68 results about "Binary particle swarm optimization" patented technology

The invention relates to the technical field of computer application engineering, in particular to a method for using an improved neural network model based on particle swarm optimization for data prediction. The method includes the steps of firstly, expressing data samples; secondly, pre-processing data; thirdly, initiating the parameters of an RBF neural network; fourthly, using the binary particle swarm optimization to determine the number of neurons of a hidden layer and the center of the radial basis function of the hidden layer; fifthly, initiating the parameters of the local particle swarm optimization. By the method for using the improved neural network model based on particle swarm optimization for data prediction, the number of the neurons of the hidden layer of the RBF neural network model can be determined easily, RBF neural network performance is improved, and data prediction accuracy is increased. In addition, the improved neural network model based on particle swarm optimization is low in model complexity, high in robustness and good in expandability.

The invention discloses a mobile edge computing task unloading method in a single-user scene, and relates to the field of treatment of a mobile computing system. The invention aims to reduce the reaction time delay and energy consumption of mobile equipment. construction of single-user scene task unloading model comprises construction of a system overall model and construction of each partial model, wherein the construction of each partial model comprises: a task queue model, a local computing model, a cloud computing model and a computing task loading model; a task unloading strategy: givinga task unloading scheme targeting at minimizing the system overall load K; executing all tasks on local CPU or an MEC server based on binary particle swarm optimization; correspondingly and locally executing a load optimal scheduling strategy, and executing a load optimal scheduling strategy by the MEC server based on pipeline scheduling. Through verification, the task unloading method in a sing-user scene provided by the invention reduces the reaction time delay and energy consumption of the mobile equipment.

The invention provides a method of reconstructing a power distribution network containing a distributed power supply. A network topology adjusting module considering various load modes is established according to actual operation conditions of the power distribution network and has good practical value. Meanwhile, topological adjusting is performed by means of a binary particle swarm optimization algorithm according to features of network topology adjustment; a dynamic inertial weight adjusting mode is provided; better coordination with global convergence speed and local convergence precision is provided; optimizing effect is better; the topological adjustment results and analysis of an IEEE-33 node typical distribution system show that the method is effective and practical.

The invention relates to an electric vehicle charging station planning method considering traffic network flow. The method comprises the following steps: referencing an intercepting location model in a traffic field, taking largest captured traffic flow, minimum network loss of a power distribution system and minimum node voltage deviation as an objective, establishing a multi-objective decision model, then determining a reasonable weight coefficient of each objective function after being normalized by using a super efficiency data envelopment analysis evaluation method, and transforming a multi-objective optimization problem to a single objective optimization problem, later putting forward an improved binary particle swarm optimization algorithm to solve a single objective optimization model. The method considers the urban traffic network traffic, electric car driving mileage and location problem of an electric vehicle charging station which affects a distribution network. The example results of the method show that the planned charging station location can provide convenient and fast charging services for more electric cars, and further improve the safety and reliability of the system operation.

The invention provides a voiceprint recognition method based on pitch period mixed characteristic parameters. The method comprises the following steps of voice signal acquisition and input, voice signal preprocessing and voice signal combined characteristic parameter extraction, i.e. a pitch period, LPCC, delta LPCC, energy, first order difference of energy and GFCC characteristic parameters are extracted to be combined into multidimensional characteristic vectors together, the multidimensional characteristic vectors are screened by adopting a discrete binary particle swarm optimization algorithm, the voice model of a speaker is obtained by introducing universal background model UBM training, and finally test voice is recognized by utilizing a GMM-UBM model. Compared with a mode that voiceprint recognition is performed through single voice signal characteristic parameter, recognition accuracy of the voiceprint recognition and system stability are effectively enhanced by adopting the combined characteristic parameters and using the voiceprint recognition system of the GMM-UBM model.

