2622 results about "Fitness function" patented technology

A method and computer program product are disclosed for determining an optimal classifier model for a pattern recognition system and updating the determined model to recognize new output classes. An initial plurality of classifier models are selected from a set of generated classifier models. An optimal representative classifier for each classifier model is selected according to an iterative optimization routine. The representative classifier having the highest associated value for a fitness function is accepted.

Optimal summaries of a linear media source are automatically produced by parameterizing a linear media source. The parameterized linear media source is used to create a similarity array in which each array element includes the value of a similarity measurement between a two portions of the parameterized media signal. A segment fitness function, adapted for measuring the similarity between a segment of the parameterized media signal and the entire parameterized media signal, is optimized to find an optimal segment location. The portion of the linear media source corresponding to the optimal segment location is selected as the optimal summary. This method produces optimal summaries of any type of linear media, such as video, audio, or text information.

The invention discloses a method and device for optimizing multi-constraint quality of service (QoS) routing selection. The method comprises the following steps: acquiring the topological structure and link parameters of an existing network in accordance with information of a prediction model; creating a corresponding multi-constraint QoS routing model in accordance with the determined topological structure and link parameters, and constructing penalty functions to transform multi-constraint conditions, as well as constructing fitness functions for evaluating paths; using a depth-first search method to acquire initial feasible paths and initializing particle swarms; calculating the fitness value of each particle, and finding out the optimal fitness value of the particle adjacent to each particle; using the generation algorithm and the genetic algorithm-particle swarm optimization (GA-PSO) to carry out iterative solution at the beginning of the initial feasible paths, and carrying out natural selection and variation operations; and finding out paths which meet conditions and are provided with the optimal fitness values, realizing optimal routing selection under the multi-constraint condition, and executing in accordance with the found routings.

A system and method for generating a neural network ensemble. Conventional algorithms are used to train a number of neural networks having error diversity, for example by having a different number of hidden nodes in each network. A genetic algorithm having a multi-objective fitness function is used to select one or more ensembles. The fitness function includes a negative error correlation objective to insure diversity among the ensemble members. A genetic algorithm may be used to select weighting factors for the multi-objective function. In one application, a trained model may be used to produce synthetic open hole logs in response to inputs of cased hole log data.

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

The invention discloses a MC/DC (Modified Condition Decision Coverage) test data automatic generation method based on a genetic algorithm, comprising the following steps of: statically analyzing a tested program to generate a control flow graph, a data flow graph, an abstract syntax tree and an abstract analysis tree; generating a MC/DC test case expected result set; executing code instrumentation on the tested program; constructing a fitness function; randomly generating the test data, and checking whether the test data satisfies an expected execution path; and obtaining proper test data through genetic operations, such as selection, crossing, mutation and the like of the genetic algorithm. In the method, for construction of the fitness function, optimization of approximate-level fitness evaluation by a method of obtaining a control node directly or indirectly influencing defective node traversing through data dependency is proposed according to the thought of a chaining method and based on the conventional fitness function. The method has greater practical value for testing a system with complex logical relations.

A Path Generator connects to a communication network and uses genetic algorithms to assign flows to paths. Genotypes encode flow to path assignments for working and protection paths. Genotype fitness functions are computed as a weighted sum of constraint fitness functions. Each constraint fitness function evaluates the degrees to which the genotype is a satisfactory solution. The system can be used for network modeling. It can also receive requests for on-demand assignment of flows and on-demand rerouting of flows.

A Soft Computing (SC) optimizer for designing a Knowledge Base (KB) to be used in a control system for controlling a motorcycle is described. In one embodiment, a simulation model of the motorcycle and rider control is used. In one embodiment, the simulation model includes a feedforward rider model. The SC optimizer includes a fuzzy inference engine based on a Fuzzy Neural Network (FNN). The SC Optimizer provides Fuzzy Inference System (FIS) structure selection, FIS structure optimization method selection, and teaching signal selection and generation. The user selects a fuzzy model, including one or more of: the number of input and/or output variables; the type of fuzzy inference; and the preliminary type of membership functions. A Genetic Algorithm (GA) is used to optimize linguistic variable parameters and the input-output training patterns. A GA is also used to optimize the rule base, using the fuzzy model, optimal linguistic variable parameters, and a teaching signal. The GA produces a near-optimal FNN. The near-optimal FNN can be improved using classical derivative-based optimization procedures. The FIS structure found by the GA is optimized with a fitness function based on a response of the actual plant model of the controlled plant. The SC optimizer produces a robust KB that is typically smaller that the KB produced by prior art methods.

