mining epistasis loci of an artificial bee colony optimized Bayesian network, including four steps S1 to S4. The method for mining epistasis loci of the artificial bee colony optimized Bayesian network, comprises firstly using three stages of expansion, contraction, and symmetry detection to calculate the Markov blanket of nodes through conditional mutual information so as to construct an initial nectar source network structure; then, based on the initial nectar source, randomly adding, subtracting and reversing edges to generate new nectar source until the maximum number of initial nectar sources is reached. The three operations (collecting bees, observing bees, and reconnoitering bees) of artificialbee colony algorithm and the BIC and MIT scoring method of the Bayesian network are configured to evolve the structure of the Bayesian network, find the optimal network structure, quickly and accurately obtain the epistasis gene loci that affect phenotypic traits, and assist the gene function mining.

The invention discloses an abnormal traffic detection device of an energy internet information support network, which realizes identification of abnormal traffic by performing traffic sampling, feature extraction and classification on the information support network. According to the method, frequency-domain features of uplink traffic and downlink traffic data between a server and a core switch are extracted based on a wavelet packet analysis theory, an energy Internet information support network traffic time-frequency domain mixed feature set is constructed, and an artificial bee colony optimized support vector machine is used as a classification model. According to the device, support is provided for accurate and rapid verification of traffic anomaly and security situation awareness of the communication network of the intelligent substation and guarantee of stable operation of an information physical tight coupling power system.

The invention discloses an ore rock strength soft sensing method based on adaptive artificial bee colony optimization. The method of the invention adopts a three-layer perception neural network as a soft sensing model for ore rock strength, and utilizes the adaptive artificial bee colony algorithm to optimize a connection weight and an offset value of the designed neural network. In the adaptive artificial bee colony algorithm, search scaling factors are generated adaptively according to the feedback information of adaptive values, and a Gaussian mutation strategy based on neighbor optimal individuals and global optimal individuals is designed to adaptively generate new individuals. The method of the invention can raise the precision of soft sensing of the strength of ore rocks, and raises the efficiency of measuring the strength of the ore rocks.

The invention discloses a rare earth ore district underground water ammonia nitrogen concentration prediction method with Gauss artificial swarm optimization. A support vector machine is used as a rare earth ore district underground water ammonia nitrogen concentration prediction model; a Gauss artificial swarm algorithm is used for optimizing and designing a penalty factor C of the support vector machine, a radial base core parameter g and a parameter Epsilon in a non-sensitive loss function. In the Gauss artificial swarm algorithm, an outstanding individual in each individual neighborhood and an individual average value in a neighborhood thereof are merged into a Gauss mutation strategy to generate new individuals; in addition, a reverse study strategy using the optimum individuals as reference points is executed in a bee searching process. The method has the advantage that the prediction precision of the rare earth ore district underground water ammonia nitrogen concentration can be improved.

The invention discloses a self-adaption artificial swarm optimization method based on historical information in the running process. The parallel research is conducted through various strategies, and the use ratio of the strategies is unceasingly regulated and guided according to the historical information in the running process. The Gaussian distribution which the important parameter Limit used for balancing the overall search cost in the method belongs to is unceasingly regulated according to the historical information in the running process. The use ratio of the search strategies is dynamically distributed through the self-adaption artificial swarm optimization method according to the historical information in the running process, and the important parameter for balancing the overall search cost is dynamically regulated. Under the condition that the numbers of the calculation times of fitness functions are the same, the method has a more superior and efficient function optimization effect.

The invention discloses a wine quality discriminating method based on reABC-SVM. The method comprises the steps of 1, extracting a sample from the wine, and measuring contents of certain materials in a wine sample for forming a training sample set; and 2, performing dynamic adjustment on a penalty factor and a nuclear function parameter of a support vector machine by means of an improved artificial bee colony optimization algorithm, outputting an optimal parameter, and establishing a wine discriminating model by means of an optimal parameter, thereby realizing quality discriminating on the wine. According to the wine quality discriminating method, an SVM parameter can be optimized by means of an improved ABC algorithm, thereby obtaining a most appropriate classification model for realizing wine quality classification, and furthermore settling a wine quality discriminating problem.

