348 results about "Support vector regression machine" patented technology

The invention provides a wind-field power combination predicting method based on a fuzzy neural network and a support vector machine, which comprises the following steps of: acquiring and pre-processing data; setting up a fuzzy neural network model by using a normalized training sample set and a prediction sample set and predicting; setting up a support vector machine model and predicting; linearly combining the prediction results of the two algorithms to obtain a wind speed prediction value; and setting up a wind speed power expert table via historical data, and inquiring the expert table according to the predicted wind speed value so as to obtain a power prediction result. By the method provided by the invention, the short-term prediction of a wind speed sequence can be effectively realized, the power prediction precision is improved, and fewer computing resources are consumed.

A method of ranking workers for an incoming task includes recording a list of completed tasks in a computer data structure, extracting first attributes from the list for the tasks that were completed during a pre-determined period, generating a first feature vector for each task and worker from the first extracted attributes, training a Support Vector Machine (SVM) based on the feature vector to output a weight vector, extracting second attributes from an incoming task, generating a second feature vector for each worker based on the second extracted attributes, and ranking the workers using the second feature vectors and the weight vector. The first attributes may be updated during a subsequent period to re-train the SVM on updated first feature vectors to generate an updated weight vector. The workers may be re-ranked based on the second feature vectors and the updated weight vector. Accordingly, the feature vectors are dynamic.

In a support vector regression approach to forecasting power load in an electrical grid, a feature learning scheme weights each feature in the input data with its correlation with the predicted load, increasing the prediction accuracy. The kernel matrix for the input training data is computed such that features that align better with the target variable are given greater weight. The resulting load forecast may be used to compute commands sent to demand response modules.

The invention provides an aquaculture dissolved oxygen concentration online forecasting method and an aquaculture dissolved oxygen concentration online forecasting system. The method comprises the following steps of: acquiring water quality index and related meteorological factor data in a scheduled time period to establish an original data set; performing data standardized pre-processing on the original data set by using a normalization method to acquire a training sample data set of a least square support vector regression (LSSVR) machine model; training the LSSVR machine model by using the training sample data set, and optimizing the parameters of the LSSVR machine model to acquire an optimal LSSVR machine model; and acquiring water quality index and related meteorological factor data of an aquaculture ecological environment on line in real time, and inputting the acquired water quality index and related meteorological factor data to the optimal LSSVR machine model to acquire a dissolved oxygen concentration forecasting value. By using the method and the system, accurate and efficient forecasting of the aquaculture dissolved oxygen concentration is realized.

The invention provides a gear fault diagnosis method based on improving multivariable predictive models. The method comprises the following steps: measuring the vibration signal of a fault object; extracting a fault characteristic value from the vibration signal, namely the instantaneous amplitude entropy of local characteristic scale decomposition; dividing the fault characteristic value into a training sample and a test sample; respectively carrying out training of multivariable predictive models based on a support vector regression machine method to the training sample to establish the best variable predictive model, and classifying the test sample according to the best variable predictive model; and distinguishing the operating state and the fault type of the fault object according to the classification result. The gear fault diagnosis method based on improving the multivariable predictive models has higher resolution in the model recognition process.

The invention discloses an indoor passive positioning method based on channel state information and a support vector machine. The method comprises the following steps: firstly preprocessing the acquired channel state information data, performing de-noising and smoothness through the adoption of a density-based spatial clustering of applications with noise and a weight-based moving average algorithm, and then using the principal component analysis algorithm to extract the features. The data after the preprocessing and feature-extracting can reflect the signal change more accurately and the dimension is greatly reduced. The passive positioning adopts two-stage positioning. In the training stage, the large positioning space is divided into sub-regions, the support vector machine classification and regression model is established for each sub-region so as to acquire a statistic model for accurately representing the nonlinear relationship between the position and the signal. The two-stage positioning firstly determines the sub-regions through the classification of the support vector machine, and the precision position is determined in the sub-region through the regression of the support vector machine. The method disclosed by the invention has the beneficial effects that the passive positioning can be performed in the absence of the active participation of the target, and the indoor positioning precision is improved to sub-meter level.