The invention relates to a gene regulation and control network constructing method based on a Bayesian network. Relationship of all gene expressions in a species or a tissue is wholly analyzed and studied in simulation by constructing a gene regulation and control network model. The construction of the gene regulation and control network model comprises the following steps: A. identifying the best node order by a binary particle swarm optimization algorithm with memories, wherein after the particle speed is updated by the binary particle swarm optimization algorithm, the speed of part particles is varied and a searching space is searched, and skipping from local is taken as optimal; and B. inputting the obtained best node order as a K2 algorithm, then executing K2 algorithm and studying the structure of the Bayesian network. The method has the advantages of fast convergence, simple calculation, easy realization and the like, provides clues for studying regulation and control of gene transcription level, shows the structure of the gene regulation and control network more accurately and has important theoretical significance and practical value in many fields, such as biology, medicine science, pharmacy and the like.

The invention relates to a power distribution network multi-period dynamic fault recovery method considering a DG (Distributed Generation) output curve. A power distribution network fault recovery model generally carries out static recovery on conditions at one moment, and an output of a DG is changed along with time, and thus, consideration to fault recovery of a power distribution network containing the DG on one time section is impractical. According to the power distribution network multi-period dynamic fault recovery method disclosed by the invention, a fault recovery period is subjected to period division in accordance with the output situation of the DG according to the actual situation on site, an optimal solution is obtained in one single period, and finally, an optimal fault recovery scheme in all periods is finally obtained, so that power distribution network multi-period dynamic fault recovery considering the DG output curve is implemented. The power distribution network multi-period dynamic fault recovery method disclosed by the invention sufficiently considers the output change of the DG, proposes combination of a load shedding strategy based on an optimal fault recovery path on the basis of improving a binary particle swarm optimization algorithm, and solves the problem of power distribution network multi-period dynamic fault recovery considering the DG output curve.

The invention discloses a method of determining locating and sizing planning of distributed power supplies in a power distribution network. The method includes the steps: for the power distribution network corresponding to a planning area, grouping distributed power supplies in the control partition into a cluster according to each control partition that is pre-divided, so as to obtain at least two clusters; acquiring the basic data of the power distribution network; according to the basic data of the power distribution network, establishing an upper layer model including each cluster and a lower layer model which performs parallel calculation on all the clusters for internal nodes in each cluster; optimizing the upper layer model by means of a particle swarm optimization algorithm, and optimizing the lower layer model by means of the particle swarm optimization algorithm and a binary particle swarm optimization algorithm; and taking the optimized result as the target planning result of locating and sizing planning of the distributed power supplies of the distribution network. The method of determining locating and sizing planning of distributed power supplies in a power distribution network can reduce the complexity of locating and sizing planning of the power distribution network.

The invention relates to a fan optimization arrangement method based on binary particle swarm optimization (BPSO), and aims to solve the problem of fan optimization arrangement of an irregularly-shaped wind field. The method comprises the following steps of: acquiring weather and geographical conditions of a wind field to be optimized and fan parameters to be selected; determining an optimized target and optimized constraint; dividing grid of a rectangular region; compiling an effective position array matrix corresponding to the shapes of the grid and the wind field, wherein in the effective position array matrix, each element is set as 1 or 0, 1 means a position at which a fan can be arranged, 0 means a position at which the fan cannot be arranged, the position at which the fan cannot bearranged is opposite to a place where the geographical condition is not suitable for arrangement of the fan in the wind field region and places outside the wind field region; constructing a solution space according to design variable to be optimized, looking for an optimal fan arrangement scheme in the solution space by adopting the BPSO according to a target function, optimizing and thus obtaining the optimal fan arrangement scheme.

The invention proposes a uncertainty based reconstruction method for an active power distribution grid. The method comprises the following steps of: building a multi-target reconstruction mode with minimum grid loss and cost; processing a load parameter of the mode by using an interval analytic hierarchy process; and acquiring the optimal solution of the multi-target reconstruction mode by combining a binary particle swarm optimization with a harmony search algorithm containing a distribution grid (DG), wherein the optimal solution comprises a switch state, and the position and the output power of the DG.

