A new support vector machine optimized by improved particle swarm optimization and its application

A new support vector machine （SVM） optimized by an improved particle swarm optimization （PSO） combined with simulated annealing algorithm （SA） was proposed. By incorporating with the simulated annealing method, the global searching capacity of the particle swarm optimization（SAPSO） was enchanced, and the searching capacity of the particle swarm optimization was studied. Then, the improyed particle swarm optimization algorithm was used to optimize the parameters of SVM （c,σ and ε）. Based on the operational data provided by a regional power grid in north China, the method was used in the actual short term load forecasting. The results show that compared to the PSO-SVM and the traditional SVM, the average time of the proposed method in the experimental process reduces by 11.6 s and 31.1 s, and the precision of the proposed method increases by 1.24% and 3.18%, respectively. So, the improved method is better than the PSO-SVM and the traditional SVM.     

LSSVM Predictive Control for Calcination Zone Temperature in Rotary Kiln with IHS Algorithm

The calcination zone temperature control is an important problem in rotary kiln production process. In order to solve this problem,a predictive control method based on improved harmony search algorithm( IHS)and least square support vector machine( LSSVM) is proposed. LSSVM is utilized to bulid the nonlinear predictive model of calcination zone temperature in rotary kiln. The calcination zone temperature can be predicted through input control variable,the error and error correction of output feedback. The performance index function is established by deviation and control variable. An IHS algorithm with better fitness and faster convergence speed is proposed. The optimal control variable can be obtained by rolling optimization through this IHS algorithm. The stability of this predictive control method is proved to be feasible. The simulation and actual experiment results show that the proposed predictive control method has good control performance.     

Study and application of time series forecasting based on rough set and Kernel method

A support vector machine time series forecasting model based on rough set data preprocessing was proposed by combining rough set attribute reduction and support vector machine regression algorithm. First, remove the redundant attribute for forecasting from condition attribute by rough set method; then use the minimum condition attribute set obtained after the reduction and the corresponding initial data, reform a new training sample set which only retain the important attributes influencing the forecasting accuracy; study and train the support vector machine with the training sample obtained after reduction, and then input the reformed testing sample set according to the minimum condition attribute and corresponding initial data. The model was tested and the mapping relation was got between the condition attribute and forecasting variable. Eventually, power supply and demand were forecasted in this model. The average absolute error rates of power consumption of the whole society and yearly maximum load are respectively 14.21% and 13.23%. It shows that RS-SVM time series forecasting model has high forecasting accuracy.     

Predicting configuration performance of modular product family using principal component analysis and support vector machine

A novel configuration performance prediction approach with combination of principal component analysis（PCA） and support vector machine（SVM） was proposed.This method can estimate the performance parameter values of a newly configured product through soft computing technique instead of practical test experiments,which helps to evaluate whether or not the product variant can satisfy the customers＇ individual requirements.The PCA technique was used to reduce and orthogonalize the module parameters that affect the product performance.Then,these extracted features were used as new input variables in SVM model to mine knowledge from the limited existing product data.The performance values of a newly configured product can be predicted by means of the trained SVM models.This PCA-SVM method can ensure that the performance prediction is executed rapidly and accurately,even under the small sample conditions.The applicability of the proposed method was verified on a family of plate electrostatic precipitators.     

Parameter Identification and Application of Slippage Kinematics for Tracked Mobile Robots

A new parameter identification method is proposed to solve the slippage problem when tracked mobile robots execute turning motions. Such motion is divided into two states in this paper:pivot turning and coupled turning between angular velocity and linear velocity. In the processing of pivot turning, the slippage parameters could be obtained by measuring the end point in a square path. In the process of coupled turning, the slippage parameters could be calculated by measuring the perimeter of a circular path and the linear distance between the start and end points. The identification results showed that slippage parameters were affected by velocity. Therefore, a fuzzy rule base was established with the basis on the identification data, and a fuzzy controller was applied to motion control and dead reckoning. This method effectively compensated for errors resulting in unequal tension between the left and right tracks, structural dimensions and slippage. The results demonstrated that the accuracy of robot positioning and control could be substantially improved on a rigid floor.     

