53 results about "Evolution algebra" patented technology

The invention discloses a structure robustness optimization design method containing interval parameter uncertainty. The method comprises the following steps that a structure robustness optimization design model based on an internal is built; sample points are obtained by adopting a Latin hypercube sampling and co-simulating technique; a Kriging proxy model for predicting a target function and a constraint function is constructed; an interval robustness optimization design model is solved by adopting a double layer-nested genetic algorithm, left boundaries and right boundaries of the target function and the constraint function are calculated in the inner layer of the genetic algorithm, and in the outer layer of the genetic algorithm, total interval constraint violation degree vectors of all design vectors are calculated, and the feasibility of the total interval constraint violation degree vectors is judged; all the design vectors are subjected to advantage and disadvantage sorting according to a superior relationship criterion based on the interval constraint violation degree vectors; when a largest evolution algebra or convergence threshold value is achieved, the optimal solution of the robustness optimization design model is output, and the structure robustness optimization design containing interval parameter uncertainty is achieved.

The invention discloses a surface roughness prediction method based on GA-GBRT and a method for optimizing process parameters. The method comprises the steps of: collecting data to construct a data set, and dividing the data set to training set data and test set data, and employing the training set data to perform training of key parameters of a GBRT model; b, performing parameter coding and population initialization: randomly generating a chromosomal sequence for increasing the number of iterations, the maximum depth of the individual regression estimator and the learning rate; c, employing the k-folded cross-validation method to train the GBRT model, and employing the genetic algorithm to calculate the fit goodness fitting value of each individual; d, when the number of cycles does not reach the maximum number of iterations, allowing the population to be selected, crossed and mutated to produce a new generation of populations, and continuously performing training of the GBRT model; and e, repeatedly performing the steps c and d until the number of cycles reaches the maximum evolution algebra or exceeds the maximum number of iterations to obtain the optimal model parameters. The surface roughness prediction method based on GA-GBRT and the method for optimizing process parameters are high in test precision and superior in prediction performance and improves the surface processing precision of the workpiece.

The invention discloses a genetic evolution image rebuilding method based on a Ridgelet redundant dictionary, and the method is used for solving the problem that the image rebuilt by the existing L0 norm rebuilding technology is poor in visual effect. A rebuilding process comprises the following steps of: clustering all partitioning observation vectors according to the level of similarity by selecting a proper clustering algorithm; initiating clusters; carrying out the common genetic evolution on the initiated clusters; rebuilding an initial image; updating by means of filtering and convex projecting; judging whether evolution algebra reaches a maximum value or not; updating the sparsity; updating the clusters; carrying out the independent genetic evolution on the image blocks; and rebuilding the image. In the method, the similar clusters of the image blocks are used, and the optimal Ridgelet redundant dictionary base atom is found for each image block of each cluster by a genetic evolution computation thought, so that the time complexity of the algorithm is reduced, the blocking effect in the rebuilt image is removed by means of filtering and convex projecting, the search space of the optimal solution is shortened, the image is high in rebuilding precision, and the image is good in rebuilding effect, so that the method can be used for the fields of image processing and computer vision.

The invention discloses a main line green wave coordination control signal time method for optimizing exhaust gas emission. The method comprises the following steps that first, the basic traffic parameter of a main line is surveyed and obtained, and a vehicle exhaust gas emission calculating platform is initialized; second, the basic parameter of a multi-objective genetic method is set, and a population of the multi-objective genetic method is initialized; third, based on the platform, the adaptive degrees of all individuals in the population are calculated; fourth, the non-domination sequence and the visual adaptive degree of the individuals in the population are calculated, a progeny population is generated through genetic section, genetic cross and genetic variation, and the adaptive degrees of all individuals of the progeny population are calculated; fifth, the population and the progeny population are combined to obtain a new population, the non-domination sequence and the congestion degree of all individuals of the new population are calculated, the individuals are chosen based on the non-domination sequence and the congestion degree, and the next generation population is obtained; six, when evolution algebra is larger than the best evolution algebra, execution of the method is completed, all the individuals with the non-domination sequence being equal to 1 in the last generation population are used as a final noninferior solution to be output, and the timing scheme in which the vehicle average delay and vehicle exhaust gas emission are comprehensively considered is obtained.

