This paper presents a fuzzy control strategy to build the driver's intentions. The driving performance is improved through the fuzzification of two parameters: the accelerator pedal opening and pedal opening rate to revise torque requirement. At the same time, the efficiency of energy recovery system is increased through brake pedal multi-parameter fuzzification.

The features of the Antilock-Braking System (ABS for short) while braking on split-mu road are analyzed in this paper. To guarantee the yaw stability and the braking distance, an identification method of split-mu road and the control algorithm are presented to control the changes of the yaw velocity and the sideslip angle. The simulation and test results show that the identification method of split-mu road and the control strategy can not only improve the yaw stability but also guarantee the braking distance. The proposed method and control strategy play an important role in improving the safety while braking on split-mu road.

Statistics shows that tire-road friction coefficient is critical importance for vehicle active control system. This paper proposes a tire-road friction coefficient estimation method for advanced vehicle steer-by-wire system. A vehicle model is built and a model-based friction coefficient estimator is proposed using Expended Kalman Filter (EKF) technique, which can be used for estimation of vehicle self-align torque and road adhesion condition. Simulation and experiment under on-center handling and steering wheel angle step input conditions were carried out. The results demonstrate the effectiveness and robust of the proposed estimation.

In order to reduce the computational cost of complex tyre model, the vehicle dynamics description is obtained based on the mixed data/physics tyre model and the mixed model is validated. Then, the nonlinear cascaded observer is designed using the modular estimation scheme. The selections of parameters and gain of the cascaded observer are discussed using randomized algorithm. For the sake of verifying the estimation results of the vehicle velocities, simulations are carried out under the normal and critical running conditions. The estimation results are compared with the nonlinear reduced-order observer and with the high precision vehicle dynamic model. The simulation results show that the observer parameters and gains obtained from the randomized algorithm are good estimations. Moreover, faster computation speed could be received using data/physics modeling method.

In order to obtain the transfer path of the steering wheel judder caused by brake, the running conditions of the road surface were simulated on the rotary drum test bench, the brake judder phenomenon recurred. The vibration accelerations of the caliper, caliper bracket, knuckle arm, tie rod, universal joint, and steering wheel were acquired. Meanwhile, the vibration signals of the steering wheel were analyzed at vertical, lateral, and radial directions. The first and second order frequencies were confirmed through order analysis. The transfer path from caliper to steering wheel was analyzed. The intermediate links were also analyzed during vibration transferring from caliper to steering wheel. The experiment results show that the transfer path of the brake judder is mainly caliper-caliper bracket-knuckle arm-tie rod-universal joint-steering wheel. The brake judder was enlarged or diminished during the transfer process, and it was transferred to the steering wheel after enlarging and diminishing at the three directions.

In order to achieve the energy management and control of the dual-mode electro-mechanical transmission system, a comprehensive control strategy is studied. After the analysis of the structural characteristics of the dual-mode electro-mechanical transmission system, a control strategy for improving the fuel economy is proposed. This strategy can realize the operating mode judgment of transmission system and the coordinated control between the different energy sources. Based on dSPACE system, the hardware-in-loop simulation platform is developed to evaluate the control strategy of the dual-mode electro-mechanical transmission. The simulation results indicate that the proposed control strategy can realize the energy management and control of hybrid vehicles, the optimization of the engine operating point and the improvement of the vehicle fuel economy.

In order to predict and evaluate the human thoracic injuries caused by rib fracture in car crashes, a finite element model of human ribs having the features of human anatomy structure is established and verified based on the CT scanning of a 50 percentile Chinese male volunteer. To improve the model's biological fidelity, the material parameters of ribs are gained by reverse experiment of animal tissue. The model is verified by the Kindig's load experiment of human body ribs and compared with the simulation data, which the material parameters were cited directly from the reference. The results show that the reaction force-displacement curve agrees well with the experimental curve and the model has a better biological fidelity when the material parameters were gained from the reverse engineering method.

