36 results about "Memetic algorithm" patented technology

The invention discloses a multi-target reactive power optimization method for an electric system, which belongs to the field of reactive power optimization for electric systems. The method includes: modifying the Memetic algorithm to adapt to multi-target optimization, applying the modified Memetic algorithm to the problem of multi-target reactive power optimization for the electric system, and working out a Pareto optimal solution of the multi-target problem; and judging whether algorithm convergence conditions are met or not, and if yes, completing optimization and outputting optimization results. The multi-target reactive power optimization method has the advantages that the algorithm for solving the problem of multi-target reactive power optimization is provided, the method is more suitable for solving the multi-target problem while giving play to existing advantages of the Memetic algorithm which integrates local searching and evolutionary computation and has high global search capacity and the like, and searching efficiency is improved while algorithm robustness is improved.

The invention provides a reconstructible assembly line balancing optimization method adopting an improved memetic algorithm. The reconstructible assembly line balancing optimization method comprises steps of coding and decoding, initial population generating, sequence adjustment, adaptation degree calculation, parent selection, population update, and local searching. When an ideal state or set iteration frequency is acquired, a chromosome having a minimal adaption degree value is output, and then a task allocation way corresponding to the chromosome having the minimal adaption degree value is acquired. A productivity and assembly line smoothness degree weight function is used as an adaption degree value function. Compared with a common memetic algorithm and a common genetic algorithm, the reconstructible assembly line balancing optimization method is advantageous in that the acquired adaption degree value is low, and the optimizing capability of the algorithm is effectively improved, and more reasonable solution is acquired.

The invention provides an SAR image search method based on sparse coding classification. The SAR image search method aims at the defects of an existing image search system and method. Through extraction of characteristics and construction of an over-complete dictionary, solution is carried out through sparse representation based on a dual Memetic algorithm, a sparse representation classifier is trained, the classification process with supervision can be achieved in the classification process, the sparse solution with global optimum can be solved fast, and then search results are output from high to low according to similarity. When the problem of image classification is solved, the method achieves the good improvement effect on classification accuracy, search content similarity, calculating complexity and result robustness.

The present invention discloses a data compression system for DNA sequence, which is a lossless compression system for DNA sequence data, based on the MA-ARV codebook, which is able to search the approximate repeat fragment of the MA-ARV code vector in the whole sequence, and use a heuristic optimization algorithm of memetic algorithm to optimize the construction process of the compressed codebook, so as to fully use the repeat nature of DNA sequence data, and eliminate the redundancy effectively.

The invention discloses a three-dimensional box loading method based on a three-dimensional moving mode sequence and a memetic algorithm. The method mainly solves the problem of low utilization rate on the volume of a three-dimensional box loading container in the prior art. The three-dimensional box loading method comprises the following realization steps that 1, each parameter is set; 2, an initial population is randomly generated, and the adaptive fitness of individuals in the population is calculated; 3, whether the termination conditions are met or not is judged, if so, the step 4 is executed, and otherwise, the step 9 is executed; 4, a binary tournament method is used for selecting the individuals; 5, the individuals are crossed, and the individual adaptive fitness value is calculated again; 6, the individuals are subjected to variation, and the individual adaptive fitness value is calculated again; 7, the individual with the greatest adaptive fitness value in the current generation is stored; 8, the number of the iteration times is added to 1, and the operation returns to the step 3; 9, a hill climbing method is used for optimizing the individuals with the greatest adaptive fitness value, and the optimized box loading result is output. The method has the advantages that the volume utilization rate of the container can be improved, and the method can be used for solving the box loading problem, and can also be used for soling other combination optimization problems.

The invention discloses a wave beam forming method of ascending MIMO-SDMA, which relates to the field of wireless communication. The invention provides a wave beam forming method based on a New-Memetic algorithm. Aiming at the defect of the traditional wave beam forming method for working out an optimal weight value based on a gradient method (such as a minimum mean square error algorithm and a linear constraint minimum variance algorithm), the method improves the processes of intersection and variation on the basis of the Memetic algorithm (a mixed inheritance algorithm), thereby enhancing the convergence speed and the local searching capability of the algorithm. The method can search the weight vector optimal value of an antenna array, carry out the self-adapting zero control of an interference direction, also enable a maximum gain main lobe to be directed at the direction of a desired signal and enhance the signal-noise ratio of the system.