The invention discloses an image enhancement method based on an adaptive immunity genetic algorithm. The image enhancement method based on the adaptive immunity genetic algorithm includes steps that S1, normalizing an image pixel gray level f(x, y) to obtain n(x, y); S2, coding parameters (alpha, beta) to be optimized, randomly generating a group of initial individuals to form an initial population, and inputting a control parameter crossover probability p<c>, a mutation probability p<m>, a population size N, a maximum running algebra G and the like; S3, judging whether an evolution algebra t is equal to G, if so, ending the algorithm, and outputting the optimal solution of (alpha, beta), otherwise, turning to the next step; S4, using a roulette strategy to select M individuals, and carrying out crossover and mutation operations on the individuals according to crossover and mutation methods in genetic operation; S5, selecting two vaccines, the individuals to be vaccinated and a vaccination point number to perform immunization, making a immunization choice after the vaccination, and using the optimal individual retention strategy for the vaccinated population; S6, obtaining the corresponding nonlinear transformation function F(u) of each group of (alpha, beta), and using the nonlinear transformation function to perform an image gray level transformation to obtain an output image g(x, y).

The invention provides a water transportation pipe network leakage positioning method based on a Bayesian decision theory and a genetic algorithm, and relates to the water transportation pipe network leakage positioning method based on the Bayesian decision theory and the genetic algorithm. The water transportation pipe network leakage positioning method aims at solving the problems that the existing water transportation pipe network adopts a sound listening leakage positioning method, the work intensity is high, and in addition, the efficiency is low. The water transportation pipe network leakage positioning method is realized through the following steps that a PDD leakage model is built according to DMA; a water transportation pipe network with leakage is subjected to water pressure signal collection; according to the Bayesian decision theory, the leakage node number and the leakage quantity are used as independent variables, and the probability density of leakage accidents is used as a dependent variable to build a target function; the genetic algorithm is utilized for solving the functional expressions, after the evolution algebra is completed, individuals with the probability greater than the given value f' in the population are output, and the possible leakage occurring position is obtained; according to calculation results, monitoring personnel are assigned to the possible leakage occurring position for checking or restoring pipelines. The water transportation pipe network leakage positioning method is applicable to the field of water transportation pipe network engineering.

The invention discloses a WSN node locating method based on heterogeneous double-population particle swarm optimization. According to the method, community self-adaptation activity behavior of animals in the nature is merged in a particle swarm optimization algorithm, a fitness function is designed based on the distance between an anchor node and an unknown sensor node, two heterogeneous child populations are used for maintaining good variety by simulating the community activity mode of the animals in the nature, a composite reverse learning strategy and an elite chaos search strategy are merged in the two child populations respectively by simulating the natural law that the animals in different communities in the nature have different preferences and habits, different search modes are carried out on the two child populations, advantage complement of various search modes is achieved, so that global searching ability is enhanced, and locating accuracy is improved. Meanwhile, communicating behavior of the animals in different communities in the nature is simulated, in appointed evolution generations at intervals, individuals are exchanged between the two child populations, sharing of high-quality search information and a guiding effect are achieved, and accordingly convergence speed is improved, and locating real-time performance is improved.

The invention discloses a V-BLAST system detection method based on discrete particle swarm algorithm, which belongs to the technical filed of artificial intelligence and aims to overcome the disadvantages in the prior art, such as high calculation complexity and code error rate. The inventive method can provide better performance of code error rate and reduce the actual complexity. The method is implemented by the steps: 1. initializing a particle swarm with a size of m, randomly generating a position vector of each particle, the dimensional number of the position vector being the same as the transmitting antenna, wherein the setup variation rate is m, the evolution algebra t is zero and the condition of ending the determined algorithm is that the maximal iterative algebra is ga; 2. calculating the particle adaptability, judging whether the particle adaptability meets the condition of iterative termination, if meeting the condition, the algorithm is ended, otherwise going to the step 3; step 3, updating the particle position, executing the variation operation, updating the individual extreme value and the overall extreme value, and going to the step 2. The invention can be used to solve the detection problem of vertical layered space-time system in the field of wireless communication.

