45 results about "Random search algorithm" patented technology

The present invention relates to a method and a system for providing an estimate of the state of a contact. The method includes the steps of sensing the state of the contact; generating signals representative of the state of the contact; and processing the signals using a random search procedure to arrive at an estimate of the state of the contact. The random search procedure may employ the simulated annealing-based algorithm methodology or the genetic-based algorithm methodologies. The system includes sensors for sensing the state of the contact and a pre-programmed computer for generating the desired contact state estimates.

A method of determining a set of parameters for modeling an active semiconductor device in which current flow through a channel or other area is regulated by voltage applied to the device terminals, for example, MOSFETs. The method comprises first providing a plurality of measured values for current as a function of voltage for a plurality of active semiconductor devices of differing geometries. There is then determined an initial population of vectors comprising individual values representing a plurality of desired active semiconductor device model parameters. Fitness is then evaluated for each of the vectors by comparing calculated values for current as a function of voltage from the population to the plurality of measured values for current as a function of voltage of the vectors, converting any current differences to voltage errors and adding any such voltage errors together to arrive at a fitness value for each vector. Vectors of best fitness are selected and at least one genetic operator is applied thereto to create a new population of the vectors. Vectors of best fitness are then selected. The steps of evaluating fitness and selecting vectors of best fitness are optionally repeated for such vectors of best fitness until a desired fitness is achieved to determine the desired active semiconductor device model parameters.

The invention mainly provides a bus stop site selection and layout optimization method based on passenger trip spatial distribution. The bus stop site selection and layout optimization method based on the passenger trip spatial distribution mainly comprises construction of a bus stop site selection optimization model and a solution algorithm thereof. The bus stop site selection optimization model takes minimization of walking distance of all the residents for taking buses as a goal and considers realistic constraint factors such as maximum walking distance between a resident dense point and a bus stop and upper and lower bounds of distance between adjacent stops, the solution algorithm of the bus stop site selection optimization model is a novel hybrid algorithm and combines advantages of a bacterial foraging optimization algorithm and a group random search algorithm, and according to problem characteristics, individual bacterium coding, initial bacterial colony generating, individual bacterium evaluation function generating and bacterial foraging operations are redesigned. The bus stop site selection and layout optimization method based on the passenger trip spatial distribution scientifically and reasonably determines bus stop positions by combining a real road topological structure according to trip spatial distribution characteristics of residents nearby a route, so that resident trips are facilitated, and bus operation efficiency is also improved.

The invention relates to a grid resource allocation method based on a parallel genetic algorithm. The method comprises the following steps: firstly, the information is initialized in a main thread, such as task collection, machine collection, an execution time matrix E of the task, and mapping of a sub-task to the machine, etc.; then a plurality of sub threads are generated and mapped to different processors, an initializing sub-population is independently generated by each sub thread, evolutionary computation is performed in parallel, the optimum individual of each generation is transferred to the main thread, the main thread performs comparison, and the optimum individual is retained; when the predetermined generation arrives, the transfer operation between the sub-groups is performed; and the operation of the main thread and all the sub-groups cannot be finished until the termination conditions are met. The genetic algorithm is taken as the most effective heuristic global stochastic searching method, and the solution of the NP problem can be performed. The quality and the speed for the algorithm for solving are improved by the parallel genetic algorithm proposed according to the natural parallelism of the genetic algorithm, and the method is an effective grid energy resource optimization method and favorable for improving the service quality of the grid.

The invention provides an oil deposit well pattern and injection-production scheme optimum design method based on a balanced water drive idea. The method comprises the steps of 1, collecting and arranging block geological and development related data; 2, setting optimizing related parameters, and completing the well pattern and injection-production optimization preparation work; 3, adopting an oildeposit engineering method, predicting the displacement condition on each direction under current well location and injection-production parameters and performing quantitative evaluation; 4, adoptinga global random search algorithm, optimizing and generating novel well location / injection-production parameters, and predicting and evaluating the displacement condition in each injection-productiondirection under novel well location injection-production parameters; 5, arranging the optimization result, forming a well pattern injection-production design scheme, and performing on-site implementation. According to the method, the water drive development index calculation method built on the basis of the oil deposit engineering idea is combined with the optimization theory, the global random search algorithm is adopted for automatic solving, the calculation efficiency is ensured while the well pattern and injection-production scheme matched with the actual oil deposit is obtained, and the oil deposit recovery efficiency is improved.