In this paper, based on the contradiction matrix and innovation principle of TRIZ theory, three strategies are proposed for the vehicle braking system. The transition wheel load is changed by controlling the damping force of the semi-active suspension shock absorber. Thereby, the outer front wheel braking force is controlled. This can improve the vehicle handling stability when the vehicle steering wheel rotation is over large. To verify the effect of the above method vehicle running test is carried out. As a reverse damping force semi-active suspension is employed to control the vertical load, so the vehicle height sensor is not needed in the test. Test result show that, by control of the vertical damping force of the semi-active suspension shock absorber, the vehicle slip angle is significantly improved.

In order to improve the vehicle radiator capacity and keep vehicle reliable, three new improvement schemes were proposed on the foundation of former research. A 3D model based on the drawing from manufactures was set up. Heat exchange model was used as substitute for radiator module. Each improved model was simulated in virtual tunnel with commercial CFD software. These schemes were assessed based on the simulations. The results show that all of three schemes could enhance radiator performance. Addition of air exits near muffler has the best effect. Compared with original model, the intercooler, water radiator and oil radiator of the improved schemes could lower the thermal fluid outlet temperature by 14.48%, 1.39% and 2.28%, respectively.

A two-degree of freedom 1/2 vehicle vibration model of axle suspension was built. Using random vibration theory, the differential equation of vehicle vibration was derived with the road roughness model as the input excitation. The vehicle suspension characteristics under different working condition were simulated and analyzed using Simulink software. Based on the analysis of the suspension characteristics, a suspension load monitoring device with overload protection function was designed, and a method to monitor the vehicle load status was proposed. Furthermore, a vehicle load status monitoring platform for FSAW truck was developed. The program of sensor signal processing and vehicle load identification was written. The system was tested by load status detecting trial and centre-of-mass position monitoring trial. The results show that the relative measurement error is less than 5%, and the horizontal centre-of-mass of the vehicle was positioned accurately. This research provides technical support for the real-time monitoring and warning of vehicle in-transit risk status.

Based on the real time information of dynamic route volume, initial queue length, and the real-time vehicle and signal controller communication within vehicle infrastructure integration environment, an route based integrated signal coordination control model is proposed to optimize the dual-direction coordinative routes, the dynamic travel speed, and the specific intersection signal timings. Thus overcomes the drawbacks of the conventional coordinated signal control model that uses those parameters as inputs based on the assumptions that the parameters are fixed and the value is optimal. The product of the output volume and the travel speed is employed as the objective of the propose model. The decision variables include the coordinative routes, travel speed and offset. A set of constraints were set up to ensure feasibility and safety of the optimal results. Compared with the results optimized by classical Maxband model and Synchro program, the proposed model can significantly improve the green wave bandwidth, decrease the number of stops, and increase total coordinated benefits. The sensitivity analyses with fluctuation of traffic flow, section length between intersections, maximum speed limit and degree of saturation further demonstrated the potential of the proposed model to be applied in various traffic conditions.

In order to analyze the randomness of the steel corrosion in reinforced concrete in the in marine environment, the performance function of the corrosion of steel bars in reinforced concrete is proposed using reliability theory. The proposed function contains two random variables: the chloride diffusion coefficient of concrete and the thickness of the protective layer. The calculation methods of the probability of steel corrosion under some environmental conditions and the service life of reinforced concrete under a certain degree of reinforcement corrosion probability are given. This method is validated with reinforced concrete experiments that simulate seawater corrosion. The results show that the use of probabilistic methods to analyze the service life of reinforced concrete and the probability of corrosion is effective. The method can be used to evaluate the reinforced concrete durability. Decrease in the concrete permeability and appropriate increase in the thickness of the protective layer of concrete reinforced concrete can be significantly improved in the marine environment durability.

To improve the power performance and reduce energy consumption and environmental pollution, a parallel hybrid loader is taken as a sample to research the automatic shift technology of hybrid power engineering vehicles. The influence of the operating load on the optimal gear was taken into account. The throttle, speed and operating load factor were considered as the control parameters for the new automatic shift control scheme. Energy saving schedule, power shift schedule and comprehensive shift schedule were proposed. Results show under the premise of meeting the normal requirements, the proposed shift schedule is accurate and effective, the gear change and load change have good consistency. The power performance is improved and the energy consumption is reduced.