The invention provides a guidance tool error distinguishing method based on experiment design and evolutionary optimization. The guidance tool error distinguishing method comprises the following steps: S1: obtaining data information; S2: preprocessing the data information; S3: setting parameters: setting a target function, setting a constrain condition, setting a search space, setting a populationrange, and setting an evolution algebra; S4: according to an experiment design method, finishing multiple-time error coefficient distinguishing, and carrying out statistics on analysis results. According to the guidance tool error distinguishing method, under a condition of inertia/ starlight combined guidance, multi-source measurement information is used, a ballistic trajectory recurrence methodbased on evolutionary optimization is adopted to carry out distinguishing on error coefficient, meanwhile, an experiment design scheme of multiple-time optimization is put forward, the optimization effect of a genetic evaluative algorithm is improved, and guidance tool error coefficients can be effectively separated. Compared with a traditional tool error distinguishing method, the guidance toolerror distinguishing method disclosed by the invention has obvious advantages.

The invention relates to a multi-core SoC software mapping method based on multi-objective optimization. The method includes the following steps that, (1) objectives of multi-core SoC software mapping are optimized, wherein (11), decision variables in the multi-core SoC software mapping are determined, (12), constraint conditions of the multi-core SoC software mapping are determined, (13), a total objective function of the multi-core SoC software mapping is determined, (14), mapping expenditure between tasks and a processor is established based on the tasks and the processor, and (15), a corresponding expenditure matrix between communication events and the processor is established; (2), multi-objective evolution of the multi-core SoC software mapping is conducted, wherein multi-objective evolution of the multi-core SoC software mapping is conducted includes the following procedures of (21), coding, (22), initial chromosome population generation, (23), constraint regulation violation inspection, (24), fitness calculation, (25), selection, (26), variation, (27), hybridization, and (28), repeating of the procedure (23) to the procedure (27) till a preset evolution algebra is acquired.

The invention is suitable for the technical field of computer, and provides a recommendation method and device. The method comprises the following steps of: receiving history recommendation preference data; obtaining an extreme point of a recommendation target; generating a corresponding current population; calculating a target function value of the current population; selecting a preset quantity of individuals from the current population according to the target function value to group a parent population; carrying out an evolution operation on the parent population; generating a child population; when the current evolution algebra does not exceed a preset evolution algebra threshold value, combining the parent population and the child population to generate a next generation of evolution population; setting the next generation of evolution population as the current population; adding 1 to the current evolution algebra; calculating the target function value of the current population again; repeating the steps until the current evolution algebra achieves an evolution algebra threshold value; and setting the parent population as a recommendation result so as to realize correct, novel and comprehensive personalized recommendation and then improve the recommendation adoption rate.

The invention discloses a radar sub-array dividing optimization method based on a difference algorithm, comprising steps of obtaining a covariance matrix of a reception signal of the radar array and a beam pointing guiding vector, respectively arranging a radar array to divide the quantity of the sub-array, the number of the difference algorithm populations, and the variation operator, and a first generation variation rate, a crossover rate and a maximal evolution algebra in the difference algorithm, obtaining a target function and a Gth generation population XG of the sub-array of the radar array on the basis of the difference algorithm, calculating the Gth generation variation population of the sub-array of the radar array on the basis of the difference algorithm according to the first generation variation rate, obtaining the Gth generation crossover population of the sub-array of the radar array on the basis of the difference algorithm, performing comparison on the target function value corresponding to each of the crossover individuals and the target function value corresponding to each of the individuals in the XG in one-to-one correspondence to obtain the Gmth generation population X<Gm> of the sub-array of the radar array on the basis of the difference algorithm, and using the individual which has the biggest target function value in the X<Gm> as the optimal sub-array of the radar array based on the difference algorithm.

The invention relates to an improved evolutionary neural network architecture search method based on a super network. The method comprises the following steps of S1, taking an input layer as a first layer, and packaging five calculation modules; S2, binarizing the connection of the computing nodes in the neural network; S3, learning a structure weight for each computing node, and S4, constructinga parent population P by adopting a binary fancy competition selection method. S5, forming a filial generation population Q; and S6, performing mutation operation on individuals in the filial generation population Q. S7, decoding each individual in the filial generation population Q into a corresponding neural network to obtain a structural weight; and S8, combining the parent population P and thechild population Q into a population R, selecting a plurality of individuals as an original population of the next generation by adopting an environment selection method, and feeding back to the stepS4 until a predetermined maximum evolution algebra is reached, and after the evolution is finished, outputting the individual with the highest fitness value as the optimal neural network architecture.