The invention discloses a network text segmenting method based on the genetic algorithm, used for segmenting short network texts. The method comprises the following steps of: evaluating a Latent Dirichlet allocation (LDA) model corresponding to a corpus by using a Gibbs sampling method, inferring latent topic information using the model, representing texts by using the latent topic information; then transforming a text-segmenting process into a multi-target optimum process by using a parallel genetic algorithm, and calculating the coherency of segmented units, the divergence among the segmented units and fitness functions by using deeper semantic information; and carrying out the genetic iteration of the text segmenting process, and determining whether the segmenting process terminates based on the similarity among multi-iteration results or the upper limit of iterations to obtain the global optimal solution for segmenting the texts. Therefore, the invention improves the accuracy for segmenting the short network texts.

A method for predicting or detecting an event in turbomachinery includes the steps of obtaining operational data from at least one machine and at least one peer machine. The operational data comprises a plurality of performance metrics. A genetic algorithm (GA) analyzes the operational data, and generates a plurality of clauses, which are used to characterize the operational data. The clauses are evaluated as being either “true” or “false”. A fitness function identifies a fitness value for each of the clauses. A perturbation is applied to selected clauses to create additional clauses, which are then added to the clauses group. The steps of applying a fitness function, selecting a plurality of clauses, and applying a perturbation can be repeated until a predetermined fitness value is reached. The selected clauses are then applied to the operational data from the machine to detect or predict a past, present or future event.

The invention relates to a cloud computing resource scheduling realization method based on an improved genetic algorithm. With the development of cloud computing, the scale of a data center is large, the energy consumption problem and the resource utilization rate become main restraining factors, so that on the basis of meeting the user requirements, how to design a reasonable resource scheduling method, to improve the resource utilization rate and to reduce the energy consumption become urgent problems to be solved, and the problems are one of the bottlenecks of the cloud computing development at present. According to the method, the genetic algorithm is improved and is applied to the cloud computing resource scheduling. SLA constraint and energy consumption constraint are used as fitness functions, so that a virtual machine can find a most proper placement strategy when being created on a physical machine; and on the basis of meeting the user requirements, the improved genetic algorithm can reduce the energy consumption and can generate the optimal economic benefit to the greatest degree.

The invention discloses a cloud computing task scheduling method based on an improved genetic algorithm. A multi-objective-based weighted optimization evaluation method is adopted as a fitness function; Coding is carried out in a real number direct coding mode; Generating an initial population by adopting a static algorithm and random generation combined method; Carrying out selection operation byadopting a hierarchical selection strategy; Performing crossover operation according to the crossover probability and the chromosome difference degree; Performing mutation operation according to thedynamic mutation probability; Judging whether the program is ended or not according to the double-termination condition; And finding an optimal scheme and distributing. Multiple steps of a traditionalgenetic algorithm are optimized and improved, and the user satisfaction degree and the algorithm execution efficiency are improved.

Provided are an excavating device and an excavating method of binary system program loopholes. The excavating device is provided with a static analysis module, a debugger module, a genetic algorithm module, a test input generation module and an abnormal monitoring module which are sequentially connected and a dynamic pollution tracking module located between the debugger module and the genetic algorithm module. The excavating device is guided to generate a test case by aid of a fitness function of the genetic algorithm, the multi-objective fitness function is designed, and the test case is evaluated in view of quantification. Dynamic pollution tracking is used for identifying key bytes in input files to narrow search space of the genetic algorithm. The excavating device combines advantages of the genetic algorithm and the dynamic pollution tracking and is guided to generate the test case, so that the generated test case is strong in pertinence, generated test data are high in accuracy and efficiency, further qualitative analysis and quantitative calculation are combined, and therefore path explosion in binary system program testing based on the symbolic execution and constraint solving technology is avoided.

The invention provides a virtual synchronous generator virtual inertia and virtual damping coefficient adaptive control method, which relates to the technical field of smart grid and intelligent algorithms. Firstly, the inverter based on the virtual synchronous generator is modeled, and the correlation between the output frequency of the inverter and the virtual inertia J and the virtual damping coefficient D is obtained. Then the fitness function of the adaptive control method of virtual inertia J and virtual damping coefficient D based on the improved particle swarm optimization algorithm isdetermined. Finally, the improved particle swarm optimization algorithm is applied to the active power. In the frequency control part, the real-time adaptive control of virtual inertia J and virtualdamping coefficient D is realized in order to minimize the frequency deviation and stabilize the system. The invention provides a virtual synchronous generator virtual inertia and virtual damping coefficient real-time adaptive control method, which fully utilizes the characteristics of the virtual inertia and introduces the virtual damping coefficient, so that the inverter is more stable and the frequency offset is smaller.

