54 results about "Bacteria foraging algorithm" patented technology

The invention discloses a reactive power optimization method of an electrical power system. The bacterial foraging algorithm and the particle swarm optimization algorithm are combined to be applied to the reactive power optimization of the electrical power system. The reactive power optimization method of the electrical power system comprises the following steps: adopting a Newton-Raphson load flow calculation procedure to provide values of each state variable and a network loss value for optimal calculation; setting a reactive power optimization model; initializing a bacterial colony; invoking load flow iterative program appraisal to record an adaptation degree and an optimal value of bacteria; carrying out chemotaxis operation; remaining a good population and breeding the population; carrying out migration operation, and the bacteria die or are born again in a certain probability; and updating the bacterial colony, and outputting a reactive power optimization result after the rated number of iterations is reached. The reactive power optimization method of the electrical power system introduces the colony in an optimal solution, avoids blind and random problems, can exceed a locally optimal solution, combines an existing bacteria foraging optimization algorithm with the particle swarm optimization algorithm, lowers transmission losses by controlling an engine, reactive power output of reactive compensation equipment and tapping points of an adjustable transformer, has the advantages of being rapid in convergence, efficient and stable, and is suitable for resolving the problem of reactive power optimization in the electrical power system.

The invention discloses a hybrid intelligent optimization method and belongs to the technical field of artificial intelligence and data mining. A genetic optimization algorithm and a bacterial foraging optimization algorithm are organically combined by the hybrid intelligent optimization method. First, a preliminary superior solution is obtained through the breadth searching capacity of the genetic optimization algorithm and serves as an initial bacterial population in the posterior bacterial foraging algorithm, by fully utilizing chemotaxis, duplicating and dispersing operation of the bacterial foraging algorithm, excellent individuals are generated continuously, and finally the solution converges towards the optimal solution. On the basis of the technical scheme, further improvements are made in the four aspects of genetic selection operators, optimal joint pints, bacterial chemotaxis and duplicating operation. Compared with the prior art, the convergence speed and the convergence precision of an optimal solution set can be improved, and the hybrid intelligent optimization method is more extensive in applicability.

The invention discloses a task scheduling method and a system in cloud computing. The task scheduling method in the cloud computing comprises performing parameterization on characteristic information of tasks and classifying the tasks, computing and obtaining an optimal working node through a bacterial foraging algorithm according a classifying result and matching the working node and the task. The task scheduling method in the cloud computing has the advantages of enabling scheduling of user task groups through the cloud computing to have advantages of allowing swarm intelligence parallel search, being easy to jumping out of a local minimum and the like and being contributed to maintenance of diversity of the task group in the cloud computing due to the fact that the task scheduling and resource allocation problem in cloud computing is achieved through the bacterial foraging algorithm and satisfying user requirements and improving satisfaction of user experience.

The invention discloses a determining method of photovoltaic transmission power limit considering variable correlation. According to the method, a random planning problem is built, the photovoltaic transmission power limit maximization is set as a target function, the correlation inside a photovoltaic power station, the correlation between region loads and the correlation of the photovoltaic power generation and the node load change are considered, the constraint conditions are inspected through the sampling on the photovoltaic power output, the traditional machine unit power output and the load level on the basis of a Latin hypercube sampling method, finally, an improved bacterial foraging algorithm is used for solving a planning problem, and the photovoltaic transmission power limit conforming to the conditions is calculated. The method provided by the method has the advantages that the problem of potential safety hazard after the photovoltaic network accessing since the photovoltaic transmission power limit does not conform to the actual production practice is solved, the planning work of the photovoltaic power station can be better guided, the sampling coverage rate is increased, the calculation quantity is reduced under the condition of not influencing the precision, the early stage optimization speed is accelerated, and meanwhile, the later stage optimization precision is also ensured.

A processing scheme optimal selection method in a cloud manufacturing environment belongs to the technical field of manufacturing resource optimization. The aim of the invention is to acquire the basic attribute of the cloud manufacturing resources as an evaluation index of the selection of the cloud manufacturing resources, and to establish a processing scheme selection mathematical model based on the optimization of bacteria foraging and to produce an optimal selection method in the cloud manufacturing environment. In the invention, a multi-objective optimization mathematical model is constructed by a production cost objective function, a production time objective function, a processing quality objective function and other evaluation index objective functions, and then optimal selection of the processing scheme in the cloud manufacturing environment is carried out. The invention aims at designing a processing scheme selection method based on a bacteria foraging algorithm for manufacture resource selection in a machinery manufacturing process in the cloud environment so as to provide a reasonable suggest for an enterprise decider in processing scheme selection and to further raise the product quality and the enterprise profit.