The invention discloses a memetic algorithm-based method for researching optimal configuration of switches of power distribution networks. The method comprises the following steps of comprehensively considering the balance of economy and reliability by taking a minimum power supply zone as a basic analysis unit, calculating the differences between a manual switch and a remote control switch, and establishing a mathematic model for optimizing a section switch and an interconnection switch at the same time on the basis of a cost-benefit theory; importing a memetic algorithm, and combining an intelligent algorithm and a local search strategy to optimize the mounting quantity and positions of the section switch and the interconnection switch; and carrying out analysis by applying an IEEE RBTS-BUS 2 system, and the results show that load capacities at two ends of the switches need to be basically balanced and then the effectiveness of the method is verified.

The invention discloses a multi-modulus blind equalization algorithm (MMA) optimized by a Memetic algorithm (MA). The concepts of individual evolution, social behaviors among individuals and the like are introduced into a blind equalization technology; the reciprocal of an MMA cost function is defined as a fitness function of the MA; individual optimal vectors are searched in a global scope by means of the population optimization mechanism and local area deep-searching capability of the MA and are taken as the initial optimization weight vectors of the MMA. Iteration is performed through the MMA to obtain an optimal weight vector of the MMA. Compared with a constant modulus blind equalization algorithm (CMA), the MMA and a genetic algorithm-based multi-modulus blind equalization algorithm (GA-MMA), the algorithm disclosed by the invention has the advantages of high convergence speed during equalization of high-order multi-modulus signals, smallest steady-state errors and clearest output signal constellation map.

The embodiments of the invention provide a wind power generation power determining method and device. The method includes the following steps that: a Gauss regression training model is constructed, and the hyper-parameters of the Gauss regression training model are determined; the hyper-parameters of the Gauss regression training model are screened by adopting a memetic algorithm, and screened-out hyper-parameters are utilized to construct a new Gauss regression training regression model; the new Gauss regression training regression model is utilized to perform computation processing on a first sample data set, so that wind speed data at a wind power generation power prediction time point is obtained; and the new Gauss regression training regression model is utilized to perform computation on the wind speed data at the wind power generation power prediction time point and a second sample data set, so as to obtain predictive wind electricity power. According to the method provided by the technical schemes of the invention, the memetic algorithm is adopt to screen the hyper-parameters of the Gauss regression training model; the screened-out hyper-parameters are utilized to construct the new Gauss regression training regression model; the new Gauss regression training regression model is utilized to predict the wind electricity power; and therefore, the efficiency of the prediction of the wind electricity power can be improved, and the accuracy of prediction results can be improved.

The invention discloses a memetic framework-based multi-target dynamic network community division method, which comprises the following three steps of: step 100, establishing a memetic algorithm framework; Step 200, under a memetic framework, weighting the modularity density function D to obtain an optimized modularity density function D, and embedding the optimized modularity density function D and the normalized mutual information NMI into a cost objective function to obtain a minimized optimization objective function; Step 300, adopting direct integer coding mode, combining an initialization mechanism based on identifier transmission, a two-way cross genetic algorithm and a searching mode of a self-climbing algorithm to obtain the optimal community structure, the population diversity ishigh, the searching space is small, fine division of the community structure can be achieved, meanwhile, the algorithm efficiency is high, and the community division precision is fine.

The invention provides an off-line service task scheduling method for remote health monitoring with hard time windows. According to the invention, firstly a mathematical model for off-line service task scheduling of remote health monitoring is built, and then a meme algorithm is proposed for solving. The off-line service task scheduling method specifically comprises the steps of (1) building a mathematical model for off-line service task scheduling of remote health monitoring; (2) performing parameter initialization; (3) randomly initializing the location of firefly individuals, and determining the brightness corresponding to each firefly individual through calculating an objective function value of each firefly individual; (4) calculating the relative attractiveness and brightness of eachfirefly individual; (5) updating the location of the fireflies; (6) updating the brightness of each firefly individual; (7) performing local search on a chaotic neighborhood solution of an optical solution by using a simulated annealing algorithm; and (8) judging whether the number of iterations satisfies the maximum value or not, and obtaining an optimal solution. The off-line service task scheduling problem of remote health monitoring can be effectively solved according to the invention.