The invention relates to a grid load predicting method based on a support vector regression machine. The method comprises the following steps: firstly, carrying out automatic regression (AR) modelingon history property data of nodes by a time sequence method; evaluating dimensions of the input vector in SVR according to orders of an AR model; performing SVR learning on the history data, and constructing a regression function of SVR; predicting the property of the node at the next time according to the regression function and the measured history property data, and regulating the regression function of SVR on line according to the regression function. The method can provide a data reference for dispatch, property optimization and the like of grid resources, avoid passive and blind task dispatch and enhance the efficiency of the grid environment.

Subsurface reservoir properties are predicted despite limited availability of well log and multiple seismic attribute data. The prediction is achieved by computer modeling with least square regression based on a support vector machine methodology. The computer modeling includes supervised computerized data training, cross-validation and kernel selection and parameter optimization of the support vector machine. An attributes selection technique based on cross-correlation is adopted to select most appropriate attributes used for the computerized training and prediction in the support vector machine

The invention relates to a target identification method based on training Adaboost and a support vector machine, relating to the technical field of image processing. The target identification method comprises the following steps: extracting haar characteristics of an original sample and using the haar characteristics for training to obtain a cascading classifier based on the Adaboost and the SVM (support vector machine) and then carrying out target identification on an image to be identified by using the cascading classifier to obtain a final identification result. The target identification method can accurately and rapidly identify targets such as people, vehicles and the like.

The present invention discloses a video pedestrian recognition method based on a convolution neural network. The method comprises a step of reading the video in a video database, intercepting a video frame, and extracting the HOG feature of the video frame, a step of constructing and training the convolution neural network, a step of selecting a plurality of character feature attributes and designing a support vector machine classifier for each character feature attribute and carrying out training, a step of inputting the HOG feature into a trained convolution neural network model, and carrying out sorting classification on each character feature . The method has the advantages that the method of the convolution neural network is employed to reflect a recognition rate well, the HOG feature is extracted, thus the amount of calculation is reduced, the speed is improved, the constructed convolution neural network has a certain depth, at the same time combined with a support vector machine, the classification is carried out for multiple times, and the recognition efficiency and accuracy are improved greatly.

The invention relates to an electromyographic signal gait recognition method based on particle swarm optimization and a support vector machine. A particle swarm optimization algorithm is utilized to optimize a penalty parameter and a kernel function parameter of the support vector machine so that the performance of the support vector machine can be optimized, and effective recognition and classification are achieved. Firstly, wavelet modulus maximum denoising is carried out on collected lower limb electromyographic signals; secondly, time domain feature extraction is conducted on the electromyographic signals after denoising is carried out to obtain feature samples; thirdly, parameter optimization is carried out on the support vector machine by means of the particle swarm optimization algorithm to obtain a set of optimal parameters with minimal errors, and a classifier is constructed; at last, a feature sample set of the electromyographic signals is input to the classifier, and then classification and recognition are conducted on gait states. According to the method, both accuracy and adaptivity of classification are taken into consideration, the computational process is simple and efficient, and the method has broad application prospects in the field of lower limb motion state recognition.

A software defect priority prediction method based on an improved support vector machine is mainly characterized in that an improved support vector machine model is used for modeling defect priority prediction and judging and predicting defect report processing priority. The software defect priority prediction method includes the steps: firstly, selecting solved, closed and determined error report as training data; secondly, extracting needed characteristics; thirdly, giving a sampling weight to each sample and training a classifier to classify the samples by the aid of the support vector machine on the samples; fourthly, redistributing weight vectors by the aid of obtained error rate in the manner of distributing larger weights to mistakenly classified samples and distributing smaller weights to correctly classified samples; and fifthly, sequentially iterating in the manner to finally obtain a strong classifier equal to the weighted sum of a plurality of weak classifiers. The classifiers are trained by means of machine learning, so that defect priority is automatically determined, and consumption of staff and cost is reduced.