The invention discloses a multi-robot cooperation odor source localization method based on an improved bacterial foraging algorithm. The method comprises the following steps: a smoke plume discovery stage, a smoke plume tracking stage and an odor source confirming stage; in the smoke plume discovery stage, when an odor smoke plume is not measured by robots, a whole work space is divided into a plurality of areas by utilizing a Voronoi diagram method, and each robot executes a random smoke plume search strategy in the own corresponding work area; in the smoke plume tracking stage, when the odor smoke plume is detected by the robots, next-search directions of the robots are determined through adoption of the improved bacterial foraging algorithm, and automatic tracking of an odor is achieved; and, in the odor source confirming stage, according to odor concentration values measured at positions where the robots locate and position change scopes of the robots, a position of an odor source is determined. Multi-robot distributed-type cooperation rapid localization of the odor source is achieved, odor source searching efficiency and odor source localization precision are substantially improved, and the multi-robot cooperation odor source localization method can be applied to occasions of harmful/poisonous gas detection, searching after a disaster, rescues after the disaster and the like.

The invention discloses a sensor node deployment strategy based on a chaotically optimized bacteria foraging algorithm. The strategy comprises the following steps of initializing, setting a cyclic variable, performing chemotaxis cycle, performing multiplication cycle, migrating, judging a condition by which an algorithm cycle ends, if the condition is met, ending the algorithm and outputting an optimal bacterium combination, and if the condition is not met, returning to set the cyclic variable. The strategy provided by the invention has the advantages that WSN nodes are uniformly distributed in a monitoring area in a node coverage scheme acquired by using the chaotically optimized bacteria foraging algorithm, less node redundancy is generated, hardly no coverage blank exists; compared with a random node deployment strategy, the sensor node deployment strategy is improved in node deployment strategy network coverage rate, the nodes are more uniformly distributed in the monitoring area, fewer coverage areas are repeated, redundancy of the nodes is extremely low, the purpose of WSN optimized coverage is achieved, and the optimized algorithm can effectively cover the monitoring area by using the fewer nodes, deployment costs are saved, and meanwhile monitoring time of the WSN is also greatly prolonged.

The invention discloses a wind-storage combined system scheduling method and apparatus capable of improving the wind power schedulability. The scheduling method includes: the available scheduling output of a wind-storage combined system is calculated via the current charge state of a storage battery and the wind power prediction output; a scheduling model containing the wind-storage combined system is established, and the scheduling model is formed by a target function and restriction conditions; the target function serves as the bacteria fitness value, an improved bacteria foraging algorithm is employed to solve the scheduling model; and if the solution result satisfies all the restriction conditions, the solution result is the optimal solution and is outputted. The scheduling apparatus comprises a calculation module, an establishing module, a solution module, and an output module. According to the method and the apparatus, based on an operation strategy of the wind-storage combined system, the scheduling model of the wind-storage combined system is established, the improved bacteria foraging algorithm is improved for solution, the economy of the operation of the wind-storage combined system is fully considered, the wind power schedulability is effectively improved, the wind power utilization rate is increased, and the method and the apparatus are applicable to the coordination scheduling of the wind-storage combined system.

The invention relates to a load frequency control method using a bacterial foraging algorithm improved based on a particle swarm. A load frequency control (LFC) system model of a single-area two-unitpower grid is built, and a bacterial foraging algorithm improved based on a particle swarm is used to optimize the PID controller parameters of the LFC system to control the LFC system. The bacterialforaging algorithm improved based on a particle swarm is to introduce global optimum, individual optimum and adaptive step size based on the standard bacterial foraging algorithm and the idea of the particle swarm optimization algorithm to redefine the health and migration of bacteria. The optimization speed of the optimization algorithm is improved, and the optimization accuracy of bacteria is greatly improved. Compared with the prior art, a good control effect is obtained, the dynamic performance of the load frequency control system is significantly improved, and the method has important practical significance for the improvement of the automatic control level of thermal power units.

The invention discloses an aluminum electrolysis production process multi-objective optimization method based on an adaptive-step bacterial foraging algorithm. The method includes the following steps that firstly, an aluminum electrolysis production index Y is determined, and a parameter X which has the largest influence on the aluminum electrolysis production index is selected; then the parameter X serves as input, the production index Y serves as output, and modeling is carried out on the aluminum electrolysis process through a back propagation (BP) neutral network, so that an aluminum electrolysis model is obtained; and afterwards, the output Y of aluminum electrolysis is used as a fitness function, the bacterial advancing step length is adjusted in a self-adaptive mode on the basis of Pareto differential entropy, the parameter X is optimized within a value range of the parameter X through the bacterial foraging algorithm, and therefore the optimal aluminum electrolysis production process parameter is obtained. The aluminum electrolysis production process multi-objective optimization method has the beneficial effects that the aluminum electrolysis parameter is optimized based on the bacterial foraging algorithm, so that the aluminum electrolysis production efficiency is effectively improved; the bacterial advancing step length is adjusted in an adaptive-step mode, so that the bacterial foraging algorithm is effectively protected against the locally optimal solution; the flora step length is dynamically adjusted through the Pareto differential entropy, and therefore the optimal parameter of aluminum electrolysis production can be quickly obtained.