The invention discloses a BGA (Ball Grid Array) welding spot structure parameter optimization method for reducing a power cyclic stress. The method comprises the following steps that: 1) establishinga COMSOL welding spot simulation analysis model; 2) obtaining the thermal stress value of a welding spot; 3) determining the influence factor of the thermal stress value; 4) determining a parameter level value of the influence factor; 5) obtaining an experiment sample; 6) obtaining a function relational expression between the influence factor and the thermal stress value; 7) carrying out regression analysis on the function relational expression to obtain a regression equation; 8) determining the correctness of the function relational expression; 9) adopting a random way to generate an initialpopulation; 10) obtaining a current evolution algebra gen and an optimal fitness value; 11) randomly forming M/2 groups of paired individuals by M individuals, and carrying out an interlace operationon each group of paired individuals; 12) carrying out a mutation operation on the paired individuals; 13) carrying out evolution reversion on the paired individuals; 14) selecting the individual withthe optimal fitness value; and 15) after the population is updated, carrying out judgment again. The method has the advantages of good robust performance and simple calculation method, and great convenience is brought to later-stage BGA welding spot structure parameter optimization design.

The invention discloses a return track way arrangement method for a subway vehicle segment. The method comprises following steps: firstly, collecting the required to-be-returned vehicle information and the required to-be-returned track way information and establishing parameter sets of practical problems; secondly, selecting a proper encoding mode to configure the chromosomes and making population initialization rules so as to generate Z chromosomes to form an initial population P; then, according to the practical problems, determining an optimal target function and a constraint; and finally, determining a selection rule of individuals and a specific operation mode of hybridization and variation, calculating an adaptive value of each individual and storing the individual with the biggest adaptive value according to a fitness function and repeating the operation until reaching to a maximum evolution algebra. By use of the return track way arrangement method, train tasks can be properly selected or rejected and an optimal track way arrangement scheme can be found under a resource constraint condition; a decision support is provided for subsequent daily inspection and shunting operation; and working efficiency of the vehicle segment is greatly increased.

The invention provides a Switch rectifier multi-objective optimization design method based on an NSGA-II algorithm, and belongs to the technical field of rectifier parameter optimization design. The method comprises the following steps: 1, determining an optimized component and carrying out power and volume modeling; 2, defining a target function; 3, initializing a population; 4, selecting; 5, selecting binary match; 6, crossing; 7, realizing variation; 8, combining the parent population and the offspring population to obtain a first generation population; 9, forming populations by utilizingthe parent population and the offspring population; 10, reaching the maximum evolution algebra gen; and 11, outputting a Pareto optimal solution leading edge and a parameter matrix, according to the selected target performance indexes, obtaining corresponding decision variable values, namely solutions of the optimization problem, and selecting and designing components according to the parameters.

The present invention provides an improved image enhancement method. The method comprises: S1, performing normalization processing of image pixel gray level f(x, y) to obtain n(x, y), wherein an improved normalization processing mode is employed: (img file='dest_path_image001. TIF' wi='255' he='183'/); S2, performing coding of parameters to be optimized (img file='dest_path_image002. TIF' wi='37' he='71'/), randomly generating one group of initial individuals to form an initial population, and inputting a control parameter crossover probability pc, a mutation probability pm, a population size N, the maximum operation algebra G and the like; S3, determining whether an evolution algebra t is equal to the G or not, if the evolution algebra t is equal to the G, finishing the algorithm, and outputting the optimal solution of the (img file='dest_path_image003. TIF' wi='37' he='88'/), or else, turning to the next step; S4, employing a roulette strategy to select M individuals, and performing crossover and mutation operation of the individuals according to the crossover and mutation method in the genetic manipulation; S5, selecting two vaccines, individual number to be inoculated and the inoculation site number to perform immunization operation, making out immunization selection after inoculation, and employing the optimal individual reservation strategy of the population after the inoculation; and S6, allowing one group (img file='dest_path_image004. TIF' wi='37' he='71'/) to correspond to one non-linear transformation function F(u), and employing a non-linear transformation function to perform image gray scale transformation to obtain an output image g (x, y).