The invention discloses a family load optimized dispatching method based on demand response. The family load optimized dispatching method is characterized by comprising the following steps: 1, dividing family loads into transferable loads and non-transferable loads according to load characteristics; 2, establishing a target function of a family load optimized dispatching model considering a power consumption cost, an incentive income and inconvenience comprehensively according to the transferable loads; 3, determining constraint conditions of the family load optimized dispatching model, and forming a family load optimized dispatching model based on demand response together with the target function; and 4, solving the family load optimized dispatching model through a binary particle swarm optimization, and obtaining an optical dispatching result of the transferable loads. By adoption of the family load optimized dispatching method disclosed by the invention, the optimized dispatching of the transferable loads in the family can be realized, the transferable loads within a peak electricity price time period are transferred to a valley electricity price time period and a common electricity price time period, thereby reducing the family power consumption cost, realizing a peak shaving and valley filling function of the power grid, and thus the operation security and stability of the power grid in the time division electricity price environment.

The invention discloses an improved binary particle swarm optimization algorithm-based power distribution network reconfiguration method. According to the method, load equalization is adopted as an objective; a power distribution network reconfiguration problem is expressed as a nonlinear optimization problem with minimizing a load equalization index as an objective function; the topological structure model of a power distribution network is simplified according to the structure features of the open-loop operation of the power distribution network; a binary particle swarm optimization algorithm is improved; and therefore, the radial structure of the power distribution network can be ensured, at the same time, the number of iterations is greatly reduced. With the method of the invention adopted, a power distribution network reconfiguration problem in load equalization can be effectively solved. The method has fast calculation speed and excellent convergence.

The invention discloses a method for selecting high-spectrum remote-sensing image wave band based on hybrid binary particle swarm optimization differential evolution (HBPSODE). The method comprises the steps of preprocessing an original high-spectrum remote-sensing image; initializing double swarm individual and algorithm parameters; iterating the double swarms in parallel by the HBPSODE; transferring the optimal solution information through the swarm; calculating the classification precision through an SVM classifier to be used as an adaptability value; updating evolution until reaching the specified evolution times or reaching the maximum precision.

The invention discloses an intelligent design method of a digital coding metamaterial unit. The method comprises the steps of 1) carrying out digital coding on a basic pattern of a 1-bit unit for constructing a metamaterial, and predicting unit polarization wave reflection phases during arrangement of different modules in the 1-bit unit by adopting a deep learning design algorithm; and 2) determining a polarization wave phase difference theta of a 1-bit unit structure for constructing the metamaterial according to a function requirement of the metamaterial, and designing the 1-bit unit structure with the polarization wave phase difference theta in combination with a binary particle swarm optimization algorithm module and a deep learning module. The design method provided by the invention realizes automatic design of an ideal reflection phase of a multi-bit unit on the basis of deep learning, has high efficiency and simplicity, is good in expansibility, can replace software to perform simulation, and lowers and shortens the corresponding complexity and time of obtaining information of a coding unit; and a multi-beam multi-polarization artificial electromagnetic surface is quickly and simply designed.

The invention discloses a PQ (Power Quality) monitoring point configuration method metering a DG (Distributed Generator). The PQ monitoring point configuration method comprises the following steps of defining related concepts in PQ monitoring point optimal configuration metering DG grid connection; defining NKCL of the minimum number meeting KCL (Kirchhoff Current Law); defining an observable voltage area; improving a BPSO (Binary Particle Swarm Optimization) model; constructing a new evaluation function; initializing the positions and the speeds of particle swarms, substituting the particle swarms into an evaluation functional expression to compute an adaptive value, and assigning an initial extreme value; updating all particles according to an iteration expression of the positions and the speeds of the particles; then substituting all the particle swarms into the evaluation functional expression to compute an adaptive value; breaking out of a loop when maximum iterations are achieved, and outputting a current global extreme value as an optimization result; otherwise, returning to step 7 to carry out continuous iteration.

The invention discloses a tumor key gene identification method based on prior information and a parallel binary particle swarm optimization. The tumor key gene identification method comprises the steps of performing preprocessing on tumor gene expression profile data, on a training set, determining an optimal gene cluster number K through a user-defined criterion function by an improved Elbow method; preferably selecting K optimal cluster centers by the particle swarm optimization (PSO), and clustering the tumor genes into K classes on the training set by a K-mean value method; on the training set, obtaining gene to class sensitivity (GCS) information and gene regulation (GR) information separately; and taking the obtained K gene clusters as a searching spacing, by combining with the obtained two kinds of prior information, identifying the key tumor genes by adopting the parallel binary particle swarm optimization (BPSO). Compared with the existing tumor key gene identification method, the probability of losing key information genes related to classes of tumors is lowered by consideration of the two kinds of prior information in the tumor key gene identification method disclosed by the invention, so that subsequent tumor identification can be improved.