The invention discloses a wireless rechargeable sensor network optimization method based on a memetic algorithm. The problem that the wireless rechargeable sensor network is low in network optimization speed and not ideal in optimization effect is mainly solved. The method includes 1) constructing a wireless rechargeable sensor network; 2) setting memetic algorithm parameters; 3) coding each individual in the population by adopting a direct coding mode; 4) performing cross operation and variation operation on the population; 5) selecting an optimal individual group from the variation population; 6) carrying out the redundant detection and hole detection on the optimal individual group; 7) selecting an optimal individual and charging the optimal individual; 8) determining whether the circulation algebra of the current memetic algorithm reaches the maximum number of iterations or not, if yes, executing the step 9; otherwise, adding 1 to the circulation algebra of the memetic algorithm, and returning to the step 4 ); and 9) outputting the service life of the wireless rechargeable sensor network. According to the invention, the optimization speed of the wireless rechargeable sensor network is increased, and the service life of the network is effectively prolonged.

The invention provides a flight conflict resolution method and a flight conflict resolution device. The method comprises steps: according to a four-dimensional trajectory of a to-be-optimized aerocraft, a flight conflict situation estimation model is acquired, wherein each individual in the model comprises the departure delay time for all to-be-optimized aerocrafts, and different individuals are different; according to a target function, a target function value corresponding to the above model is acquired, wherein the target function is built according to the flight conflict situations between all to-be-optimized aerocrafts; all individuals in the above model are divided into M groups, and as for each group, a memetic algorithm is adopted for preset times of variation; the M groups are arranged in a cycle, the optimized individual in each group is copied to the next group sequentially to replace the poorest individual in the next group, an updated model and the corresponding target function value are acquired, and the model corresponding to the larger value in the two target function values is kept. The computing efficiency is high, the flight conflict resolution efficiency is improved, and the average departure delay time is low.

The invention discloses an argon atom cluster structure optimization technology based on a Memetic algorithm. The breadth advantage of group algorithm search and the depth advantage of a local search algorithm are combined, Lennard-Jones potential energy is taken as an evaluation function of an optimization algorithm, the diversity of particles is increased through cross and variation of individuals, the optimization search efficiency is improved by using preferential selection and local search of the individuals, performance in aspects of computation complexity, rapid convergence, the global situation and the like are considered comprehensively, and the global optimum stable structure of an atom cluster is obtained.

The invention discloses a metabolin MS / MS (mass spectra) computer simulation method. According to the method, optimization design is conducted on a fragmentation rule through an efficient Memetic algorithm, and the molecule mass spectra specificity serves as the fitness function value of an optimizing individual, so that the formed simulation method has the theoretically global optimum mass spectra distinguishing ability, and the accuracy of the metabolin identification can bee improved effectively. In the optimizing process, a sparse fitness function value is added to be used for guiding the optimizing individual, and it can be guaranteed that the formed fragmentation operation tree can be provided with the minimum redundancy. Accordingly, in the less molecule operation steps, the identification mass spectra with higher specificity is obtained, and robustness brought by the complex analysis process in an existing algorithm is avoided effectively. Finally, the method is not dependent on the real mass spectra and molecule data input particularly, and the formed simulation mass spectra data have the generality and can be used for establishing a general metabolin identification data base.

The invention discloses a transfer function identification method based on a memetic algorithm. The method includes the steps of generation of data, encoding, generation of an initial group, crossover, mutation, calculation of fitness functions, selection, local search and judgment of whether a stop condition is met. The method combines the breadth advantage of a group search algorithm and the depth advantage of a local search algorithm, a transfer function output response error is used as an evaluation function for algorithm optimization, the diversity of particles is improved through crossover and mutation of individuals, the optimal parameter of a transfer function is determined through the minimum evaluation function, and then the system transfer function fit with experimental data is obtained through speculation.

The invention discloses an optimal control method for an integral firing neuron model, which utilizes the advantages of the breadth of the group search and the depth of the local search in the Memetic algorithm, and uses the potential firing time error of the integral firing neuron model as the evaluation function of the optimization algorithm , the diversity of particles is increased through the crossover and mutation among individuals, and the optimal control sequence of the integral firing neuron model is determined by minimizing the evaluation function, so as to achieve the control purpose of completing the firing of neurons at the desired time.