The operating characteristic of moving construction machinery is analyzed and the energy saving potential is evaluated. In view of the characteristics that moving construction machinery frequently start and stop, different energy saving schemes are proposed. Regenerative braking strategy and energy reuse strategy are designed based on the advantage of high power density of hydraulic accumulator. Prototypes of hydraulic hybrid loader and hydraulic hybrid concrete mixer truck are developed. Experiment results show that the hydraulic hybrid system can effectively recover and reuse the braking energy. The control strategies reasonably switch among various operation modes, reduce fuel consumption and improve the dynamic performance of the machine.

The damage failure problem of the final drive of tracked vehicle was investigated. It was found that the interference contact between the final drive and the walking system is the key to solve the problem. First, the interference contact parts are taken as rigid bodies, and the mechanical model and mathematic model of the interference contact are established using the deformation compatibility relationship. Second, the finite element simulation method is employed to verify the proposed model using parameter of the practical vehicle. The finite element analysis results are in good agreement with that of proposed model, proving the effectiveness of the proposed method. Finally, it is found that the main reason of the final drive damage is the insufficient strength of the wheel hub of the large gear ring, whose factor of safety is only 2.6. This leads the failure of the wheel hub of the large gear ring under bad working condition, resulting in the failure damage of the final drive. This study provides theoretical basis for the optimization of the final drive of tracked vehicle.

A boom potential energy recovery scheme for hybrid excavator is proposed, which is based on hydraulic accumulator and power battery. The multi-objective optimization method is used to match the parameters of key components of the energy recovery system. Furthermore, for the recovery system, a simulation model and control strategies are established to analyze the effect of parameter matching and control methods by simulation. Finally, a hybrid excavator test platform of boom potential recovery system is built up to study the recovery effect and handling performance. Experiment results show that not only the recovery program is feasible but also the system operation performance is stable, the parameter matching effect is good and energy recovery rate is high.

In this paper, a bionic airfoil mimicking the wing of a long-eared owl is proposed. On this basis, a bionic slat without cove and multi-element airfoil is built. In order to reveal high-lift effect of the bionic slat, the corresponding quasi-two-dimensional models are manufactured by rapid manufacturing and prototyping system. Experiments are conducted in a low-turbulence wind tunnel. The results show that the lift coefficient of the bionic airfoil is larger when the angle of attack is less than 5°, but lift coefficient of the bionic multi-element airfoil with slat is larger s when the angle of attack is greater than 5°. The bionic slat can increase the stall angle and the maximum lift coefficient; at the same time, it can also delay the decline of the lift coefficient curve slope in order to prevent the leading-edge separation within a certain range of angle of attack. Furthermore, the flow field around the models is visualized by smoke wire method, which shows the leading-edge separation of the bionic airfoil at low Reynolds numbers. However, the finding does not occur in the flow field of the bionic multi-element airfoil at the same conditions. This superiority may be used as reference in the design of the leading-edge slat or slot.

In order to analyze and simulate the dynamic characteristics of the electro-hydraulic servo loading system of the servo turret reliability test bench, a dynamic characteristics model of the electro-hydraulic servo loading system was established. The parameters of the dynamic model were estimated by the least square algorithm. For avoiding the system from the noise and preventing the Unscented Kalman Filter (UKF) algorithm from the ill-conditioned covariance matrix, a SVDUKF algorithm was proposed to smooth the force feedback signal of the hydraulic servo loading system. The filtering performances of the SVDUKF algorithm and the EKF algorithm were compared by experiments. The results show that the dynamic model with parameters estimated by the least squares algorithm has high precision; and the filtering performance of the SVDUKF algorithm has high-efficiency, which can improve the stability of the system.