The invention discloses a WSAN actuator task distribution method based on BA-BPNN data fusion, and the method employs a BA optimization BP neural network to build a data fusion model. The method specifically comprises the steps: employing a bat algorithm to optimize the weight value and threshold value of the BP neural network, building a data fusion model, carrying out the data fusion of the sensor node information, and obtaining the task distribution information of an actuator node. The bat algorithm is a meta heuristic type group intelligent optimization algorithm, employs an echo positioning method of a miniature bat under the condition of different transmitting speeds and responses, can achieve a precise capturing and obstacle avoidance random search algorithm. The BP neural network is a multilayer feedforward neural network which can search a global optimal value in a training process, and can increase the convergence rate of the network. The method searches the optimal parameter of the BP neural network through the positioning updating of bats, is more precise in data fusion, and is more reasonable in task distribution of an actuator.

The invention discloses a heuristic one-dimensional blanking method based on a stratified random search algorithm. The method includes: A, using a parameterized model to signify one-dimensional blanking problems; B, preprocessing billets for combination; C, obtaining various stock layout modes through the stratified random search algorithm combined by random search and deep search; D, picking out an optimal stock layout mode according to a heuristic rule; E, adding the optimal stock layout mode and a time which does not exceed the current needed maximum time of reusing the billets into a current stock layout scheme, and updating a billet set to be laid; F, repeating processes of C, D and E until the total length of the billets to be laid less than the length of raw materials, and outputting the current stock layout scheme; and G, repeating processes of B, C, D, E and F for many times, and then performing compared screening on all stock layout schemes so as to obtain an optimal stock layout scheme. The heuristic one-dimensional blanking method based on the stratified random search algorithm can avoid blindness of a traditional random search algorithm and is high in computation speed, and the obtained stock layout scheme adapts to practical production needs.

According to some embodiments, a thermostat obtains real-time energy prices from a electricity grid. It may also obtain additional data from external data sources, such as predicted energy prices or weather predictions. The thermostat attempts to find a control strategy for when to switch available aggregates that may include furnaces and air conditioners on and off. In order to solve this integer programming problem, the thermostat uses a random search algorithm. According to some embodiments, various data sources, such as day-ahead prices and real-time prices, are combined into forecasts of electricity prices for the present and future time periods. In some embodiments, a thermostat selects predictively between heating or cooling by letting outside air in or by using heating and cooling aggregates. Additional embodiments are discussed and shown.

The invention belongs to the technical field of robot task allocation, and particularly relates to a tight coupling task allocation method based on an improved ant colony algorithm. Aiming at the problems of the ant colony algorithm in the field of multi-robot task allocation at the present stage, a strategy of searching robots from tasks by ants is adopted, and a reference set is provided for updating pheromones of the ant colony algorithm in combination with a random search algorithm, so that the convergence speed is increased, and the task allocation solving accuracy is improved. Accordingto the method, the random search algorithm is introduced into the traditional ant colony algorithm, so that the early-stage convergence speed of the traditional ant colony algorithm is increased, thesituation that the traditional ant colony algorithm is prone to falling into local optimum is optimized, multi-robot task allocation is effectively achieved, and the defects that the traditional ant colony algorithm is slow in convergence, prone to falling into local optimum and the like are well overcome.