The invention discloses a three-dimensional box loading method based on a three-dimensional moving mode sequence and a memetic algorithm. The method mainly solves the problem of low utilization rate on the volume of a three-dimensional box loading container in the prior art. The three-dimensional box loading method comprises the following realization steps that 1, each parameter is set; 2, an initial population is randomly generated, and the adaptive fitness of individuals in the population is calculated; 3, whether the termination conditions are met or not is judged, if so, the step 4 is executed, and otherwise, the step 9 is executed; 4, a binary tournament method is used for selecting the individuals; 5, the individuals are crossed, and the individual adaptive fitness value is calculated again; 6, the individuals are subjected to variation, and the individual adaptive fitness value is calculated again; 7, the individual with the greatest adaptive fitness value in the current generation is stored; 8, the number of the iteration times is added to 1, and the operation returns to the step 3; 9, a hill climbing method is used for optimizing the individuals with the greatest adaptive fitness value, and the optimized box loading result is output. The method has the advantages that the volume utilization rate of the container can be improved, and the method can be used for solving the box loading problem, and can also be used for soling other combination optimization problems.

The invention discloses an image retrieval method based on memetic algorithm, and relates to shape-based image retrieval. The image retrieval method comprises setting parameters; generating an initial population; calculating antibody affinity; cloning; executing clonal variation based on probability; performing clonal selection; recombining; subjecting the antibody to local search operator optimization based on simulated annealing algorithm; optimizing superior antibodies by using a local search operator (1) and a local search operator (2); and repeating the operations to realize rapid and effective image retrieval. By combining the clonal selection algorithm with the local search operators, the image retrieval method provided by the invention has high global search capacity, high convergence rate and high image retrieval efficiency. The local search operators with high local search capacity can further improve the retrieval result of the clonal selection algorithm so as to improve the accuracy of the image retrieval results. In addition, the method can overcome the difficulty in determining the number of classes by using a coding method based on class marks. Based on the advantages of high efficiency and high accuracy, the image retrieval method provided by the invention can be used for retrieving and classifying network pictures.

The invention relates to an improved memetic algorithm-based project code module redundancy optimal distribution protection method. The method comprises the steps of 1, performing module-based multilevel segmentation on a project code to obtain a multilevel structure corresponding to the code; 2, encoding a solution of the multilevel structure and building a reliability model; and 3, solving the reliability model by utilizing an improved memetic algorithm, wherein an obtained optimal solution is an optimal redundancy set of modules in the multilevel structure. The improved memetic algorithm improves a crossover operator and a local search operator in a MA to improve the optimization precision and speed; and a higher-reliability solution can be obtained when a target model is optimized by utilizing the improved algorithm.

The invention discloses an image retrieval method based on memetic algorithm, and relates to shape-based image retrieval. The image retrieval method comprises setting parameters; generating an initial population; calculating antibody affinity; cloning; executing clonal variation based on probability; performing clonal selection; recombining; subjecting the antibody to local search operator optimization based on simulated annealing algorithm; optimizing superior antibodies by using a local search operator (1) and a local search operator (2); and repeating the operations to realize rapid and effective image retrieval. By combining the clonal selection algorithm with the local search operators, the image retrieval method provided by the invention has high global search capacity, high convergence rate and high image retrieval efficiency. The local search operators with high local search capacity can further improve the retrieval result of the clonal selection algorithm so as to improve the accuracy of the image retrieval results. In addition, the method can overcome the difficulty in determining the number of classes by using a coding method based on class marks. Based on the advantages of high efficiency and high accuracy, the image retrieval method provided by the invention can be used for retrieving and classifying network pictures.

The invention discloses a method for optimizing microbial fermentation control based on a Memetic algorithm, which comprises the following steps: establishing a microbial fermentation data set, constructing a BP neural network, using the training data set to train the BP neural network, and controlling parameters for microbial fermentation Binary coding, execution of crossover operator, execution of local search operator by hill-climbing algorithm, execution of mutation operator, execution of local search operator again, the selected better individual as the current solution, and the qualified BP neural network trained in step 4 As a fitness function, calculate the fitness value of each individual in the population, and then update the population through the selection operator, find the optimal individual from the new population, and record the optimal combination of control parameters. The present invention can acquire the optimal control parameter combination according to the existing fermentation data without redesigning the experiment.