The invention discloses an artificial bee colony optimized rare-earth mine underground water total nitrogen concentration soft measurement method which adopts a support vector machine as a soft measurement model for total nitrogen concentration of underground water of a rare-earth mine, and a penalty factor C and a radial basis kernel parameter g of the support vector machine and a parameter epsilon are optimized and designed by using an adaptive artificial bee colony algorithm. In the adaptive artificial bee colony algorithm, search zoom factors are adaptively adjusted by using feedback information of an adaptive value, and information of an optimal individual of an adjacent domain and an optimal individual of a global domain is infused to a Gaussian variation strategy to generate new individuals adaptively. By adopting the method, the soft measurement precision of the total nitrogen concentration of the underground of the rare-earth mine can be improved.

The invention discloses a three-dimensional image registration method based on a reselection point strategy and an artificial bee colony optimization, which comprises the following steps: (1) samplinga dynamic point cloud to obtain a sampling point set; (2) generating a certain number of bee colonies and initializing the position of the individual bee colonies; (3) determining the Euclidean transformation moment of each honeybee, and transforming the position of the sampling point set according to the Euclidean transformation matrix; 3) transform that sample point set by the Euclidean transform moments and calculate the objective function value; (5) comparing the variation of the optimal objective function value, and if the variation is smaller than the threshold value for successive times, reselecting the point operation to obtain a new sample point set, otherwise, entering the step (6); 6) if that maximum evolutionary algebra is reach, entering the step 7), otherwise return to the step 3); (7) obtaining the optimal Euclidean transformation moment from the optimal solution of the population; And moving the dynamic point cloud to complete image registration. In this method, the re-selection strategy is introduced into the sampling process and the bee-colony algorithm is combined to effectively reduce the time of image registration and improve the performance.

The invention discloses a blind source separation main lobe interference resisting method and device based on an improved artificial bee colony, and relates to the technical field of signal processing, and the blind source separation main lobe interference resisting method based on the improved artificial bee colony comprises the steps: firstly receiving a to-be-processed mixed signal; carrying out decentration preprocessing on the to-be-processed mixed signal to obtain a preprocessed signal; further, determining a separation matrix corresponding to the preprocessed signal according to a preset artificial bee colony optimization algorithm, a preset blind source separation algorithm and a preset cost function; and finally, according to a blind source separation algorithm, the separation matrix and the preprocessed signal, calculating a main lobe interference resistant target signal, so that the main lobe interference resistant processing of the signal can be realized, the iteration speed is high, and the main lobe interference resistant effect is good.

The invention discloses a simple and efficient optimization method for improving an artificial bee colony. The method comprises the following steps of (1), initializing an algorithm; (2), calculating honey amount of each honey source according to an improved honey source initializing formula; (3), making bees search a new honey source according to a searching strategy based on normal distribution, and performing preferential selection on the new honey source and the old honey source; (4), calculating probability at each honey source position, and selecting a guiding bee for being followed by following bees according to the probability; (5), after number of searching times of one honey source reaches a cycling upper limit and updating does not occur, changing the guiding bee to a spy bee for searching; and (6), if a highest number of evolution generations is reached, outputting the position coordinate of an optimal honey source, and otherwise, returning to the step (3). According to the method of the invention, a normal distribution theory is introduced into an initializing process and a searching optimizing process of the artificial bee colony algorithm, thereby effectively improving global optimization solving capability and optimization precision.

The invention discloses a domestic intelligent microgrid optimization configuration method. The method comprises the steps of analyzing actual conditions of local wind and solar energy resources; performing investigation and survey on domestic load electricity consumption data; determining types and capacity limit of distributed power supplies, and types of inverters and controllers of the domestic intelligent microgrid; taking the lowest total operating cost of the domestic intelligent microgrid as an optimization target function; setting constraint conditions for the domestic intelligent microgrid to satisfy requirements of domestic load on power quality and technology; performing an artificial bee colony optimization algorithm; and outputting an optimization result with the lowest target function value and obtaining a configuration scheme. According to the domestic intelligent microgrid optimization configuration method, the artificial bee colony algorithm is applied to the optimization configuration of the domestic intelligent microgrid, so that the optimization speed is relatively high and the working efficiency is improved; a collector for sampling the domestic load electricity consumption is designed based on LABVIEM virtual instrument visual software, so that the domestic load electricity consumption can be monitored and sampled in real time, and the peak-valley periods of electricity consumption can be observed, so that evidence can be provided for realizing reasonable load access.