The invention relates to the technical field of exercise fitness equipment, in particular to an interactive fitness system and method based on virtual scenes. The system comprises fitness equipment, aKinect sensor, a large-screen displayer, a computer host, a loudspeaker and a fitness management system. Fitness users can select different instruments to be combined with the virtual scenes, and byintroducing action evaluation and correction and feedback of multiple aspects of vision, hearing and tactile sense, fitness actions are corrected and standardized in time, so that the effect of quickly developing standard fitness is achieved. The fitness actions of the users are evaluated in real time, action trigger thresholds or program types are self-adaptively changed according to completion effects so as to adjust the training difficulty coefficient, and meanwhile, the training difficulty can be adjusted by adjusting the training frequency and training time and selecting different instruments. When the system detects that the users complete a certain fitness action completely correctly, the number of training is automatically increased by one until the users start a next training program after the users complete training in a targeted number set by the training program, and then the training rhythm is actively controlled.

The invention provides a power distribution network fault positioning method based on improvement of a binary particle swarm algorithm, the conventional binary particle swarm algorithm is improved, and the method is applied to positioning of power distribution network faults. The method comprises following steps: firstly, determining parameters including the particle swarm scale and the maximum iteration frequency etc.; then forming an expectation function of a switch according to fault information of the switch, and constructing a fitness function of power distribution network fault positioning; initializing a particle swarm, setting particle positions, and setting the speed of the particles as 0; calculating the fitness values of the particles according to the fitness function, and setting an initial global extremum; updating an individual extremum and the initial global extremum; updating the speed and position of the particle swarm; and stopping calculation when reaching the maximum iteration frequency, and outputting the global optimal position of the particle swarm, namely the practical fault state of each feed line section of a target power distribution network. According to the method, the problem of premature convergence of the conventional method can be overcome, and the convergence and the stability of the algorithm can be further improved.

The invention discloses an on-line calibration system for an external parameter of a monocular camera. The system comprises a straight line detection module, a lane extraction module, a fast-perspective-removal (FPR) module, an optimal value iterative module, and an external parameter output module. The straight line detection module obtains an effective straight line position of an inputted infrared image; the lane extraction module carries out effective straight line searching and determination at a possibly occurring lane position; the FPR module carried out perspective transformation removing on a lane based on an imaging principle; the optimal value iterative module carries out loop iteration to enable a fitness function to be optimized, thereby obtaining an angle with the smallest error relative to a practical value; and the external parameter output module is used for outputting the angle obtained by optimal iteration. No calibration object is needed; the influence on the calibration precision by any external factor can be reduced; dependence on the straight line information of the lane is only needed during the calculation process; the flexibility is high; and on-line calibration on the camera can be realized. Moreover, no control by an external device and no strict experiment condition are needed during the calculation process. The system and method are simple and convenient to realize.

Sensors are selected from a sensor network for tracking of at least one target. The sensors are selected using a genetic algorithm construct having n chromosomes, wherein each chromosome represents one sensor, defining a fitness function based on desired attributes of the tracking, selecting one or more of the individuals for inclusion in an initial population, executing a genetic algorithm on the initial population until defined convergence criteria are met, wherein execution of the genetic algorithm has the steps of choosing the fittest individual from the population, choosing random individuals from the population and creating offspring from the fittest and randomly chosen individuals. In one embodiment, only i chromosomes are mutated during any one mutation, wherein i has a value of from 2 to n−1.

The invention relates to a power control optimization algorithm in a cognitive radio network, which belongs to the field of system resource allocation. The algorithm comprises the following steps: 1, initializing the iteration number of the algorithm, the positions and speed of particles and the basic parameters of the particle swarm; 2, calculating a fitness function value, setting the position Xa of an individual particle as the initial best position, and setting the particle with the best function value in the swarm as the initial best swarm position Gbestk; 3, searching based on a PSO algorithm, updating the best positions of the particles and the swarm and updating the speed and positions of the particles by using a fundamental formula of the particle swarm; and 4, setting a termination standard. The invention conducts study on the non-convex optimization problem controlled by the cognitive radio power and puts forward an improved particle swarm-based power control algorithm which allows utility functions such as an S-type function a convex function and the like to be non-concave, thereby conforming to the actual network better. Parameter adjustment is performed by the particle swarm algorithm to guarantee the global astringency of the algorithm. The algorithm of the invention has better validity and rapidity.

A method for using machine learning to solve problems having either a “positive” result (the event occurred) or a “negative” result (the event did not occur), in which the probability of a positive result is very low and the consequences of the positive result are significant. Training data is obtained and a subset of that data is distilled for application to a machine learning system. The training data includes some records corresponding to the positive result, some nearest neighbors from the records corresponding to the negative result, and some other records corresponding to the negative result. The machine learning system uses a co-evolution approach to obtain a rule set for predicting results after a number of cycles. The machine system uses a fitness function derived for use with the type of problem, such as a fitness function based on the sensitivity and positive predictive value of the rules. The rules are validated using the entire set of training data.