The invention discloses an optimization algorithm of a low-voltage transformer area line loss neural network, and the algorithm comprises the steps: 1) carrying out the preprocessing of the original distribution transformer side data and user side data of a low-voltage transformer area, and obtaining the line loss characteristic indexes of the low-voltage transformer area; 2) performing clustering analysis on the low-voltage transformer area data, dividing the data into four classification samples, and performing modeling on the whole sample and the four classification samples by adopting linear regression, r-tree and K-nearest neighbor algorithms; 3) screening out four characteristic parameters related to the grid structure and the load of the transformer area through a main factor analysis method; and 4) establishing a BP neural network model, setting four input ends and one output end which are respectively corresponding to the four characteristic parameters and the line loss rates, and optimizing input variables of the neural network by applying a genetic algorithm and / or a particle swarm algorithm until data convergence. According to the characteristics of low-voltage transformer area line loss calculation, two random search algorithms, namely the genetic algorithm and the particle swarm algorithm, are applied to optimization of the initial threshold value and the weight value of the BP neural network, and the precision and the speed of network training are improved.

The invention provides a harmony search algorithm for solving mixed loading of various large-batch cold-chain articles. The method comprises the following steps: firstly, acquiring various parametersof to-be-conveyed cold-chain articles and a used single-temperature-zone refrigerator car, establishing a mixed loading model when different cold-chain articles are assembled to the single-temperature-zone refrigerator car, then establishing an initial harmony memory bank, obtaining an initial solution by adopting a random search algorithm, and repairing the solution through a repairing operator until a required number of solution vectors are obtained; secondly, calculating a new solution according to the parameters and harmony search rules and repaired, and finally obtaining a cold-chain article mixed loading scheme meeting the conditions. Under the condition that the temperature constraint, the volume constraint and the weight constraint of the single-temperature-zone refrigerator car are met, a harmony search algorithm is adopted for automatic calculation, and an assembling scheme with the minimum humidity difference, the minimum shelf life difference and the minimum mixed loading type difference as targets is generated.

The invention relates to the field of bearing fault prediction and diagnosis and discloses a bearing fault diagnosis method based on random search and a convolutional neural network, which optimizes hyper-parameters in combination with a random search algorithm and establishes a convolutional neural network model for intelligent diagnosis of rolling bearing faults; problems that a traditional method in the prior art is not enough in accuracy and manual parameter adjustment in an intelligent method is tedious and time-consuming are solved. The method comprises the following steps of (1) initializing a hyper-parameter combination; (2) configuring a distribution function of random search; (3) continuously updating the distribution function of random search; and (4) selecting optimal hyper-parameter configuration; and training to obtain a final bearing intelligent diagnosis network model. The method is advantaged in that through an alternate connection structure of two convolution layers and a single pooling layer, the convolution layer performs convolution operation and data feature learning on input data, the pooling layer is designed to be maximum pooling, and pooling kernel operation of the maximum pooling layer can enhance data features obtained by learning of the convolution layer.

The invention provides an elastic vector wave field numerical simulation method and system based on a random search algorithm, computer equipment and a computer readable storage medium, and relates tothe technical field of seismic exploration. The method comprises the following steps: constructing an improved particle swarm algorithm; According to the improved particle swarm optimization algorithm, carrying out optimization solution on an objective function containing a finite difference coefficient to obtain an optimized finite difference coefficient; And carrying out numerical simulation onthe elastic vector wave field according to a finite difference operator formed by the optimized finite difference coefficient. The invention designs a random search algorithm. the algorithm is an improved particle swarm algorithm; According to the elastic vector wave field numerical simulation method, the objective function containing the finite difference coefficient is optimized and solved through the improved particle swarm algorithm, the optimized finite difference coefficient is obtained, then the finite difference operator composed of the finite difference coefficient is used for conducting elastic vector wave field numerical simulation, and the elastic vector wave field numerical simulation precision and efficiency are improved.

According to some embodiments, a thermostat obtains real-time energy prices from a electricity grid. It may also obtain additional data from external data sources, such as predicted energy prices or weather predictions. The thermostat attempts to find a control strategy for when to switch available aggregates that may include furnaces and air conditioners on and off. In order to solve this integer programming problem, the thermostat uses a random search algorithm. According to some embodiments, various data sources, such as day-ahead prices and real-time prices, are combined into forecasts of electricity prices for the present and future time periods. In some embodiments, a thermostat selects predictively between heating or cooling by letting outside air in or by using heating and cooling aggregates. Additional embodiments are discussed and shown.

