98 results about "Chromosome encoding" patented technology

A system, method and program product for selecting an algorithm and feature set to solve a problem. A perpetual analytics system is disclosed that provides a genetic algorithm for jointly selecting an algorithm and feature set to solve a problem, comprising: an evolutionary computing engine for processing data encoded as chromosomes, wherein each chromosome encodes an algorithm and a feature set; a domain knowledge store that maintains a plurality of algorithms and a plurality of features; a system for applying a generation of chromosomes to a set of data to provide a set of results; and a fitness function for evaluating the set of results to rate a performance of each chromosome in the set of chromosomes; wherein the evolutionary computing engine is adapted to evolve a subset of the set of chromosomes into a new generation of chromosomes.

The invention relates to the technical field of artificial intelligence algorithms, and provides a method for optimizing a disaster emergency decision system path. The method comprises the steps that S11, an objective function and a fitness function are defined; S12, chromosome coding is carried out; S13, operators are selected according to the fitness function; S14, the operators are crossed; S15, the operators are mutated according to the fitness function and pheromone updating guiding mutation rules; S16, a plurality of sets of optimal solutions are generated, and the best solution is output through the fitness function. The quality and efficiency of the solutions are effectively improved through the pheromone guiding mutation rules of an ant colony algorithm. A large amount of useless redundant iteration of a genetic algorithm is avoided through the positive feedback mechanism and the parallelism of the ant colony algorithm, and the solution speed of the genetic algorithm is further improved. The decision model with a single saving point and multiple disaster points based on continuous supply losses is more suitable for emergency situations, and practical significance is achieved.

The invention relates to a resource allocation global optimization method of an intelligent scheduling system. The resource allocation global optimization method is implemented by adopting a new genetic algorithm. The resource allocation global optimization method considers the difference between a full-load driving path and a no-load driving path of an AGV, adds a conversion coefficient alpha before the no-load driving path, the value of the conversion coefficient alpha ranges from 0 to 1, and establishes a new mathematical model for AGV global optimization scheduling f=min max{d1,d2,...,dn}.Based on the improvement of a crossover operator, a best-worst crossover mode is proposed on the basis of a hybrid crossover mode combining two-point crossover and BCBRC, thus the problem that a non-feasible solution is prone to generate due to changes of chromosome coding rules is avoided; and a variation pattern of gene segment random exchange is adopted, thereby avoiding the problem that the conventional variation mode is prone to generate a non-feasible solution containing the same nodes when a real number system coding method is adopted.

The invention discloses a multi-target workshop scheduling method based on a multicolor set genetic algorithm, and the method comprises the steps: converting multi-target optimization into single-target optimization according to constraint conditions and a target function through a random weight coefficient method; building a process-equipment contour matrix constraint mode, carrying out the chromosome coding, and reducing the size of a GA search range. The method can assign a reasonable weight for each target according to the management demands through employing a preference matrix, and enables a multi-target scheduling problem to be converted into a single-target scheduling problem, thereby reducing the scheduling complexity. In addition, the method carries out the chromosome coding through the process-equipment contour matrix constraint mode, enables the size of the GA search range to be precisely and effectively reduced, and improves the solving efficiency and precision.

The invention provides an optimization algorithm based on multi-objective resource-constrained project scheduling model. The scheduling model requires scheduling the start time of each activity to achieve the optimal goal under the condition of satisfying the relevant constraints. Based on the RCPSP model, the invention introduces the optimal resource balance as the objective, and expands the model to a multi-objective model. The solution of RCPSP is mainly based on the heuristic algorithm. When the task list is used to encode chromosomes in the heuristic algorithm, the task list initialized randomly may not satisfy the constraint relation between the top and bottom. The invention provides an individual generation mode based on control relation, which presents a new crossover operator andmutation operator based on the NSGA-II algorithm. The invention can greatly reduce the time complexity of the algorithm, realize the balanced allocation of resources, improve the production efficiencyand save the production cost while ensuring the solution precision of the algorithm, thereby improving the economic benefit of the resource scheduling production process.

The invention encompasses a recombinant bacterium that comprises at least one vector capable of expressing a nucleic acid sequence encoding an antigen. In particular, the bacterium comprises at least one chromosomally encoded essential nucleic acid that is altered so that it is not expressed, and at least one extrachromosomal vector.