The expanding tube connection by pushing through balls with multi-step for hollow assembled camshaft was presented and numerically analysed by finite element simulation. The single-step, double-step and triple-step were contrastive studied from the aspects of press-fit force, residual strain, axial loading force and maximum torsion. Static torsional strength of the assembled camshaft on expanding tube connection by multi-step pushing through balls was determined experimentally.The results show that multi-step tube expanding assembly with equal amount of interference can increase the press-fit force and connecting strength between cam and tube, improve tube expanding conditions and reduce balls load. Static torsion of assembled camshaft by the double-step or triple-step tube expanding is 12% larger than the single-step one. The new technique of assembled camshaft is suggested for its simple procedure and connecting reliability.

Finite element model was proposed to simulate the flexible roll forming process. Based on theoretical analysis of bending process, the longitudinal curvature equations of the formed part was set up. The method to design maximum height difference and the compression ratio difference in the cross-section was given. Numerical simulation was carried out, and the simulation results are in good agreement with the theoretical values. A small experimental device was developed and flexible roll forming tests were carried out. The results demonstrate that flexible roll forming can be used for forming three-dimensional surface parts efficiently.

In view of the limitation of the Forming Limit Diagram (FLD), which is used to evaluate the formability of AA7075 under hot stamping conditions, the continuum damage mechanics was introduced into the prediction of the FLD. A set of uniaxial damage constitutive equations was presented and the material constants were optimized. Based on the analysis of the features of Forming Limit Curves (FLCs) for AA7075 at different temperatures, a plane-stress damage equation was proposed to take account of the damage of material at different stress-strain state. Then the FLCs were predicted on the condition of four sets of temperatures and three sets of strain rates. The results indicate that the stress-strain curves of the numerical analysis of uniaxial damage constitutive equation and the experimental curves are basically consistent. Using the plane-stress unified viscoplastic damage constitutive equations can predict the FLD at different temperature and strain rate in the hot stamping process and provide the failure criteria.

P-doped nanosized TiO₂ powders (P-TiO₂) were synthesized by a sol-gel method with tetrabutyl titanate and phosphoric acid as Ti and phosphorus resources, and then immobilized on diatomite (P-TiO₂-Dt). P-doped TiO₂ powders and P-TiO₂-diatomite were characterized by XRD, FTIR, SEM and UV-Vis. The photocatalytic activities of P-TiO₂ and P-TiO₂-Dt were investigated by the degradation of methyl orange and methylene blue under visible light. The results show that phosphorus is incorporated into the TiO₂ lattice. The transformation temperature from anatase-to-rutile phase and the crystal growth of TiO₂ are remarkably improved. The morphology of diatomite is not destroyed by supported P-TiO₂. According to the diffuse reflectance UV-Vis spectra, P doping and supported by diatomite both lead to absorption edge of TiO₂ and shift to the visible-light region. This implies that P-TiO₂ and P-TiO₂-Dt have photocatalytic activities under visible-light irradiation. As a result, the P-TiO₂-Dt photocatalyst exhibits higher visible-light catalytic effect than unloaded P-TiO₂, which is due to the interactions between diatomite support and P-TiO₂. The photocatalytic experiments prove that the sample with P-TiO₂-Dt content of 20% exhibits the best photocatalytic activity.

A complex interactive activity recognition approach with spatial-temporal relation and an extended Hidden Markov Model with multi-observations and multi-layers is presented here. Interactive activities involving multi-objects are naturally hierarchical and related with spatial-temporal relationship. The multi-granularity features (group, two persons and single person) are used. A new model, Multi-observations Three-Layers Hidden Markov Model (MTHMM), corresponding to these features is put forward. The experiments show that the new feature extraction method and the new model have a good performance and a fair robustness in complex interactive activity recognition.