The invention discloses a binary tree-based SVM (support vector machine) classification method. The binary tree-based SVM classification method comprises the following steps: 1, acquiring signals, namely detecting working state information of an object to be detected in N different working states through a state information detection unit, synchronously transmitting the detected signals to a data processor, and acquiring N groups of working state detection information which corresponds to the N different working states; 2, extracting characteristics; 3, acquiring training samples, namely randomly extracting m detections signals to form training sample sets respectively from the N groups of working state detection information which are subjected to the characteristic extraction; 4, determining classification priority; 5, establishing a plurality of classification models; 6 training a plurality of classification models; and 7, acquiring signals in real time and synchronously classifying. The binary tree-based SVM classification method is reasonable in design, easy to operate, convenient to implement, good in use effect and high in practical value; and optimal parameters of an SVM classifier can be chosen, influence on the classification due to noises and isolated points can be reduced, and classification speed and precision are improved.

The invention provides a no-reference video quality evaluating method, and aims to solve the problems of low relevance between an objective MOS value and a subjective MOS value output by the conventional method, weak prediction accuracy and weak generalization ability thereof. The method comprises the following steps: extracting blocking effect parameters, blur parameters and code rate parameters of video received by a receiving end; setting motion complexity parameters for the video according to the decoded time domain of the video; outputting the objective MOS value by using an evaluation model which is acquired in advance based on support vector (SV) regression (SVR) according to the blocking effect parameters, the blur parameters, the code rate parameters and the motion complexity parameters; and acquiring higher relevance with the subjective MOS value. On characteristic parameter selection, TS stream parameters are integrated with main sense injury parameters; and the evaluation model determining method adopts a method of the support vector regression. The method is applicable to videos in different resolution ratios, in particular to videos with encoder injury.

The invention discloses an analytical prediction module of disease incidence affected by environmental change. An association condition between outpatient capacity data and meteorology change and environmental pollution data is analyzed, a quantitative analysis module is built based on the incidence of the disease incidence caused by the environment change of a non-parameter Poisson regression model, and a prediction module which affects the outpatient capacity based on support vector regression is built based on the quantitative analysis module, and the weekly outpatient capacity of each department of a hospital is predicted. Therefore, a patient can avoid the influence on bad weather conditions and environmental pollution factors in advance and arrange the daily activities reasonably; furthermore, a hospital can reasonably configure the medical resources and manpower of all departments aiming at top diseases; furthermore, the emergency preparation can be done by public hygiene departments in advance, and the special crowds can be intervened in advance, therefore the morbidity can be reduced and the human living quality can be improved.

The application provides an air quality forecasting system which comprises an environment quality detection sensor and a cloud computing data processing platform, wherein the environment quality detection sensor is configured at a position of a pollution source and is used for acquiring air monitoring data and transmitting the acquired air monitoring data to a cloud computing data processing platform; and the cloud computing data processing platform is used for receiving the air monitoring data, predicting air quality of a monitored region by adopting an air quality prediction model based on a resource allocation neural network and a smooth support vector regression machine so as to obtain air quality predication data of each monitoring station, pushing the air quality predication data to an associated intelligent terminal, calculating correlation between the monitoring data of the pollution source and the air quality predication data of the monitoring stations according to the air monitoring data of the pollution source and the air quality prediction data of each monitoring station by utilizing a Gaussian point source diffusion model, and carrying out visualized presentation.