The invention discloses a third-party software testing system and method in a cloud computing environment. Third-party software testing is carried out by using a cloud computing method, so that a usercan carry out software testing anytime and anywhere, the technical defect that a software testing environment needs to be maintained regularly in the conventional technology is overcome, and the pressure on testing resource management caused by the testing environment is reduced. As the cloud computing method can be used for testing anytime and anywhere, a user can control the testing time autonomously in the using process, and the technical defects of long testing period, high testing cost and the like in the conventional technology are overcome. The layout of the software test items of thecloud network is realized by adopting the data set of the big data storage data structure, the retrieval of the retrieved software test items is realized, the query time of the test items is reduced,and the type data layout of the software test items is improved. Optimal search of various software test items is realized through a random search algorithm model, so that the software test time is greatly saved.

The invention relates to a path planning method and device passing through a necessary point set and having extra hard constraint, wherein the path planning method comprises the following steps: S1, a preprocessing stage: converting the actual application environment information into mathematical description, and completing the processing and modeling of an undirected weighted graph of a problem; and S2, a shortest path solving stage: carrying out shortest path solving by using a random search algorithm, carrying out real-time judgment on the feasibility of the path in the search process, solving the shortest path problem of the necessary point set under the condition of meeting the requirement of extra hard constraint, and finally obtaining the shortest path meeting the requirement. According to the method, the additional hard constraint can be added, and the calculation result meeting the engineering requirements under the actual physical limitation condition can be ensured to be obtained within the short calculation time.

The invention discloses a large-scale offshore wind power cluster fan arrangement optimization method and system. The method comprises the following steps: S1, dividing a large-scale wind power cluster area into a plurality of same small-scale wind power plant areas; S2, determining a medium-sized wind power plant of which the scale is smaller than that of the large-sized wind power cluster from the large-sized wind power cluster as an optimization object, wherein the shape of the optimization object is the same as that of the large-sized wind power cluster; S3, determining fan arrangement of the small wind power plant area with the minimum wake flow loss from the fan arrangement candidate points by using a random search algorithm; and S4, calculating the wake flow loss of the large-scale wind power cluster area, and when the wake flow loss of the large-scale wind power cluster area does not conform to the expectation, expanding the scale of the optimization object, and skipping to the step S3 until the wake flow loss of the large-scale wind power cluster area conforms to the expectation. The small-scale wind power plant is used for replacing the whole wind power cluster to serve as an optimization object, the optimization time and cost can be greatly shortened, and the calculation speed and wake flow loss are well balanced.

The invention relates to a method for solving the selected harmonic eliminated switching angle of a ladder wave multilevel converter. Power reduction is performed on an equation set by use of an elementary symmetric polynomial, and the equation set is triangularized by use of a Groebner basis algorithm; the solution of a multivariate higher-order polynomial equation set is converted into the solution of two univariate higher-order polynomial equations and a set of single-variable linear equations; all solutions of a harmonic elimination equation set can be obtained based on constraint conditions; and the optimal switching angle can be worked out by evaluating the total harmonic of each set of solutions. Compared with the current commonly-used numerical algorithm and random search algorithm, no initial value needs to be given, and all solutions of the harmonic elimination equation set can be obtained. Compared with other existing algebraic algorithms, the amount of calculation is greatly reduced, more switching angles can be processed, and the method has higher practical values.

Described here are systems and methods for knowledge-based brachytherapy planning. These systems and methods are capable of automatically generating treatment plans for prostate brachytherapy in clinically relevant times. Primitive features computed from anatomical contours of the target organ are used to retrieve a template plan from a database. The template plan is then adjusted in a stochastic search algorithm.