The invention provides a test data automatic generation method based on an independent path in order to solve the problem of test case generation in software testing. The method comprises the following steps: firstly, carrying out static analysis on a tested program to obtain a control flow diagram of the tested program, and exporting all independent paths through loop complexity; secondly, encoding specific problems of the tested program to obtain an input range and a path code; then, according to a path analysis and branch distance calculation method, performing instrumentation on the testedprogram to obtain a fitness function; generating uniformly distributed initial populations through a chaotic sequence, and selecting the initial population with a high fitness function value as the initial population of the bacterial foraging algorithm; and finally, carrying out iterative updating by utilizing the improved bacterial foraging algorithm and the fitness function until the test casesof all the independent paths are solved or the maximum number of iterations is reached, and recording and outputting the test cases covering the independent paths. According to the method, on the premise that effective coverage is guaranteed, the test cases can be rapidly obtained with few test cases.

The invention discloses an extended black-start scheme bi-level programming optimization method capable of improving the safety of a recovery process, provides a new idea for the recovery control of an initial stage after blackout, enables the safety factors affecting the recovery process to be introduced to an optimization target so as to improve the safety of an extended black-start scheme in anactual recovery process, gives consideration to the main-slave logic relation between unit recovery and path recovery, and achieves the building of an extended black-start scheme bi-level programmingoptimization model. The upper level optimization takes the maximum comprehensive efficiency of the scheme under the safety condition as the target, and a decision making variable is a started unit. The lower level optimization takes the solving of a recovery path, which contains a key path, is high in recovery safety and gives consideration to the recovery of an important load, as the target, anda decision making variable is a to-be-recovered path. A hybrid algorithm which combines an improved bacterial foraging algorithm with a shortest path method based on a Dijkstra algorithm is employedfor solving the model. The recovery process is higher in safety, the recovery effect is improved, and the method has the practical engineering application value.

The invention discloses a method for optimizing a BP neural network based on an improved bacterial foraging algorithm. The method comprises the steps that: a neural network structure is determined, and related parameters of the neural network structure are set; the positions of bacteria are initialized; a bacterial component is converted into the weight and threshold of the neural network; the bacteria are turned over, so that a fitness value after one-time turning over is obtained, and if the fitness value becomes better, the bacteria are moved by a corresponding step length according to a turning-over direction; bacterial energy is obtained, one half of poor bacterial energy is eliminated, the other half of the poor bacterial energy is bred, and child bacteria have the same position andstep length as mother bacteria; the pheromone concentration of the current bacteria is acquired, the transition probability is calculated, and the positions of the bacteria are updated, so that an updated fitness value is obtained; and a group optimal solution is obtained, and the optimal solution is converted into the weight and threshold of the neural network. The method optimizes the weight andthreshold of the neural network, improves the performance of the neural network, and enables a prediction result to be more accurate.

The invention relates to a control method of an autonomous underwater vehicle, in particular, and a mechanism model-based autonomous underwater vehicle predictive S-plane control method. The inventionaims to solve the problem that the motion control effect of an existing AUV S-plane control method is affected due to difficulty in obtaining optimal control parameters or difficulty in adapting to complicated and varied marine environments. According to an AUV control model, a classical S-plane control method is adopted to perform closed-loop control on an AUV; control quantities are outputted from an S-plane control link in each control beat; and the control parameters k1 and k2 of the internal S-plane control link of a controller are set and adjusted by a prediction structure in each parameter setting beat through a bacterial foraging algorithm. The method of the invention is applicable to the control of autonomous underwater vehicles.

The invention discloses a microbial fermentation optimizing method based on a bacterium foraging algorithm. The microbial fermentation optimizing method comprises the following steps that a microbial fermentation data set is established and randomly divided into a training data set and a testing data set, a BP neural network is structured and trained through the training data set, binary encoding is carried out on each microbial fermentation control parameter to obtain an initial bacterium population, a bacterium chemotaxis operator and a reproduction operator are executed, and a migration operator is executed on each bacterium according to the probability. By means of the microbial fermentation optimizing method based on the bacterium foraging algorithm, the optimal control parameter combination can be obtained according to existing fermentation data, and an experiment does not need to be redesigned.