The invention provides an LSSVM (least squares support vector machine) wind speed forecasting method based on integration of GA (genetic algorithm) and PSO (particle swarm optimization). The method comprises the following steps: finite wind speed samples are divided into a training set and a testing set, and normalization processing is performed; GA and LSSVM related parameters are initialized; chromosome coding is performed, and initial population is generated randomly; the fitness corresponding to each chromosome is calculated, if requirements are met, the PSO in the fifth step is started directly, and if the requirements are not met, selection, crossover and mutation operation of the GA are performed; optimum parameter combination obtained with the GA is used for initializing the PSO related parameters; the optimum position fitness value of each particle is compared with the optimum position fitness value of the swarm; the final optimum parameter combination is output, and an optimized LSSVM model is obtained; a forecast wind speed time history spectrum is obtained. The LSSVM wind speed forecasting method based on integration of GA and PSO has the characteristics of high optimization precision, high convergence precision, fewer iterations, high success rate and the like.

The invention relates to a method for using a gen etic algorithm to solve an optical covering problem of a wireless sensor network node, which comprises dividing a problem model to an 0/1 programming problem when using the gen etic algorithm as an optimal tool to solve problems, and then, using a gen etic algorithm of binary code to solve the problems, coding colored bodies of the gen etic algorithm to a 0/1binary string in the algorithm, and then, optimizing through an evolutionary mechanism. Sensors with the number of N are scattered at random, and an individual coding is a 0/1binary string with N bit length. When the sensors are optimally selected, if a sensor is selected, and then, a relative bit in the individual coding is set to be 1, or the bit is set to be 0. An individual coding and a real network structure are directly corresponding to each other through the coding mode, the gen etic algorithm not only is easy to understand, but also is simple to achieve, and is convenient to use.

The invention discloses a power distribution network planning method based on an improved genetic algorithm and a PRIM algorithm. The method comprises: establishing a power distribution network planning model; secondly, using an improved genetic algorithm to solve the optimal station address and number of the medium-voltage power distribution station and the capacity of the selected transformer, and enhancing the genetic algorithm by improving chromosome coding, a fitness function and a genetic operator; thirdly, using an improved PRIM algorithm to solve feeder line optimal paths between the high-voltage transformer substation and the medium-voltage transformer substation, between the medium-voltage transformer substation and the load center and between the medium-voltage transformer substation and the load center; fourthly, executing a power distribution network planning method based on an improved genetic algorithm and a PRIM algorithm on the test network to obtain an optimal arrangement planning scheme of the transformer substation and the medium-voltage feeder line, and determining an optimal power distribution network planning scheme by calculating economic and reliability indexes; and fifthly, performing load flow calculation by adopting a forward-backward sweep method to verify the practicability of the planning scheme. The method has the advantages of being high in searching speed and suitable for solving the planning problem of the large-planning power distribution network.

Provided is an integrated scheduling method for key loading and unloading resources of an automatic container terminal. The integrated scheduling method comprises the steps of setting chromosome coding modes, initializing population, defining the fitness function, setting the interlacing operation guideline and the mutation operation guideline, setting the parental generation selection strategy and the filial generation receiving strategy, using the fitness function to conduct chromosome iterative computation, and when meeting the iterative stopping criterion, stopping iteration and outputtingthe optimum solution. The integrated scheduling method can reasonably control device operation and space resource distribution of the container terminal, improves the operation efficiency of the automatic container terminal, reasonably distributes devices and space resources, avoids congestion, shortens the driving time of three types of devices, and optimizes space resource usage of a storage yard.

The present invention discloses a site operation intelligent scheduling method. The method comprises the steps of: respectively establishing corresponding mathematic models for a plurality of site operation targets, wherein the site operation targets comprise the shortest task completion time, the highest task completion quality, the highest resource utilization rate, the shortest queuing time andthe most balance load; collecting tasks, personnel and resources in the site operation; randomly generating a plurality of populations according to the tasks, personnel and resources in the site operation; performing numbering of the tasks, personnel and resources in the populations, and generating a plurality of initial chromosome codes; and performing solution of the mathematic models corresponding to the targets according to a preset improved genetic algorithm and the initial chromosome codes, and obtaining an optimal solution as a site operation scheduling scheme. The method can achieve multi-target optimization of site maintenance, the algorithm process of target optimization is simple, and the efficiency of the site maintenance is improved.