The invention provides a danger assessment method of power grid blackout based on generalized extreme value theory and analytic hierarchy process, which relates to the technical field of electric power system risk assessment. The method comprises the steps of S1 extracting the sample data of load loss; S2 selecting an appropriate extreme value distribution model; S3 conducting parameter estimationof the generalized extreme value distribution model by means of genetic binary particle swarm optimization based on effective population; S4 analyzing the accident probability of load loss; S5 establishing evaluation index and calculating weight vector by means of analytic hierarchy process; S6 calculating the danger index of load loss. The method is advantageous in that the same type of accidents which happen in different areas can undergo parallel comparison; a clear accident result seriousness grade evaluation index can be established.

The invention discloses a distribution network reconfiguration method based on an immune binary particle swarm algorithm. Includes such steps as simplifying distribution network, numbering nodes, branches and switches, analyzing distribution network topology by using depth-first search method, calculating distribution network power flow by using forward-backward substitution method to obtain linepower flow and loss; 2) aim at minimizing that network los in normal operation of the distribution system, establishing a distribution network reconfiguration model and determining a fitness function;3) that immune binary particle swarm algorithm is implemented, the state of the connection switch in the distribution network is taken as a control variable to code, the initialization of the population is completed, the affinity, the concentration and the selection probability of the particles are calculate, and the global optimal solution is outputted. The invention improves the speed of the distribution network reconfiguration, prevents the problem of premature convergence, and improves the searching ability of the algorithm in the whole solution space.

The invention discloses a method for designing a multi-beam multi-polarization artificial electromagnetic surface based on deep learning, and the method comprises the following steps: 1) predicting apolarized wave reflection phase of a 1 bit unit through a deep learning design method; 2) combining with a binary particle swarm optimization algorithm module and a deep learning module to design a 1bit unit structure with polarized wave phase difference theta; 3) according to radiation beam design requirements of the artificial electromagnetic surface, selecting the 1 bit unit with correspondingpolarized wave phase difference to perform array coding, and obtaining the multi-beam multi-polarization artificial electromagnetic surface satisfying design requirements. The design method providedby the invention realizes automatic design of ideal reflection phase of a multi-bit unit based on deep learning, is highly efficient and convenient, has excellent expansibility, can substitute software simulation, reduces corresponding complexity and time of obtaining coding unit information, and quickly and easily designs the multi-beam multi-polarization artificial electromagnetic surface.

The invention discloses an evaluation method for determining a backbone grid line of a power grid. The method comprises the following steps of obtaining configuration information of the power grid and determining a bus set and a branch set for forming the power grid; and searching various key lines included in the power grid by adopting a binary particle swarm optimization (BPSO) rule on the basis of various preset operation modes of the power grid, various pre-arranged initial fault lines of the power grid, the bus set of the power gird and the branch set of the power grid to obtain a branch set occupied by various key lines, wherein the branch set occupied by various key lines forms the backbone grid line of the power grid. According to the evaluation method for determining the backbone grid line of the power grid, the key lines included in the power grid are searched by adopting the binary particle swarm optimization (BPSO) rule through the operation modes and the initial fault lines of the power grid, thereby determining the backbone grid line of the power grid.

The invention discloses a non-invasive load monitoring method, device and equipment as well as a storage medium. The method comprises the following steps: selecting a plurality of different load conditions, and acquiring characteristic values of the load conditions, wherein the characteristic values comprise steady state fundamental wave active power in the load conditions and amplitude of an oddharmonic of a steady state harmonic current; establishing a load characteristic database; and selecting a globally optimal extreme value and a globally optimal position of the characteristic values byutilizing a binary particle swarm optimization algorithm, and constructing a load condition recognition algorithm. The method disclosed by the invention performs active power estimation by utilizingthe load characteristic database and the load condition recognition algorithm, outputs a load structure corresponding to the current load condition, utilizes the steady state fundamental wave active power and the amplitude of the odd harmonic of the steady state harmonic current as the characteristic values, can rapidly recognize different load types from the total current and reduces calculated amount of a system during load recognition.