The invention aims to provide a ship electric propulsion system fault diagnosis method based on an ABC-SVM expert system, knowledge and experience of experts are stored in a knowledge base, a supportvector machine optimized by using an artificial bee colony is used in an inference machine, the knowledge in the knowledge base is preprocessed and then is used for performing classification trainingon the inference machine and storing a trained model. real-time data collected from the ship electric propulsion system is stored in a dynamic database, feature value extraction is conducted, the datais input into a trained inference engine, whether the ship electric propulsion system breaks down or not is judged, which device breaks down is judged, and the reason why the fault occurs is explained. Practicability and real-time performance of an expert system are greatly improved, the problem of a local optimal solution of the support vector machine is solved to a great extent, and the accuracy of fault diagnosis is improved.

The invention relates to the technical field of image processing, and provides a pantograph-catenary image registration method and device optimized by chaotic heuristic search, aiming at solving the technical problems of slow convergence speed and insufficient registration precision when artificial bee colony optimization algorithm is applied to pantograph-catenary image registration. For this purpose, the chaotic heuristic search optimized pantograph-catenary image registration method of the invention comprises the following steps of generating an initial population according to a preset reference pantograph-catenary image, wherein each individual in the initial population stores a predetermined number of registration parameters; optimizing the registration parameters of pantograph-catenary images by a chaotic heuristic search algorithm; obtaining the pantograph and catenary images to be registered by spatial transformation of the registration parameters. The invention can enhance thelocal search ability of the algorithm, accelerate the convergence speed of the algorithm, and improve the accuracy of pantograph-catenary image registration.

The invention relates to the technical field of alternating-current asynchronous motor control, and particularly relates to an artificial bee colony parameter optimization-based direct torque controlmethod for an alternating-current asynchronous motor slip film variable structure. The artificial bee colony parameter optimization-based direct torque control method for the alternating-current asynchronous motor slip film variable structure specifically comprises the following steps of establishing a stator flux linkage equation, an electromagnetic torque equation and a magnetic chain amplitudesquare expression of an alternating-current asynchronous motor; establishing a direct torque controller equation of the slip film variable structure; designing a calculation method for calculating thetarget function and the fitness value of an artificial bee colony optimization algorithm; initializing a nectar source position of a standard artificial bee colony algorithm; and finally, determiningthe parameters of a slide film variable structure controller equation through the search of the standard artificial bee colony. According to the method, a torque and magnetic chain slip film variablestructure-based controller is designed to replace a traditional direct torque controller. Meanwhile, the optimal parameters of the slip film variable structure-based controller are determined throughthe search of the artificial bee colony. Therefore, the variable-frequency speed regulation performance of the alternating-current asynchronous motor is effectively improved.

The invention discloses a gesture recognition method based on variational mode decomposition and a support vector machine, which comprises the following steps: firstly, performing decomposition and noise reduction on surface electromyogram signals by using a variational mode decomposition algorithm, and meanwhile, realizing optimal selection of parameters of the variational mode decomposition algorithm by using an improved artificial bee colony optimization algorithm, thereby avoiding blindness of manual selection; then, on the basis of the decomposed variational mode components, 4-order autoregression model parameters and fuzzy entropy are extracted, a multi-scale feature set is constructed, and surface electromyogram signal features can be effectively extracted; and finally, performing classification and gesture recognition by using a multi-classifier constructed by a support vector machine optimized by an improved cuckoo algorithm, and realizing a relatively good classification effect under the condition that the number of samples is relatively small. According to the method, the parameter selection problem of the variational mode decomposition algorithm is solved, the characteristics of the electromyographic signals can be more effectively obtained, and the accuracy and speed of gesture recognition based on the surface electromyographic signals are improved.