In order to resolve the low accuracy and inefficiency of reconstruction of Gene Regulatory Networks (GRNs) in system biology, we proposed a novel inference algorithm from gene expression data based on differential equation model. In this algorithm, two methods are employed for inferring GRNs. One is Singular Value Decomposition (SVD) method and the other one is Gravitation Field Algorithm (GFA). The SVD method is used to decompose gene expression data, determine the algorithm solution space, and get all candidate solutions of GRNs. The GFA is the kernel part of the proposed algorithm. The GFA is divided into four parts: initialization, division of solution space, movement operator and absorption. Random group method is used in division of solution space. Every element movement method is used in movement operator. The proposed algorithm is validated on both the simulated scale-free network and real benchmark gene regulatory network in network database. Both genetic algorithm and simulated annealing are also used to evaluate GFA. The cross-validation results confirm the effectiveness of the proposed algorithm, which outperforms significantly other existing algorithms.

To overcome the defect of the real-time system fault-tolerant model that could only tolerate one fault occurred during the task execution process, a task Local Optimum Checkpoint Interval (LOCI) iterative algorithm was proposed. The algorithm is based on the checkpoint and rollback recovery fault tolerant model to make the task of the real-time system could tolerate multiple faults occurred during the task execution process. Then the Checkpoint Interval Global Optimization (CIGO) algorithm of the task set was proposed. The schedulability of the real-time system was analyzed based on the different checkpoint interval optimization mechanisms according to the calculation formula of the task worst case response time. The simulation results show that the strategy based on the GIGI can significantly improve the system fault resilience compared to that based on the LOCI.

Despite the factor that Support Vector Domain Description (SVDD) model is an effective method for describing a set of training data, one inherent drawback is that the description is very sensitive to the selection of the trade-off parameter, which is hard to estimate in practice. To solve the difficulty, we proposed a novel Variable Trade-off parameter Support Vector Domain Description (VT-SVDD). In the proposed VT-SVDD, first, we assigned a position-based variable trade-off parameter to each data point. Then we computed a convex constrained quadratic programming based on the variable trade-off parameters. Finally, we can obtain a spherical data domain description for the training data. Experimental results demonstrate that the VT-SVDD can significantly improve the accuracy and robustness on UCI data sets.

In order to improve the detection accuracy of spliced images, a new blind detection based on the Visual Attention Model (VAM) was proposed in this study. First, the Edge Conspicuous Map (ECM) is created by an improved Order Statistics Filter (OSF) based nonlinear filtering approach; then, the ECM fixations are extracted by VAM, and the fixations on the boundaries are located by conspicuous edge positioning method, accordingly the key feature fragments are captured. Second, the Extended Hidden Markov Model (E-HMM) features are extracted from each wavelet reconstructed image of Cr channel of the fragments, and their dimensions are reduced by SVM-RFE. Finally, the above features are trained and classified using SVM, by which the spliced images can be identified from the natural ones. The experimental results show that, when testing on the Columbia image splicing detection dataset, the detection accuracy of the proposed method is 96.32%.

In this paper, an anonymous area estimation method of path data based on Bayesian network is proposed. This method is based on the path network model, combined with road network data, and the maximum likelihood theory is used to determine the minimum anonymous region estimation. Simulation results show that the proposed anonymous area estimation method can provide timely and proper anonymous region suggestion for the privacy protection based on the position service, thus, the service quality in terms of anonymous time and precision. The proposed method also makes use of the known path data for objective reasoning and providing reasonable suggestion to individuation anonymous demands, which solves the major uncertainty problems in the mobile user queries.

Based on the research and analysis of probabilistic classification and its influence on the overall accuracy,a new online traffic classification method is proposed. First,the Relevant Vector Machine (RVM) is used to classify traffic flows and output probabilistic classification.Then,the flows, whose classification probability is in doubting interval[0.1,0.9],are re-identified by using port & Deep Packet Inspection (DPI).If the predicted probability is in the interval [0,0.1] and [0.9,1],the classification is totally accepted.Experiment studies demonstrate that the proposed method can reach the overall accuracy of 98%,and perform well in online network traffic classification.