The invention discloses a video multicast transmission method based on a TV white frequency band. The video multicast transmission method comprises the five steps: building a wireless video multicasttransmission network, performing SVC encoding processing on the video by a CR base station, performing spectrum detection and realizing the optimal resource scheduling. The utilization rate of the TVwhite frequency band is improved by using the cognitive radio technology. The SVC technology is applied to encode the video so that the user can obtain the video matching the channel quality. The AMCchannel encoding technology and the video encoding SVC technology are matched in the physical layer to ensure the link quality. The heuristic algorithm of the binary particle swarm optimization algorithm with low complexity is applied to obtain the sub-optimal solution of the optimal resource scheduling and solve the resource scheduling problem based on the maximum proportional fairness as the resource scheduling criterion. The method improves the throughput of the system and the quality of the received video and realizes the purpose of receiving the matching high-quality video according to the user channel quality.

The invention discloses a 5G network edge server deployment method for mobile users. The 5G network edge server deployment method comprises the following main steps: S1, dividing mobile user data in one day into a plurality of network snapshots in different time periods; s2, taking user delay and edge server deployment cost as optimization objectives, and obtaining edge server deployment schemes of different network snapshots through an improved discrete binary particle swarm optimization algorithm and a recent association algorithm; s3, obtaining a set C containing all network snapshot edge server positions; s4, calculating the position numbers of the edge servers of different network snapshots, and obtaining the maximum value K of the position numbers of the edge servers in all the network snapshots; and S5, selecting an edge server deployment position meeting the movement requirement of the user by adopting an alternate replacement mode to obtain an edge server deployment scheme. According to the method, the user delay is reduced, the mobility of the user is effectively considered, and the overall performance of the system is improved; the method is simple, and an edge server deployment scheme can be obtained more quickly.

The invention discloses a wireless sensor network node scheduling method based on improved binary particle swarm optimization, and the method comprises: firstly carrying out coding of positions and speeds of particles in an improved binary particle swarm optimization algorithm population on the premise that the coverage rate and connectivity constraints of a wireless sensor network are met; evaluating a fitness value of each particle, setting a current position of each initial particle as an individual optimal position, and setting the position of the particle with the maximum fitness value inthe particle swarm as a global optimal position; updating the speed and the binary position of each initial particle; executing mutation operation; judging whether the coverage rate and connectivityconstraints of newly generated particles are met or not; and when the number of iterations reaches a preset maximum value, stopping, and taking the finally obtained global optimal position as a node scheduling scheme. The node scheduling method can meet the coverage rate and connectivity constraints, and meanwhile, the network energy consumption is optimized and the network survival time is prolonged.

The invention provides a dynamic extension and placement method and device for edge computing services, and the method comprises the steps: carrying out the dynamic extension and placement of the edge computing services according to the current respective work load intensity prediction result of each micro-service corresponding to each application in an edge computing platform and the current respective work performance evaluation result of each edge node; performing automatic expansion processing on the number of the micro-service copies according to the target data processing request so as to determine the scaling optimization number of the micro-service copies; and adopting a preset self-adaptive discrete binary particle swarm optimization algorithm to obtain a mapping relationship between each micro-service copy and each available edge node according to the scaling optimization number of the micro-service copies, the number of currently placeable edge nodes and the performance information, so as to respectively place each micro-service copy to the corresponding edge node. According to the invention, the automatic expansion reliability and effectiveness of the edge computing service can be improved under the conditions of unbalanced edge load and unreliable network state in the edge environment, and the accuracy and reliability of edge computing service placement are improved.