The invention relates to an intelligent test paper method based on a genetic particle swarm optimization algorithm, comprising: generating the objective function corresponding to each constraint condition according to the constraint condition corresponding to test paper attribute information, and calculating the fitness function of test paper according to the objective function corresponding to each constraint condition; obtaining test questions from an item bank to form a plurality of pieces of test paper, and performing chromosome coding on each piece of test paper, wherein each piece of test paper corresponds to a chromosome, the chromosome includes a plurality of segments, each segment of chromosome corresponds to a type of test questions, and includes a plurality of genes, and each gene corresponds to a test question; obtaining an initial population through a particle swarm algorithm; and processing the initial population through a genetic algorithm to obtain a new population to output test paper therein. According to the technical scheme, the method employs test paper attribute information as constraint conditions to generate a fitness function, and performs particle swarm algorithm and genetic algorithm treatment on test paper according to the fitness function, thereby obtaining test paper meeting user needs.

The invention discloses a genetic algorithm and variable precision rough set-based PET/CT high-dimensional feature level selection method. According to the method, on one hand, a chromosome coding value, a minimum reduction number of attributes, attribute dependency and the like are comprehensively considered to construct a universal fitness function framework, and different fitness functions arerealized by adjusting weight coefficients of factors; and on the other hand, for the limitation of a Pawlak rough set model, a classification error rate beta is introduced for broadening strict inclusion of lower approximation in the Pawlak rough set model to partial inclusion, so that the concept of an approximation space is perfected, the noise processing capability is enhanced, and the beta range is continuously changed to realize different fitness functions. Experimental results show that different weight coefficients greatly influence the results under the condition of consistent classification error rate; and likewise, under the condition of consistent weight coefficient, the classification error rate is increased constantly, the experimental results have relatively large difference,and a parameter combination most suitable for the method can be found according to data in the method.

The invention discloses a clustering method based on an ecological niche genetic algorithm with a diverse radius technology. The clustering method based on the ecological niche genetic algorithm with the diverse radius technology comprises the following steps that (1) chromosome coding and population are initialized; (2) the individual fitness is calculated; (3) the position, content and number of the ecological niches in the population are identified by adopting a dynamic identification method; (4) the radius information of each ecological niche is adjusted by executing the diverse radius mechanism; (5) the new individual fitness is recalculated by applying a fitness sharing function; (6) selection, intersection and mutation operations are executed; (7) an elite strategy is executed to replace the worst individual in the population; (8) if a termination condition is met, the operation is terminated, otherwise the step (5) is executed. The clustering method based on the ecological niche genetic algorithm with the diverse radius technology has the advantages that the clustering effect is good, and the stability is good.

The invention discloses a method for carrying out AFDX network path optimization by a genetic algorithm. The method is virtual link path optimization carried out on an AFDX network configured with a VL path. The method disclosed by the invention comprises the following steps of: establishing a connection matrix and a virtual link path group of a switch at first; and then carrying out chromosome coding on a virtual link path; finally carrying out genetic operations of crossing, variation and selection on a chromosome group, obtaining the optimal chromosome in case of meeting an end condition, and extracting out a valid virtual link path, wherein the path is used as the optimized virtual link path. According to the method disclosed by the invention, path optimization for the virtual link VL in the configured AFDX network is solved, and real-time path optimization for the virtual link VL in the configured AFDX network is carried out by applying the genetic algorithm, thus improving the message transmission real-time performance of the AFDX network.