According to the remote sensing and monitoring global target space coverage optimization method disclosed by the invention, aerial targets in a certain height range can be effectively sensed. The method can be realized through the following scheme: firstly, screening out sensors capable of sensing a to-be-sensed target space, forming a three-dimensional space directed sensor network of sensing coverage, and establishing a set coverage model; counting the coverage range of discrete points of the three-dimensional to-be-sensed space, calculating the sensing coverage rate, and iteratively optimizing the fitness function. The directional sensors are adjusted to solve the sensing coverage rate of the whole directional sensor network to the space to be sensed, the pitch angle and the azimuth angle of each sensor are iteratively optimized by using a random search algorithm, and the specified three-dimensional aerial target on the earth surface is sensed and covered until the coverage rate requirement or the iteration frequency requirement is met. Optimizing parameters are output through the intelligent optimization algorithm to adjust the azimuth angle and the pitch angle of the sensor, and global optimal space coverage is obtained by optimizing the main sensing direction.

The invention relates to a query optimization method based on a simulated annealing algorithm. The method comprises the steps that a data query optimization process is divided into a model building part, a strategy space resolving part and an optimization part, then the simulated annealing algorithm is led in, all strategy space subsets are searched in a parallel mode, a final solution is obtained from each subset, and the optimal solution is obtained after the final solutions are compared. Compared with other intelligent optimization algorithms, the simulated annealing algorithm can effectively avoid a local extremum and shorten the optimization time. In addition, due to the utilization of parallel searching, the searching range of the simulated annealing algorithm is enlarged, and the influence on searching precision by local search characteristics of the simulated annealing algorithm can be reduced. Compared with the probability for searching an optimal strategy by a traditional local random searching algorithm, the probability for obtaining the optimal strategy by the query optimization method based on the simulated annealing algorithm is improved obviously. The query speed of a database is improved, the relative time of query optimization is shortened, and the probability for obtaining the optimal strategy is improved.

The invention provides an oil deposit well pattern and injection-production scheme optimum design method based on a balanced water drive idea. The method comprises the steps of 1, collecting and arranging block geological and development related data; 2, setting optimizing related parameters, and completing the well pattern and injection-production optimization preparation work; 3, adopting an oildeposit engineering method, predicting the displacement condition on each direction under current well location and injection-production parameters and performing quantitative evaluation; 4, adoptinga global random search algorithm, optimizing and generating novel well location / injection-production parameters, and predicting and evaluating the displacement condition in each injection-productiondirection under novel well location injection-production parameters; 5, arranging the optimization result, forming a well pattern injection-production design scheme, and performing on-site implementation. According to the method, the water drive development index calculation method built on the basis of the oil deposit engineering idea is combined with the optimization theory, the global random search algorithm is adopted for automatic solving, the calculation efficiency is ensured while the well pattern and injection-production scheme matched with the actual oil deposit is obtained, and the oil deposit recovery efficiency is improved.

The invention provides a public transportation multi-source data fusion based IC card swiping station matching method. The method is characterized in that data resources of a card swiping system and a GPS vehicle movement recording system are fused; and the optimal time lapse amount is obtained based on an adaptive random search algorithm to eliminate the time difference between the card swiping system and the GPS vehicle movement recording system so as to realize IC card swiping station matching. According to the IC card swiping station matching method, equipment regulation and changing is not required, while equipment adding is not required as well; therefore, low computation burden and high computation speed are realized; and the mutual matching between the IC card numbers and card swiping stations can be effectively realized, so that guest flow statistics and data mining can be facilitated.

The invention provides a method for improving transportability of an image recognition model. An improved minimum weight random search algorithm is used; endowing each attribute with a searched weightvalue; wherein the more the searched times are; weight increase, conversely, the smaller is, calculating the next searched probability of each attribute according to the weight; wherein the smaller the weight value is, the higher the searched probability is, otherwise, the lower the searched probability is, and further, according to the searched probability, the searching direction can be deviated to the attribute with the smaller weight value, namely, the attribute with the smaller searching frequency and the object with the larger weight are properly ignored, so that the purpose of searching balance is achieved; through an E-S judgment method, complex operation of further calculation accuracy is reduced, Meanwhile, the purpose of screening objects is achieved; by increasing the complexity of each attribute combination and applying a convolutional neural network which is constructed based on a Leaky Relu activation function and is provided with three convolutional layers, the purposeof fully extracting image features is achieved.