In this paper, a lightweight certificateless signature scheme is proposed for Wireless Sensor Networks (WSNs) under the condition of limited recourses. The signature scheme is more efficient by using online/offline approach and has no bilinear pairing operation during signing and verifying process compared with the Xirrus Management System (XMS) and Paterson scheme. Based on this signature scheme, a key agreement scheme is also proposed, which can be used by every two nodes in WSN to establish a session key safely. Finally, this signature scheme is proved to be securing that can defense the two attacks proposed in Al-Riyami and Paterson security model, and the security proof can be reduce to discrete logarithm problem.

In HMIPv6 networks with multi-level MAP architecture, it remains a significant challenge to select an optimal serving MAP for the mobile terminal to optimize the whole network performance. By analyzing and modeling the behavior characteristics of the mobile terminals, mainly including the mobility characteristics and the service characteristics, the overall performance objective is determined. Then, an adaptive MAP selection strategy with multi-objective cooperation is proposed (A-MAP), in which the load balancing among the available upper-layer MAPs is comprehensively taken into account. Considering the proportional allocation of mobile terminals with different velocities, a set of experimental scenarios is elaborately designed based on the NS-2 network simulation platform, and three typical MAP selection strategies are selected for the comparison. The performance efficiency of the proposed A-MAP on system cost, load balancing, and average handover latency is verified by adjusting the session arrival rate.

For multi-relay-assisted Orthogonal Frequency Division Multiplexing (OFDM) cooperative systems with Amplify-and-Forward (AF) mode, when relay nodes are close to the source node, higher system gain can be achieved, otherwise, the system will get lower gain. Under the condition that each relay node has individual power constraint, the more subcarriers relay nodes forward, the less power of each sub-carrier is allocated. Greater power gain can be acquired when some of the sub-carriers are forwarded by idle relay nodes instead of the previous busy node. Therefore, we propose an improved algorithm. In this algorithm, when relay nodes are close to the source node, we make full use of the power of the free relay nodes to increase system capacity. When relay nodes are far from the source node, the relay node number is reduced to decrease system complexity. Simulation results show that the improved algorithm can obtain higher system capacity when the relay nodes are close to the source node, and it also can approximately maintain system capacity unchanged with lower system complexity when relay nodes are far from the source node.

This paper presents two methods of high-resolution direction-of-arrival estimation based on signal sparse representation. First, a coarse-refined grid separation is created, which is used to structure the sparse representation of array covariance column vectors corresponding to overcomplete basis. Then, a removing and differencing method is employed to suppress the inference of noise. A regularized FOCUSS algorithm with lp norm constraint is applied for sparse reconstruction. Finally, the Singular Value Decomposition (SVD) is adopted to remove the singular value, which is less than the threshold to reduce complexity as well as ill-posed problem of overcomplete basis. Simulation results validate the effectiveness and robustness of the proposed method.

In order to improve the resolution of spatial spectrum estimation multiple signal classification (MUSIC) algorithm in the presence of colored noise fields, a new spatial spectrum estimation method was proposed. This method is based on the receiving characteristic of pulse signals in an actual direction finding system, when decoy signals overlap radar signals. In this method, first, the receiving data are divided into several segments; then the number of signals of every receiving data segment is estimated; finally, the covariance matrix of every receiving data segment is normalized by covariance matrix diagonal elements. The covariance matrix of every signal is given by using covariance difference MUSIC algorithm. By searching spectrum peak, the result of Direction of Arrival (DOA) estimation can be obtained step by step. Both simulation and test based on actual direction finding system prove the proposed method realizable.

In this paper, the problem of symbol detection in the presence of unknown carrier phase is studied. A joint symbol detection and phase estimation algorithm based on the Expectation-Maximum (EM) algorithm is developed, which is different from the traditional optimum receive algorithm. The algorithm does not need to estimate the carrier phase and obtain symbol information directly. After the algorithm convergence, it obtains signal information, at the same time estimates the maximum-likelihood (ML) of unknown carrier phase, and consequently realizes the joint symbol detection and phase estimation. Because the algorithm does not require synchronization parameters estimation, it can reduce the loss of performance, which is brought by the hierarchical processing of the traditional method. Simulation results show that, in comparison with the traditional optimum receive algorithm, which has to estimate the phase by non-data-aided first, then detect symbol information, the performance of the proposed algorithm is improved significantly.