The invention provides a site selection-distribution method and system based on a bacterial foraging algorithm and an ant colony algorithm. According to the embodiment of the invention, a site selection-distribution model meeting the capacity constraint of a distribution center is formed through a constructed upper-layer target function and a lower-layer target function and the corresponding constraint conditions, the site selection problem of each distribution center is solved through the bacterial foraging algorithm, and to-be-distributed clients are reasonably distributed according to the capacity constraint of the distribution center and the demand of the clients; An optimal distribution scheme of each distribution center is solved through an ant colony algorithm according to the vehicle capacity constraint and the client time window. Conditions such as distribution center capacity, customer demand, time window and the like are considered, the optimal upper-layer objective functionis finally achieved, i.e., the total cost of site selection-distribution is lowest.

The present invention discloses a raster map copyright protection method based on a bacterial foraging algorithm (BFA) and a least significant bit (LSB). The method comprises: carrying out edge detection on the raster map by using the sobel operator, and extracting map edge data; dividing the edge image into a plurality of blocks according to the number of bits of watermark information; using the BFA algorithm to optimize the edge points in each block, and extracting optimal watermark embedding points until the number of extracted points is equal to the number of bits of the watermark information; and according to the position of the extracted points, embedding the watermark information into the corresponding pixel of the original raster map for each block by using the LSB algorithm. According to the method disclosed by the present invention, the generated raster map digital watermark can better resist the noise attack, the filtering attack and the geometric attack, has good concealment, and can effectively protect the raster map copyright.

The invention discloses a power distribution network reconfiguration optimization operation analysis method and device. The method comprises the steps that A, a power distribution network reconfiguration model is established, wherein an objective function and a constraint condition are determined; B, specific parameters of a power distribution network and a bacterial foraging algorithm based on adaptive step size and crossed replication are determined; C, decimal encoding rules based on a looped network are used to generate an initial population; D, transcoding rules are used to convert the initial population of decimal encoding into an initial population using real number encoding; E, the bacterial foraging algorithm based on adaptive step size and crossed replication is used for solving;and F, optimization analysis is performed to obtain a reconfiguration optimization result. Through the power distribution network reconfiguration optimization operation analysis method and device, anoptimization operation problem of power distribution network reconfiguration can be efficiently solved, the algorithm is easy to implement, the convergence rate is high, and optimization precision ishigh.

The invention discloses a node positioning method based on a wireless sensor network DV-Hop ranging algorithm. The method comprises the steps that a beacon node carries out the self-positioning according to GPS equipment carried by the beacon node, obtains the current position information, and builds a data information package for broadcasting according to the initial position information, after receiving the data information packet, a neighbor node modifies internal data, performs secondary transmission, and selectively reserves a hop count value of the beacon node closest to the neighbor node; a global hop distance and a local hop distance are calculated according to hop values in the data information packet, and the global hop distance and the local hop distance are averaged to obtain an average hop distance; and finally, according to a bacterial foraging algorithm, accurate optimization is performed on estimated coordinates by utilizing a three-step optimization method of chemotaxis, duplication and dispersion of the bacterial foraging algorithm. According to the positioning method provided by the invention, the positioning information of unknown nodes can be obtained while thenumber of the beacon nodes is reduced, the use cost and the energy consumption are reduced, and the feasibility and the effectiveness are good.

With the development of science and technology and the advent of the era of big data, the problem of social information security receives much concern. For example, the public security department identifies suspects from the streets, banks need identity authentication of customers, and the customhouse identifies the identity of exit-entry people. The means of identity authentication emerges in endlessly, such as password, fingerprint, ID card, RFID card and other ways of recognition, wherein face recognition is a popular research area at present. In recent years, many excellent talents have conducted research in the field of face recognition, and have made great achievement. In view of the weaknesses and shortcomings of the methods put forward by predecessors, the traditional bacteria foraging method is improved in the invention, and the improved algorithm is applied to face recognition, namely, classification is performed on an original face database by the improved bacteria foraging method first, and then, a target face is recognized by a comparison method. Experiments show that a method of the invention has very high recognition rate. The method of the invention mainly comprises a face input module, a face image preprocessing module, a face feature extraction module, a bacterial foraging algorithm training module, a target face image input module, and a target face image recognition module.

A method for locating a solid state transformer considering distribution network reconfiguration comprises the following steps: A. establishing a bi-level optimization model for locating the solid state transformer considering distribution network reconfiguration, comprising an objective function and a constraint condition; B. determine that specific parameters of an improved bacterial forage algorithm which integrates the learning of a variable speed society in a distribution network; C. that solution is carry out by using an improved bacterial forage algorithm fuse with variable speed sociallearning; D. analyze that result to obtain the optimal installation position of the solid-state voltage transformer and the result of the distribution network reconstruction. The invention realizes the optimal location of the solid-state transformer and improves the operation stability and economy of the distribution network through the location selection method of the solid-state transformer considering the distribution network reconfiguration. At the same time, the improved bacterial foraging algorithm has the advantages of fast convergence speed and high searching precision when calculating the network with more ring networks and more complex topology.