The invention discloses a sensor deployment method and system based on BIM and simulation. The method is based on the information of BIM. In the simulation environment, the sensor data is represented by the indoor environment variable in the divided hot zone, and the environment variable of the hot zone is used. The influence component on the HVAC terminal is used as a decision variable, with the minimum HVAC misadjustment rate and the minimum number of configured sensors as the optimization goals, and a random search algorithm is used to obtain the optimal deployment scheme of indoor sensors. The invention enables the deployment of indoor sensors to be completed at the architectural design stage, which significantly saves the cost of data collection and scheme testing; at the same time, it adopts a clear and adaptable optimization target, making the optimization result more practical.

The invention discloses a method for determining a variable cross-section pipe distribution mapping relation, which comprises the following steps of: S1, adopting an artificial potential function method to quantify the mapping relation, and defining a potential energy state function when a pipeline or a heat exchange pipe passes through a space; the mapping relation is quantified by adopting an artificial potential function, the target of random search is to enable a potential energy state function to take a minimum value, and at the moment, the corresponding mapping relation is a potential selectable distribution mapping scheme. The method is mainly used for determining a pipe distribution scheme of dense pipelines or heat exchange pipes in special-shaped equipment, and compared with a traditional mode of manually setting the mapping relations of the pipelines or the heat exchange pipes one by one, the pipe distribution efficiency can be effectively improved by adopting an artificial potential function method and a genetic algorithm or other random search algorithms, and the pipe distribution efficiency is improved. Through random search in the mapping space, an optimal mapping relation scheme can be obtained.

The invention discloses a coal-fired boiler exhaust gas temperature prediction method and system based on a LightGBM and a random search method, and solves the problems that an existing neural network model is liable to fall into a local minimum value and is liable to over-fit, and a support vector machine model is not suitable for large sample learning. The method comprises the steps of collecting historical operation data, performing data cleaning and normalization, performing feature selection according to mutual information entropy, constructing a model by adopting a LightGBM algorithm, optimizing hyper-parameters by adopting a random search algorithm, and obtaining an optimal model for verification application. According to the method, the LightGBM and the random search algorithm are adopted to establish and optimize the prediction model, the overfitting phenomenon is effectively prevented, the model generalization ability is excellent, a large sample learning strategy is supported, training is more efficient, the calculation speed is higher, lower model deviation can be achieved, meanwhile, the random search algorithm is combined, an optimal hyper-parameter combination is found, and the prediction accuracy is improved. The precision of the model is further improved, and a high-performance coal-fired boiler exhaust gas temperature prediction model is obtained.

The invention discloses a method and system for optimizing the arrangement of large-scale offshore wind power clusters. The method includes: S1: dividing a large-scale wind power cluster area into a plurality of identical small wind farm areas; S2: determining from the large-scale wind power cluster The medium-sized wind farm of the wind power cluster is used as the optimization object, and the shape of the optimization object is the same as that of the large-scale wind power cluster; S3: Use the random search algorithm to determine the wind turbine arrangement in the small wind farm area with the smallest wake loss from the candidate points of the wind turbine arrangement cloth; S4: Calculate the wake loss in the large-scale wind power cluster area. When the wake loss in the large-scale wind power cluster area does not meet expectations, expand the scale of the optimization object, and skip to step S3 until the wake loss in the large-scale wind power cluster area meets the expectations. expected. By replacing the entire wind power cluster with a smaller-scale wind farm as the optimization object, the invention can greatly shorten the optimization time and cost, and better balance the calculation speed and the wake loss.

Described here are systems and methods for knowledge-based brachytherapy planning. These systems and methods are capable of automatically generating treatment plans for prostate brachytherapy in clinically relevant times. Primitive features computed from anatomical contours of the target organ are used to retrieve a template plan from a database. The template plan is then adjusted in a stochastic search algorithm.