Automatic target recognition of moving target based on empirical mode decomposition and genetic algorithm support vector machine

In order to improve measurement accuracy of moving target signals, an automatic target recognition model of moving target signals was established based on empirical mode decomposition(EMD) and support vector machine(SVM). Automatic target recognition process on the nonlinear and non-stationary of Doppler signals of military target by using automatic target recognition model can be expressed as follows. Firstly, the nonlinearity and non-stationary of Doppler signals were decomposed into a set of intrinsic mode functions(IMFs) using EMD. After the Hilbert transform of IMF, the energy ratio of each IMF to the total IMFs can be extracted as the features of military target. Then, the SVM was trained through using the energy ratio to classify the military targets, and genetic algorithm(GA) was used to optimize SVM parameters in the solution space. The experimental results show that this algorithm can achieve the recognition accuracies of 86.15%, 87.93%, and 82.28% for tank, vehicle and soldier, respectively.     

A Generalized Acceleration and Deceleration Approach for Continuous Small Line Blocks with Look-ahead

Generally complex 3D contours are divided into a lot of continuous small line blocks by CAD/CAM software. When these small line blocks are used in conventional way,machine tool has to stop at the end of one move before continuing on to the next to meet accuracy requirement,which results in inefficiency.Look-ahead is an intelligent function that aims at adjusting the feed rate automatically to achieve maximum productivity while maintaining accuracy.By now most researchers just utilize the simplest linear acceleration(ACC)and deceleration(DEC)to deal with look-ahead intelligence.A generalized ACC/DEC ap- proach and corresponding optimal look-ahead algorithm based on dynamic back tracking along a doubly linked list are proposed.An improved rounding strategy for reducing interpolation errors is also presented.By using the proposed techniques,arbitrary velocity profiles that offer look-ahead feature and have the desired ACC/DEC characteristics for movement of a lot of continuous line blocks can be generated efficiently.Both simulations and experiments showed the productivity was dramatically increased without sacri- fice of accuracy.     

Heating load interval forecasting approach based on support vector regression and error estimation

As the existing heating load forecasting methods are almostly point forecasting,an interval forecasting approach based on Support Vector Regression (SVR) and interval estimation of relative error is proposed in this paper.The forecasting output can be defined as energy saving control setting value of heating supply substation;meanwhile,it can also provide a practical basis for heating dispatching and peak load regulating operation.By means of the proposed approach,SVR model is used to point forecasting and the error interval can be gained by using nonparametric kernel estimation to the forecast error,which avoid the distributional assumptions.Combining the point forecasting results and error interval,the forecast confidence interval is obtained.Finally,the proposed model is performed through simulations by applying it to the data from a heating supply network in Harbin,and the results show that the method can meet the demands of energy saving control and heating dispatching.     

Wind speed prediction by chaotic operator network based on Kalman Filter

A novel prediction network composed of some chaotic operators is proposed to predict the wind speed series.Training samples are constructed by the theory of phase space reconstruction.Genetic algorithm is adopted to optimize the control parameters of chaotic operators to change the dynamic characteristic of the network to approach to the predicted system.In this way,the dynamic prediction of wind speed series can be completed.The wind acceleration series can also be predicted by the same network.And the prediction results of both series can be fused by Kalman Filter to get the optimal estimation prediction result of the wind speed series,which is superior to the result obtained by each single method.Simulation results show that the prediction network has less computation cost than BP neural network,and it has better prediction performance than BP neural network and autoregressive integrated moving average model.Kalman Filter can improve the prediction performance further.     