The invention discloses a multi-target positioning method and device based on compressed sensing and a binary particle swarm. The method comprises the following steps: carrying out mesh generation ona monitoring region, building a positioning model based on compressed sensing, and converting a multi-target positioning problem into a classic 0-1 knapsack problem; searching a drop point area of a to-be-positioned target on a divided grid by adopting a preset algorithm so as to reduce the dimensionality of the optimization problem; solving a corresponding optimization problem by adopting a binary particle swarm optimization algorithm so as to reconstruct a sparse signal x; and determining a grid where the to-be-positioned target is located according to the reconstructed sparse signal x, andtaking a representative position of the grid as an estimated position of the to-be-positioned target. According to the method, the multi-target positioning problem under a wireless sensor network is converted into the 0-1 knapsack problem based on the compressed sensing theory, the estimated position of the target is obtained according to the reconstructed signal, the positioning precision is improved, and the method and the device can be widely applied to a multi-target positioning technology of a wireless sensor network.

The invention discloses a method for controlling an internal pressure of a water / vapor receiver by optimal scheduling of a mirror field. According to the method, energy received by the receiver is changed and the internal pressure of the receiver is kept constant by carrying out optimal scheduling on the mirror field of a tower solar energy heat power station, carrying out partitioning on the whole mirror field of the tower solar energy heat power station, establishing the scheduling optimization problem of the mirror field, solving the scheduling optimization problem and changing a heliostat state of the mirror field according to an optimization result. 0-1 integer programming in the scheduling optimization problem is solved by adopting binary particle swarm optimization. By utilizing the mirror field optimal scheduling method disclosed by the invention, the heliostat state of the mirror field can be rapidly obtained and pressure stability of a receiver system can be well controlled.

The invention aims to provide a discrete binary particle swarm optimization algorithm and fuzzy control coupled hybrid operation optimization control method. By adopting the method, the traditional single optimization algorithm is improved in solving the optimal problem with constraint conditions, the discrete binary particle swarm optimization algorithm and fuzzy control are subjected to couplingoperation, and a coupling relation exists between input and output of the two algorithms, so that the two algorithms in each step length are compatible. The algorithm inherits the characteristic of high robustness of the fuzzy algorithm, and also inherits the advantages of the traditional particle swarm algorithm in the aspect of the optimization effect.

The invention relates to the technical field of computer application engineering, in particular to a method for using an improved neural network model based on particle swarm optimization for data prediction. The method includes the steps of firstly, expressing data samples; secondly, pre-processing data; thirdly, initiating the parameters of an RBF neural network; fourthly, using the binary particle swarm optimization to determine the number of neurons of a hidden layer and the center of the radial basis function of the hidden layer; fifthly, initiating the parameters of the local particle swarm optimization. By the method for using the improved neural network model based on particle swarm optimization for data prediction, the number of the neurons of the hidden layer of the RBF neural network model can be determined easily, RBF neural network performance is improved, and data prediction accuracy is increased. In addition, the improved neural network model based on particle swarm optimization is low in model complexity, high in robustness and good in expandability.

The invention discloses a method for optimizing a minimum broken point set based on characteristics of spatial distribution. The method for optimizing minimum broken point set based on the characteristics of the spatial distribution comprises the first step of dividing the whole network into k communicated subnets at the positions of a node-cutting point and an edge-cutting set, the second step of dividing four gathered subspaces of the i connected subnets into a first subpopulation, a second subpopulation, a third subpopulation and a fourth subpopulation, the third subpopulation and the first subpopulation are mutually symmetrical, the fourth subpopulation and the second subpopulation are mutually symmetrical, the third step of working out minimum broken point sets of the first subpopulation and the second subpopulation by means of binary particle swarm optimization, the fourth step of obtaining minimum broken point sets of the third subpopulation and the fourth subpopulation according to the spatial symmetrical characteristics of the minimum broken point sets, the fifth step of repeating the step 2 to the step 4 and getting the minimum broken point sets of the k connected subnets, the sixth step of conducting union set operation on the minimum broken point sets of the k connected subnets and obtaining multiple sets of the minimum broken point sets of the whole net. The method for optimizing the minimum broken point set based on the characteristics of the spatial distribution solves the problem of 'curse of dimensionality', and rapidly obtains multiple sets of the MBPS which are quite different in positions by means of the spatial symmetrical characteristics.