The invention discloses a public opinion propagation inhibition method based on a smart community big data knowledge graph. The method comprises the following steps: S1) constructing a knowledge graph and a community network based on a smart community public opinion propagation inhibition model; S2) dividing the community network into a plurality of communities by adopting a community detection algorithm, and converting a community detection problem into an optimization problem by adopting a BGLL algorithm; S3) selecting a corresponding node to obtain a candidate immune node set; and S4) optimizing a propagation threshold function by using an improved Memetic algorithm, selecting a final immune node from the candidate immune node set, and inhibiting the propagation of bad public opinions in the smart community. Deep structured analysis is carried out on the big data knowledge graph of the smart community, low-cost, small-influence and high-efficiency public opinion suppression can be realized through the big data knowledge graph based on the smart community and an artificial intelligence public opinion suppression algorithm, the management effect of a manager is improved, and the management burden is reduced.

The invention provides a face image gender detection method and system wherein the face image gender detection system comprises an image obtaining module, a controlling and processing module and a saving module. The method comprises the following steps: extracting the Gabor wavelet characteristic vector of an original face image and the Gabor wavelet characteristic vectors of all the face images in a face image library through the use of the Gabor algorithm and the Memetic algorithm; establishing a characteristic vector set; finding from the characteristic vector set the characteristic vector that is closest to the Gabor wavelet characteristic vector of the original face image; and using the gender of the face image corresponding to the closet characteristic vector as the gender of the original face image. The method and system do not require dimensionality reduction through the means of PCA, and can obtain identification data with better discrimination ability in a short processing period. Therefore, the identification performance is increased and so is the identification accuracy.

The invention discloses a multi-target reactive power optimization method for an electric system, which belongs to the field of reactive power optimization for electric systems. The method includes: modifying the Memetic algorithm to adapt to multi-target optimization, applying the modified Memetic algorithm to the problem of multi-target reactive power optimization for the electric system, and working out a Pareto optimal solution of the multi-target problem; and judging whether algorithm convergence conditions are met or not, and if yes, completing optimization and outputting optimization results. The multi-target reactive power optimization method has the advantages that the algorithm for solving the problem of multi-target reactive power optimization is provided, the method is more suitable for solving the multi-target problem while giving play to existing advantages of the Memetic algorithm which integrates local searching and evolutionary computation and has high global search capacity and the like, and searching efficiency is improved while algorithm robustness is improved.

The invention discloses a wireless rechargeable sensor network optimization method based on a memetic algorithm. The problem that the wireless rechargeable sensor network is low in network optimization speed and not ideal in optimization effect is mainly solved. The method includes 1) constructing a wireless rechargeable sensor network; 2) setting memetic algorithm parameters; 3) coding each individual in the population by adopting a direct coding mode; 4) performing cross operation and variation operation on the population; 5) selecting an optimal individual group from the variation population; 6) carrying out the redundant detection and hole detection on the optimal individual group; 7) selecting an optimal individual and charging the optimal individual; 8) determining whether the circulation algebra of the current memetic algorithm reaches the maximum number of iterations or not, if yes, executing the step 9; otherwise, adding 1 to the circulation algebra of the memetic algorithm, and returning to the step 4 ); and 9) outputting the service life of the wireless rechargeable sensor network. According to the invention, the optimization speed of the wireless rechargeable sensor network is increased, and the service life of the network is effectively prolonged.

The invention provides an SAR image search method based on sparse coding classification. The SAR image search method aims at the defects of an existing image search system and method. Through extraction of characteristics and construction of an over-complete dictionary, solution is carried out through sparse representation based on a dual Memetic algorithm, a sparse representation classifier is trained, the classification process with supervision can be achieved in the classification process, the sparse solution with global optimum can be solved fast, and then search results are output from high to low according to similarity. When the problem of image classification is solved, the method achieves the good improvement effect on classification accuracy, search content similarity, calculating complexity and result robustness.

The present invention discloses a Memetic algorithm based microbial fermentation control and optimization method comprising the following steps: establishing a microbial fermentation dataset, constructing a back-propagation (BP) neural network, using the training dataset to train the BP neural network, performing binary coding on microbial fermentation control parameters, executing a crossover operator, adopting a hill-climbing algorithm to execute a local search operator, executing a mutation operator, executing the local search operator again, taking the selected preferred individuals as a current solution, taking the BP neural network that is qualified in the training in step four as a fitness function to calculate the fitness value of each individual in a population, then updating the population by virtue of a selected operator, finding optimum individuals from a new population and recording an optimum control parameter combination. The Memetic algorithm based microbial fermentation control and optimization method can be used for obtaining the optimum control parameter combination according to the existing fermentation data, and the redesign an experiment is not needed.