Mandarin Digits Speech Recognition Using Support Vector Machines

A method of applying support vector machine (SVM) in speech recognition was proposed, and a speech recognition system for mandarin digits was built up by SVMs. In the system, vectors were linearly extracted from speech feature sequence to make up time-aligned input patterns for SVM, and the decisions of several 2-class SVM classifiers were employed for constructing an N-class classifier. Four kinds of SVM kernel functions were compared in the experiments of speaker-independent speech recognition of mandarin digits. And the kernel of radial basis function has the highest accurate rate of 99.33 %, which is better than that of the baseline system based on hidden Markov models (HMM) (97.08%). And the experiments also show that SVM can outperform HMM especially when the samples for learning were very limited.     

Improved channel estimation based on adaptive SVM for NC-OFDM systems

Non-contiguous OFDM (NC-OFDM) system is popular as its characteristics of avoiding interference by spectrum sensing. The traditional channel estimation methods can not be applied to the NC-OFDM systems directly as its special scheme. The improved time-domain channel estimation method based on adaptive support vector machine (SVM) and signal to noise ratio (SNR) estimation is proposed for NC-OFDM systems in this paper. It adopts the spectrum sensing based on adaptive SVM at the receiver. The simulation results show that the mean square error (MSE) and bit error rate (BER) performance outperform than the conventional schemes under the same condition.     

Hybrid Chinese Information Retrieval Model Based on the Combination of Keyword and Concept

A hybrid model that is based on the combination of keywords and concept was put forward. The hybrid modelis built on vector space model and probabilistic reasoning network. It not only can exert the advantages of keywordsretrieval and concept retrieval but also can compensate for their shortcomings. Their parameters can be adjustedaccording to different usage in order to accept the best information retrieval result,and it has been proved by our     

Structural-acoustic coupling characteristics of honeycomb sandwich plate based on parameter sensitivity analysis简

The structural-acoustic coupling model for isotropic thin elastic plate was extended to honeycomb sandwich plate （HSP） by applying Green function method. Then an equivalent circuit model of the weakly-strongly coupled system was proposed. Based on that, the estimation formulae of the coupled eigenfrequency were derived. The accuracy of the theoretical predictions was checked against experimental data, with good agreement achieved. Finally, the effects of HSP design parameters on the system coupling degree, the acoustic cavity eigenfrequency, and sound pressure response were analyzed. The results show that mechanical and acoustical characteristics of HSP can be improved by increasing the thickness of face sheet and reducing the mass density of material.     

Study on the Dynamic Characteristic of Lathe Headstock

In order to provide reliable data for the dynamic design or modification of a tool machine,the dynamic character- istics of the headstock,which is the main component to bear moment,must be obtained precisely.In the paper,the method based on the combination of calculation mode and experiment mode is proposed to analyze the dynamic characteristics of the headstock.The modal parameters and the mode shapes are calculated by ANSYS7.1 software.According to the FEM calculating results,the ex- periment parameters can be selected correctly.The modal parameters of the headstock have to be calculated and identified precisely. On the basis of these modal parameters,the faults of the headstock are shown and its weak points of design are illustrated.A con- clusion is drawn that some reasonable reinforce positions could greatly improve the dynamic characteristics of the system and this ap- proach is proved to be precise and reliable.     

Technological parameters modeling method of automotive ignition coils

To conform the design requirements of automotive ignition coils,a multi-input multi-output(MIMO) model is proposed to estimate the technological parameters,such as coils windings,wire diameters and iron core lengths.For small sample size properties,support vector regression (SVR) is utilized to establish the model of ignition coils,and it is verified to be more effective than artificial neural networks (ANN) in this paper.The experimental data are obtained from the typical samples of ignition coils which are specially manufactured and measured.Appropriate SVR parameters and kernel functions are determined to improve the accuracy of the model by experiments.Furthermore,an improved decomposing training algorithm is designed to increase the automation degree of sample choosing and global accuracy of the model.Simulation results verify the rationality and accuracy of the model,which shows that the proposed model can provide guidance for the design of ignition coils,and analyze the relationship between technological parameter with spark time,spark current and ignition energy.     