The invention discloses an aquaculture water quality short-time combination forecast method on the basis of multi-scale analysis. The method includes the steps that water quality time sequence data are acquired online and repaired; through empirical mode decomposition, the selected water quality time sequence sample set data are decomposed into IMF components and residual rn components, wherein the IMF components and the residual rn components are different in frequency scale; the IMF components and the rn components are classified, a manual bee colony optimization least square support vector regression machine, a BP neural network and an autoregressive sliding average model are respectively selected for forecast according to classifying features, and finally, all results are weighed and summed to obtain a water quality time sequence forecast result. According to the method, the original water quality time sequence data are decomposed into the components different in time frequency through the empirical mode decomposition, and change conditions in original water quality sequences can be mastered more accurately; advantages of the manual bee colony optimization least square support vector regression machine, advantages of the BP neural network and advantages of the autoregressive sliding average model are complemented and combined, and thus performance of a combined forecast model is effectively improved.

The invention provides a method for evaluating service reliability of numerical control equipment. According to the original input and output vectors of data after zero dimension treatment, an optimized nonlinear regression function for the input-output of the numerical control equipment is obtained by using a support vector regression machine for training, and furthermore the indexes of the reliability such as point estimation, confidence interval and reliable sensitivity are solved by the self-adapting selective sampling method and bootstrap method. The method can accurately evaluate and predict the influence of different factors on the service reliability of the numerical control equipment under the condition of a small sample, find out a weak link of the numerical control equipment and indicate a direction of improving the design, manufacturing, process, maintenance and the like of the numerical control equipment.

The invention discloses a method for recognizing road speed limit signs by optimizing an SVM (support vector machine) by adaptive mutation particle swarm optimization based on a GPU (graphics processing unit). The category of the road speed limit signs is quickly and accurately recognized by using PSO (particle swarm optimization) to optimize parameters of an SVM. Based on the characteristics of particle swarm in optimizing parameters of the SVM, such as high data processing capacity and long computation time, the operating speed of PSO algorithm is increased by parallel computing of the GPU. The method has the advantages that the method of optimizing the support vector machine by GPU-accelerated ALTM PSO (adaptive local tone mapping-particle swarm optimization) is superior to the traditional SVM in accuracy of recognizing road speed limit signs and is superior to the standard PSO-SVM in algorithm convergence and operating speed.

The invention relates to an unmanned vehicle intersection detection method based on three-dimensional laser radar. In the traveling process of an unmanned vehicle, the laser radar collects ambient environment data and inputs the data to a support vector regression machine, thereby obtaining front intersection branch information. The training process of the support vector regression machine includes the following steps: S1. a laser radar installation error is corrected; S2. the unmanned vehicle travels along a road, the laser radar performs data collection on ambient environment, point cloud data are obtained, and an intersection node in front of the unmanned vehicle is found; S3. grid division is performed on a region of interest of each branch, thereby obtaining multiple frames of height information map; and S4. a pixel point sequence in the height information map is used as a feature vector, and characteristics of branches are used as output, thereby training the support vector regression machine. Compared with the prior art, a method for modeling different types of intersections is designed, different types of intersections can be effectively detected, and the method is suitable for different types of laser radars, and can achieve the same detection effect.

The invention discloses a parameter selection method for a support vector machine based on a hybrid bat algorithm. Regularization parameters and RBF kernel parameters have great influences on the learning performance and computation complexity. On the basis of analyzing the advantages and disadvantages of some classical parameter selection methods, an intelligent optimization algorithm is introduced to perform optimization on the parameters. The bat algorithm has the advantages of concurrency, high convergence speed and strong robustness. The bat algorithm is firstly utilized to perform optimization on the SVM parameters, then crossover, selection and mutation operators of differential evolution algorithm are introduced in allusion to a defect of early maturing of the bat algorithm, the position is further adjusted according to the three operators in each iteration process by using a bat individual, the search ability of the algorithm is enhanced, the algorithm is avoided from prematurely falling into a local optimal solution, and finally the SVM parameter selection is optimized by using an improved DEBA algorithm to obtain an excellent effect.