The invention relates to a method and a device for selecting histidine terahertz absorption spectrum wavelength based on differential evolution. The method comprises the following steps of firstly, performing differential-based mutation operation and crossing operation on an initial population, so as to obtain a crossing population; respectively utilizing the initial population and the crossing population to select from the terahertz absorption spectrum of a histidine sample, and utilizing a constructed fitness function to respectively calculate the fitness of each individual of the initial population and the crossing population; reserving the individual with higher fitness, so as to obtain a new generation of population; finally, using the new generation of population as a new initial population to perform evolution and iteration, until an evolution algebra reaches the setting threshold value, and selecting the individual with highest fitness value in the final population as the selected optimal solution of the histidine terahertz absorption spectrum wavelength. The method has the advantage that by performing gradual-point depth selection on the terahertz absorption spectrum of the histidine sample, the useful information is selected, so as to obviously improve the accuracy of quantitative analysis, and obtain a good application effect.

The invention is suitable for the technical field of a computer, and provides a multi-jobshop scheduling method and apparatus. The multi-jobshop scheduling method includes the steps: when receiving a request of multi-jobshop scheduling, establishing a mathematical model of multi-jobshop scheduling, and using a preset dynamic population strategy and a dual-mode difference evolution operator to perform evolution on the individuals in the current population of the mathematical model; and when the current evolution algebra of the current population does not satisfy a preset population evolution termination condition, setting the non-dominated individuals in the evolved current population as the current population of the next generation of mathematical model, repeatedly executing the step of evolving the individuals in the current population of the mathematical model until the current evolution algebra satisfies the population evolution termination condition, and then setting the non-dominated individuals in the evolved current population as the solution of a parameter to be solved in the mathematical model so as to obtain the solution of multi-jobshop scheduling, thus improving the convergence capability and robustness during the multi-jobshop scheduling process.

The invention discloses a forest fire monitoring system node effective dynamic coverage method based on Internet of Things. The method comprises the following steps: setting parameters: a population size NP, a scale factor F, a cross probability CR and a maximum evolution algebra Gmax; Randomly generating an initial population with a population size of NP, and solving a fitness value of an NP individual; Judging whether a termination criterion is met or not, and performing DE variation operation, DE crossover operation and DE selection operation on the current population individual; Executinga local search algorithm SQI; And finally outputting the obtained best result. According to the invention, the problems of high calculation cost and low efficiency of effective coverage of forest firemonitoring Internet of Things nodes in the prior art are solved.

The invention belongs to the technical field of computers, and discloses a super-multi-objective optimization method, system and terminal based on dynamic decomposition and selection, and the method comprises the steps: randomly initializing a population P with N individuals; selecting N excellent individuals as the next generation evolution filial generation P by using the DDS, and calculating the distance between the selected individual and the hyperplane and the distance between the reference points corresponding to the individuals; initializing a filial generation population O as a null set; aiming at N individuals in the parent generation, starting circular processing; initializing R used for storing offspring individuals, and selecting a mating individual from the current parent by utilizing MatingSelection; using SBX and PM for generating a filial generation R for the two parent generations, and adding the filial generation R into the filial generation population O; and repeatedly carrying out population selection until the maximum evolution algebra is reached. According to the method, the convergence of each generation of individuals and the diversity among the individuals are used as information of mating variation of the individuals, so that progenies can be promoted to evolve towards a better direction.

The invention provides a rapid high-path coverage rate test case generation method, which comprises the following steps of: obtaining a control flow graph of a target program, and determining father-child relations in multiple nodes in the flow graph; judging whether each node is a branch node or not; obtaining a test case set, taking each test case as an individual in a genetic algorithm, and forming an initial population by a plurality of individuals; constructing a branch crossing matrix; calculating a branch deviation degree of crossing any one branch node in the current generation population according to the constructed branch crossing matrix; Calculating the branch deviation degrees of all the branch nodes in the program, and taking the sum of the branch deviation degrees of all thebranch nodes as the program deviation degree of the individual traversing program in the current generation of population; and performing iterative optimization according to the constructed branch crossing matrix and the program deviation degree by using a genetic algorithm, and obtaining the next generation of population and the program deviation degree of the next generation of population crossing the tested program until a test case covering the target path is generated or the maximum evolution algebra of the genetic algorithm is reached.