The invention belongs to the technical field of antenna array arranging optimization, especially to an optimal arranging method of antenna arrays of a sub-array level, and relates to a sparse sub-array and sparse array element antenna array integrated arranging method based on a heredity algorithm. A sub-array level array idea and sparse array element arrangement are combined to establish an arranging model, the peak side-lobe level (PSLL) of the antenna array is set as a fitness function to establish an optimization model, and the heredity algorithm is employed to solve the optimal result of the optimization model. The combination of the sub-array level array idea and the sparse array element arrangement are combined to overcome the landform problem, sub-arrays can be flexibly divided for arrangement. The optimization method combining sparse sub-arrays and sparse array elements effectively lowers PSLL, which is substantially reduced when compared with a molecule array arranging method. The convergence speed of the method is five to ten times that of a conventional sparse array arranging method, and the convergence speed is substantially improved.

The invention relates to a detection method for software vulnerability, and belongs to the technical field of information security. The detection method comprises the following specific steps of: (1) collecting a software source code program containing known vulnerability; (2) obtaining a vulnerability code table; (3) obtaining a control flow graph CFG and an assembly code of the program to be tested; (4) determining key analysis code segments in the assembly code of the program to be tested and obtaining all paths of passing through nodes corresponding to the key analysis code segments through the control flow graph CFG of the program to be tested; and (5) for each suspicious path, an improved Fuzzing test method is used for detection to obtain the test results of the program to be tested. In the invention, when a fitness function is determined, the dependence on the source code of the tested program is eliminated, and the availability of information is kept; and the redundant information in an abstract syntax tree is eliminated and the integrity of useful information is kept, therefore, the practicality is better, and the efficiency is higher.

A Soft Computing (SC) optimizer for designing a Knowledge Base (KB) to be used in a control system for controlling a suspension system is described. The SC optimizer includes a fuzzy inference engine based on a Fuzzy Neural Network (FNN). The SC Optimizer provides Fuzzy Inference System (FIS) structure selection, FIS structure optimization method selection, and teaching signal selection and generation. The user selects a fuzzy model, including one or more of: the number of input and/or output variables; the type of fuzzy inference model (e.g., Mamdani, Sugeno, Tsukamoto, etc.); and the preliminary type of membership functions. A Genetic Algorithm (GA) is used to optimize linguistic variable parameters and the input-output training patterns. A GA is also used to optimize the rule base, using the fuzzy model, optimal linguistic variable parameters, and a teaching signal. The GA produces a near-optimal FNN. The near-optimal FNN can be improved using classical derivative-based optimization procedures. The FIS structure found by the GA is optimized with a fitness function based on a response of the actual suspension system model of the controlled suspension system. The SC optimizer produces a robust KB that is typically smaller that the KB produced by prior art methods.

The invention discloses a mixed energy storing capacity optimization method for optimizing micro-grid call wire power. A fitness function of a genetic algorithm is constructed through a mixed energy storing capacity optimization model and sample data of mixed energy storing total charging/discharging power is obtained; a frequency range is subjected to binary coding to generate a first-generation dividing frequency fP group; active powers PBESS and PSC of a storage battery and a super capacitor of any one individual in the first-generation group are respectively calculated according to the mixed energy storing capacity optimization model; rated capacities of the storage battery and the super capacitor are respectively calculated through the active powers PBESS and PSC of the storage battery and the super capacitor; an engineering period net present value of a mixed energy storing system, the service life of the storage battery and the service life of the super capacitor are obtained according to the rated capacities of the storage battery and the super capacitor and the active powers of the storage battery and the super capacitor; an individual with the highest fitness in the final generation is an optimal individual; and meanwhile, an optimal mixed energy storing capacity combination is obtained. An obtained capacity result is accurate and the estimation of the service life of energy storing equipment is realized.

The invention discloses a rolling bearing fault diagnosis method based on improved variational model decomposition and an extreme learning machine. The method comprises: vibration signals of a rollingbearing under different types of faults are collected, the vibration signals are filtered by means of maximum correlation kurtosis deconvolution, parameter optimization is carried out on the maximumcorrelation kurtosis deconvolution method by using a particle swarm algorithm, and an enveloped energy entropy after signal deconvolution is used as a fitness function; the mode number of variationalmodel decomposition is improved by an energy threshold and improved variational model decomposition of the filtered vibration signals is realized to obtain mode matrixes of the corresponding vibrationsignals; singular value decomposition is carried out on the mode matrixes to obtain a singular value vector and a rolling bearing fault feature set is constructed; and the fault feature set is trained by using an extreme learning machine and a rolling bearing fault diagnosis model is established. Therefore, stable feature extraction of the complex vibration signal of the rolling bearing is realized, so that the diagnostic accuracy is improved.