The invention provides an invasion detection method based on channel state information and a support vector machine. No special hardware facility is needed, an existing wireless network is fully used, and a common business router is used to realize security monitoring function. The coverage scope is wide, and privacy exposure can be prevented. The invasion detection method includes the steps of after obtaining CSI original data, conducting clustering and de noising for the subcarrier data in a channel by using a density-based clustering algorithm DBSCAN, smoothing the denoised data by using weight-based sliding average algorithm, and extracting characteristic values for data by using major constituent analyzing algorithm after data pre-processing. Data subjected to pretreatment and feature extraction can more accurately reflect the main change of signals and greatly reduce number of dimensions. The invasion detection precision is improved and the calculating complexity is reduced. The method uses an SVM classification algorithm to obtain a statistics model of non-linear dependence relation between an invasion state and a signal fingerprint, thereby achieving the purpose of invasion detection.

The invention discloses an indoor positioning method based on manifold learning and an improved support vector machine. The method comprises a step of determining a positioning area, dividing the positioning area according to an indoor structural characteristic and a layout characteristic, and obtaining a classification result, a step of obtaining offline training data, and collecting hotspot RSS signal values which can be received by the reference points in different classification area as a training data set, a step of using an isometric mapping algorithm to carry out training data characteristic extraction, a step of using the training data to carry out support vector machine classified training, using a taboo search algorithm to carry out support vector machine classification hyper parameter searching, and establishing the support vector regression model of each category at the same time, a step of carrying out online positioning, collecting the RSS signal value of each hotspot of a target, using a support vector machine classification model to carry out classification, and obtaining the rough positioning area of the target, and a step of carrying out the accurate positioning of the target by using the support vector regression model according to the classification result. According to the method, the time-varying characteristic of the wireless signal intensity is effectively suppressed, and the precision is obviously improved.

The invention discloses a target positioning and tracking method for a wireless sensor network. The method comprises the following steps: at any positioning time, estimating a position of a target according to measurement information of a sensor node, building a learning region covering the estimated position of the target, selecting an arbitrary quantity of position points in the learning region, obtaining a decision function by use of a polynomial kernel function and a mapping relationship of the distance vector from an approaching position point of an epsilon-support vector machine for regression to the sensor node and a coordinate of the position point; and obtaining an estimated value of the position of the target by inputting a distance vector from the sensor node to the target into the decision function and transmitting an estimated value of the position of the target to a base station which then fits the historical data of the position of the target and upgrades the motion patch of the target to complete target tracking. The method obviously reduces the influence of measurement errors of the sensor node on the target positioning and tracking and improves the accuracy of target tracking.

An SVR antifriction bearing performance degradation prediction method based on a krill-herd algorithm belongs to the field of functional approximation rotating machinery prediction methods. The method comprises the following steps: firstly analyzing time domain, frequency domain and time-frequency domain feature indexes, and proposing a feature extraction method based on combination of CEEMD and wavelet packet half-soft threshold noise reduction to perform fault diagnosis of an antifriction bearing; performing comprehensive evaluation of the fault degradation feature of the antifriction bearing for multiple feature parameters, and proposing a method of combining the LLE nonlinear feature dimension reduction method with the fuzzy C mean value; and finally, introducing the basic theory of the support vector regression machine, and proposing the prediction model of multivariable support vector regression machine based on the krill herd algorithm, optimizing parameters of the SVR, and selecting the optimal C, [sigma] parameters. The method is advantaged by high prediction precision, short calculation time, and good feature value prediction effect after clustering. The degradation process of the antifriction bearing can be precisely predicted through the abovementioned three steps.

The invention relates to the technical field of tunnel excavators, in particular to a TBM cutting tool life prediction method based on a data driven support vector regression machine. The method comprises the following steps that 1, data of the TBM cutting tool excavation site is collected; 2, driving factors that affect the TBM cutting tool life is determined, and a sample data set of the driving factors is established as a training set; 3,a prediction model of a multi-kernel support vector regression machine is constructed, the training set is input, training is conducted on the prediction model to determine corresponding optimal parameters and penalty function C and insensitive loss function parameter epsilon; 4, an optimal kernel function of the prediction model is determined; 5, the sample data set of the driving factors of the cutting tools to be predicted serves as a prediction sample set, the prediction model is input, and a prediction result is obtained. According to the TBM cutting tool life prediction method based on the data driven support vector regression machine, a large amount of data of the excavation site is selected as parameters, and on the basis of a model based on support vectors regression machine is constructed, and the accuracy of a prediction tool is improved.