The invention relates to a two-layer genetic integer programming-based complex system DSM (Design Structure Matrix) reconstructing method which can be applied to the industrial fields of aerospace, cars, ships and the like. According to the method, the simultaneous optimization of the element sequence and the clustering scheme in a DSM is realized by adopting a double-segment chromosome coding technique; and layered solving is carried out on a DSM clustering problem by adopting an integer genetic programming algorithm so as to obtain a reconstructed optimal DSM. The method comprises the following steps of firstly building an optimization model through taking DSM-based contact information flow as output and carrying out two-layer optimization on the model; obtaining a preliminary DSM clustering scheme by adopting a genetic integer programming method in the first-layer reconstruction; and carrying out a second search on each cluster in the preliminary scheme by adopting the same algorithm in the second-layer reconstruction so as to obtain a final DSM reconstruction result. Therefore, the method has the advantages of simplifying the design process, shortening the development time and increasing the resource utilization rate.

The invention relates to the technical field of radar communication and discloses a method for generating an orthogonal phase coded signal. The method specifically comprises the following steps of: 1, constructing a Walsh orthogonal matrix; 2, generating a chromosome code; 3, forming an initial population and decoding; 4, calculating a sum E of self-correlation side lobe energy and cross-correlation energy of a signal obtained by decoding the chromosome code in the population; 5, selecting an individual of which the E is small; 6, performing interlace operation on the individual; 7, performing mutation to generate a new individual; and 8, updating the population and repeating the operation in the steps from 4 to 7, and thus obtaining the best chromosome. The orthogonal matrix is subjectedto random column rearrangement and random row extraction by using a genetic algorithm, so the self-correlation side lobe energy and the cross-correlation energy of the orthogonal signal can be optimized on the premise of no change of the zero orthogonality, and the orthogonal phase coded signal with high performance can be acquired with lower calculation cost.

The invention discloses an unreliability test optimizing method based on the grouping genetic algorithm. The method comprises taking all tests under every test procedure as a group; applying the genetic algorithm to every group to obtain the optimal test set under the corresponding test procedure, wherein in the genetic algorithm, the setting rules of an individual fitness function includes that, when a target detection rate selected to be tested in a test set corresponding to an individual chromosome code does not reach the target detection rate in a preset test procedure, the fitness value is zero, otherwise, the lower the test cost sum selected to be tested in the test value corresponding to the individual chromosome code is, the greater the value of the fitness function is; combining the optimal test set obtained under every test procedure to obtain a general test set, and if the detection rate of the general test set does not reach a general target detection rate, adding in unselected tests one by one until the detection rate of the general test rate meets the general target detection rate. The unreliability test optimizing method based on the grouping genetic algorithm improves the optimizing efficiency under the condition of ensuring the precision.

Disclosure is a multi-objective semiconductor product capacity planning system and method thereof. The system comprises a data input module, a capacity planning module and a computing module. The machine information of the production stations, the product information and the order information are input by the data input module. According to the demand quantity of order, capacity information and product information, the capacity planning module plans a capacity allocation to determine the satisfied quantity of orders. The capacity allocation information is used to form a gene combination by chromosome encoding method. The computing module calculates the gene combination several times to generate numerous candidate solutions by a multi-objective genetic algorithm. The numerous candidate solutions sorts out and generates a new gene combination, and repeats the calculation to form candidate solution set until stop condition is satisfied. The candidate solution set is transformed into numerous suggestive plans as options.

The invention relates to a genetic-algorithm-based energy efficiency routing spectrum allocation method for multi-cast optical forest optimization. According to a multi-cast request, a plurality of shortest paths, meeting a service need, between a source node and all multi-cast destination nodes are calculated and the multi-cast destination nodes are divided to obtain all optical sub trees of an optical forest; a chromosome coding format of a genetic algorithm is designed to express destination node division of the optical forest and an optical path set of the optical forest; an energy efficiency fitness function of the optical forest is designed and a routing, modulation and spectrum allocation plan, needing the smallest spectrum number and lowest transmitter power consumption, for optical forest transmission is selected based on the fitness function; on the basis of corresponding crossover and mutation operations of probabilistic gene bits of the genetic algorithm, a new multi-cast optical forest is obtained and an optical forest plan with excellent energy efficiency is selected by using a lowest fitness function value; and when multi-cast transmission does not complete and other requests for transmission completion are found out in an optical network, transmission from the optical forest to optical trees of the multi-cast unit is reconfigured, so that resources occupied by the optical forest are released and low-energy-efficiency transmission is realized.