Angular velocity dynamics identification of small unmanned helicopter using fuzzy model

Attitude identification method for unmanned helicopter based on fuzzy model at hovering is presented.The dynamical attitude model is considered as basis for attitude control and it is very complex.To reduce the complexity of model,nonlinear model of unmanned helicopter with unknown parameters are to be determined by fuzzy system first and then derivative based gradient method is used to identify unknown parameters of model.Gradient method is used due to ability that fuzzy system is not necessarily to be linear in parameters,therefore all fuzzy sets for input and output can be adjusted.The validity of the proposed model was verified using experimental data obtained by the commercially available flight simulator X-Plane.The simulation results showed high accuracy of the modeling method and attitude dynamics data matched well with experimental data.     

An Improved Stator Flux Observation Method of Permanent Magnet Synchronous Motor

The stator flux and electromagnetic torque observation is the basis of direct torque controlled permanent magnet synchronous motor (PMSM) drive system. However, the traditional stator flux observer based on voltage model is affected by integral initial values and integral drift, that based on current model is affected by the parameters of PMSM, so a new stator flux observation method is proposed based on an improved second-order generalized integrator (SOGI). Compared to the stator flux observation method based on the conventional SOGI, the proposed method can not only overcome the influence of integral initial values and integral drift, but also completely eliminate the DC offset’s influence. Therefore, the observation accuracy of stator flux is further improved. The simulation and experimental results both show that the proposed method has a higher stator flux and electromagnetic torque observation precision.     

Compressive Measurement Identification of Linear Time-Invariant System Application in DC Motor

In traditional system identification (SI), actual values of system parameters are concealed in the input and output data; hence, it is necessary to apply estimation methods to determine the parameters. In signal processing, a signal with N elements must be sampled at least N times. Thus, most SI methods use N or more sample data to identify a model with N parameters; however, this can be improved by a new sampling theory called compressive sensing (CS). Based on CS, an SI method called compressive measurement identification (CMI) is proposed for reducing the data needed for estimation, by measuring the parameters using a series of linear measurements, rather than the measurements in sequence. In addition, the accuracy of the measurement process is guaranteed by a criterion called the restrict isometric principle. Simulations demonstrate the accuracy and robustness of CMI in an underdetermined case. Further, the dynamic process of a DC motor is identified experimentally, establishing that CMI can shorten the identification process and increase the prediction accuracy.     

Prediction of Restriction Width of Carcass by Belts for Radial Tires: Theory, Experiment and Simulation

The restriction width of carcass by the belts( RWCB) as an important parameter of radial tire design has been neglected for a long time. In order to improve the accuracy and efficiency of tire profile design,the calculating method of RWCB is proposed. The equilibrium profile is calculated by geometric model and variational approach,based on it,the predicted model of RWCB is developed for tire design. Finally,four different designs of 12R22.5 tires are investigated by experiment and finite element method,which is used to validate the accuracy of the theoretical method. Results indicate that experimental and finite element analysis results are found to be in good agreement with theoretical results; linear relationships are existed between the cord length and RWCB,and also existed between the position of belt and RWCB; tires designed by the methods have smaller and more uniform displacement,so the method can be used for tire optimized design.     

Calibration of parallel kinematics machine using generalized distance error model

This paper focus on the accuracy enhancement of parallel kinematics machine through kinematics calibration. In the calibration processing, well-structured identification Jacobian matrix construction and end-effector position and orientation measurement are two main difficulties. In this paper, the identification Jacobian matrix is constructed easily by numerical calculation utilizing the unit virtual velocity method. The generalized distance errors model is presented for avoiding measuring the position and orientation directly which is difficult to be measured. At last, a measurement tool is given for acquiring the data points in the calibration processing. Experimental studies confirmed the effectiveness of method. It is also shown in the paper that the proposed approach can be applied to other typed parallel manipulators.