The invention discloses an SVM (support vector machine) parameter selection method based on multi-kernel function adaptive fusion. Compared with the prior art, the method achieves the specific analysis of the characteristics of a local kernel function, a global kernel function, a mixed kernel function and a multi-kernel function. The method comprises the steps: combining all fusion coefficients, kernel function parameters and regression parameters of the multi-kernel function together to serve as a parameter state vector, thereby enabling a model selection problem to be converted into a state estimation problem of a nonlinear system; carrying out the parameter estimation through fifth-order volume Kalman filtering, and achieving the adaptive fusion of weighted coefficients of the multi-kernel function and the selection of kernel parameters and regression parameters.

The invention discloses a non-standard character recognition method based on a convolution neural network and a support vector machine. The method comprises steps of 1, acquiring image signals of non-standard characters to serve as sample data; 2, establishing a convolution neural network and carrying out initialization; 3, passing the trained sample data through the convolution neural network so as to finish forward propagation; 4, carrying out error calculation and gradient calculation on a multi-layered perceptron in the step 3, and if errors are converged, extracting characteristic data and entering the step 6, or else, entering the step 5; 5, using a back propagation algorithm to propagate the errors and the gradients obtained in the step 4 to a network base layer through the convolution neural network layers by layers, judging whether the grid base layer is an input layer, and if yes, entering the step 3, or else, continuing to judge whether the next layer is the input layer until the input layer is determined and entering the step 3; 6, transmitting the characteristic data to a support vector machine for training and establishing a non-standard character recognition training model; and 7, inputting to-be-recognized non-standard character signals into the non-standard character recognition training model for recognition.

The invention provides a method for monitoring the state of a gearbox of a wind power generation set. The method includes the following steps of: collecting historical data of a wind power generation set SCADA system, screening out the active power, wind speed, cabin temperature, principle shaft rotation speed and gearbox oil temperature of the set under the healthy operation condition, and establishing a standard expert database; optimizing the penalty coefficient and the nuclear parameter of a least squares support vector regression machine by using a gravitational search algorithm, and establishing a gearbox oil temperature mapping model under the healthy operation condition by taking the active power, wind speed, cabin temperature, principle shaft rotation speed in the expert database as inputs and the gearbox oil temperature as an output and based on the optimized vector machine model; monitoring the gearbox of the wind power generation set in real time by using the mapping model, inputting the actually measured values of the active power, wind speed, cabin temperature and principle shaft rotation speed to obtain the predicted value of the gearbox oil temperature, defining the quotient of the predicted value of the oil temperature and the actually measured value as a judging index, and judging that a failure occurs in the gearbox of the wind power generation set and giving an alarm, if the statistical property of the judging index is abnormal. The method can be widely applied to early warning of the gearbox of the wind power generation set.

The invention discloses an iron steel electricity consumption forecasting method and device. The forecasting method comprises the following steps: acquiring iron steel historical data of an area to be forecasted within a preset period of time to serve as sample data, wherein the iron steel historical data comprises electricity consumption index and iron steel electricity consumption; performing dimensionless treatment on the sample data to obtain normalized electricity consumption index and normalized iron steel electricity consumption; adopting the normalized electricity consumption index as an input variable, and the normalized iron steel electricity consumption as an output variable to realize network training so as to build a nerve network model; adopting the normalized electricity consumption index as the input variable and the normalized iron steel electricity consumption as the output variable to build a support vector regression model; combining the nerve network model and the support vector regression model to obtain a combined model, and using the combined model to forecast the iron steel electricity consumption. According to the invention, the demand on the forecasting precision of the iron steel electricity consumption is satisfied to the utmost extent, and reference is provided for economical and reasonable planning of a power grid.