The invention discloses a flexible job shop scheduling method based on an improved genetic algorithm. The method can solve the problem that an existing genetic algorithm is insufficient in usability in a discrete flexible job shop scheduling problem. The traditional genetic algorithm has a good global search capability, but the local search capability is insufficient, the premature convergence iseasy, the optimal solution set is difficult to find, the Powell search has a strong local search capability, but the defect that the Powell search is liable to be trapped in the local optimization exists. According to the genetic algorithm scheme combined with the Powell search method, the excellent global search capability of the genetic algorithm can be fully utilized, meanwhile, the local search capability of the whole algorithm is enhanced through the Powell search method, early maturing of the algorithm is avoided, and the quality of a scheduling scheme is improved. In consideration of the particularity of a chromosome coding scheme of a flexible job shop scheduling genetic algorithm, a traditional Powell search method is improved to avoid generation of an infeasible solution, so thatthe robustness and the search efficiency of the algorithm are improved.

The invention discloses a planning method for underground logistics paths based on a genetic algorithm. The method comprises the steps of chromosome coding, adaptation calculation and genetic operation. Through the method, the problems that general path planning algorithms are unreasonable in path planning and insufficient in extensibility are solved.

The invention relates to a cross-region power system protection communication network planning method, comprising the steps of initializing a first generation of populations through adoption of a variable-length chromosome coding method; setting a fitness function, and evaluating fitness of chromosomes according to features of target populations; selecting parental chromosomes according to the fitness function, specifically, selecting the parental chromosomes through adoption of a selection operator, thereby enabling the parental chromosomes to be inherited by a next generation; carrying out cross processing on the parental chromosomes, and generating new chromosomes, thereby improving diversity of species; carrying out variation processing on the parental chromosomes after the cross processing, thereby improving searching capability of the populations; and combining the parental chromosomes and the newly generated chromosomes, as descendants of a current generation of populations, carrying out a next round of evolution until the preset generation number is realized, exiting a loop, and obtaining a final planning scheme, namely the chromosomes with the highest fitness in the last generation of populations. According to the method, a demand of line protection business for delay can be satisfied.

According to the vehicle scheduling method designed by the invention, KMEANS algorithm optimization is carried out before chromosome coding; the poor locality search capability is preliminarily weakened; and precocity is easily generated. A brand new coding mode is adopted to reduce the coding redundancy,convergence speed is further accelerated, in addition, a new fitness function is also designed; differential amplification selection operation is adopted, manual adjustment and variation operation are carried out after intersection, no intersection route is generated on the whole, when one target distribution point is completed, the route process of the next target distribution point is directly entered, no redundant route exists in the whole process, the standard of the optimal solution is achieved, and therefore shorter-distance route distribution can be obtained.

The invention provides an active distribution network united planning method based on an active management mode. The method comprises the steps that (1) the access capacity and the new branch capacity of a distributed power supply are used as optimization variables to carry out chromosome coding, and an evolutional algebraic threshold value is set; an initial population is randomly generated according to the constraints of distributed power supply constant volume programming, and an evolutional algebra is set to 1; (2) the comprehensive cost of the planning scheme corresponding to each chromosome in the current population is calculated; whether the current evolutional algebra reaches the evolutional algebraic threshold value is judged, and if so, the planning scheme of the lowest comprehensive cost is selected and used as the final active distribution network joint planning scheme to end, otherwise a next step is carried out; and (3) chromosomes in the current population are selected, crossed and mutated to acquire the next generation of population; the evolutional algebra is added with 1; step (2) is back. According to the invention, the running cost of a planned distribution network is determined by considering the active management mode of the active distribution network; the method is in line with actual operation; and the scheme is reasonable.

The invention provides an overhaul plan optimization method containing a distributed power distribution network. According to the overhaul plan optimization method provided by the invention, an equipment overhaul start time and a load transfer path are comprehensively considered from the angle of a network side of the power distribution network, and an NSGAII algorithm is used for processing the overhaul plan problem of the power distribution network. On the respect of chromosome encoding form, the overhaul start time and a switch action combination are jointly encoded, and meanwhile, the relation of overhaul variable sets within an entire overhaul time period is coordinated by use of multi-period overall encoding. Therefore, multi-objective and multi-constraint optimization can be realized, and the actual needs of power supply enterprises are better satisfied.