Traditional DV-Hop algorithm can cause greater positioning errors. To overcome such problem, the traditional DV-Hop algorithm was analyzed and simulated, and two factors affecting positioning accurate were acquired. Then an improved positioning algorithm was proposed. Two communication ranges were used by beacon during localization. Using this method more accurate hops between beacon and unknown node can be gotten and the estimating distances between beacon and unknown node are close to real distances. Numerous simulation results show that the relative positioning error of the improved algorithm is 13% to 15% less than that of traditional DV-Hop algorithm. Moreover, the differences of positioning errors in different network topologies are decreased by using this improved algorithm.

Traditional methods for lung nodule recognition need to extract the features of the Region of Interests (ROIs), which usually leads to loss of some implicit structure information. To avoid this problem, a novel Multiple Kernel Learning method based on Matrix Least Square Support Vector Machine (MKL-MatLSSVM) is proposed. This method combines the advantages of both MKL method and MatLSSVM, and supports direct matrix input, suitable for image identification. To verify the effectiveness of the proposed method, it was applied to identify lung nodules in CT images of 20 patients, where the extracted ROIs contain 80 nodules and 190 false positives. The results show that when using hybrid or Radial Basis Function (RBF) kernels in MKL-MatLSSVM, the resulting sensitivity, accuracy and specificity can be balanced, and the area under the Receiver Operating Characteristic (ROC) curve can reach 96%, better than other two previous Support Vector Machine (SVM) methods that include MatLSSVM.

In order to analyze human activities accurately, a new analysis approach is proposed combining with real-time water and electrical events. First, applying infrastructure-mediated sensing technology, a system of sensing residential water and power consumption is designed to obtain the water and electrical usage data. Then, using the power spectrum vector chasing algorithm and k-Nearest Neighbor algorithm, working devices in the home can be recognized and classified according to the abundant of high electromagnetic interference noises generated during the switching of power supply devices. The fixtures in the home can be classified using sliding window technique and pattern recognition algorithm based on the inner product of vector. Via being deployed in real-houses, this approach successfully classifies consumption events in device-level with 95% accuracy. Finally, the analysis and post-process results can be transmitted to a clouding storage via internet. Disaggregated residential water and power consumption is the fundamental of energy monitoring and human activity analysis in a home.

A novel fusion algorithm based on nonsubsampled Bandelet transform is proposed for the multifocus images of same scene. First, the average image is calculated from the source images, after that the best geometric flow is computed using the wavelet coefficients of the image and a Bandelet basis is constructed. Secondly, the source images are represented using the Bandelet basis, and then the obtained coefficients are fused respectively according to their physical significance. Finally, the fused coefficients are applied on the inverse nonsubsampled Bandelet transform to obtain the fused image. The experiment results indicate that, compared with the fusion methods based on Wavelet transform and nonsubsampled Wavelet transform, respectively, the proposed method can provide better fusion quality in terms of both visual sense and quantified measure.

In order to improve the multicell downlink capacity, an interference alignment scheme for multicell multiuser system through Tomlinson-Harashima Precoding (THP) is proposed. In this scheme, the appropriate feedback triangular matrices, feedforward matrices and receiving matrices are designed to align the interferences, thus effectively improving each cell's capacity. The design criteria of the proposed scheme are given. Theoretical analysis and simulation demonstrate that the proposed scheme can obtain the same degree of freedom (DOF) and more capacity compared with existing schemes.

The control of the Air-Fuel Ratio (AFR) for SI engines is investigated. First, the traditional Volterra model is improved, and a modified multi-input multi-output Volterra model with variable sampling intervals is proposed. This model is employed to achieve engine AFR adaptive modeling online. Then, on this basis, a nonlinear model predicting control method based on the modified Volterra model is developed. Finally, using Matlab, simulation of the proposed control model is carried out with a widely accepted SI engine benchmark, and compared with the currently used PI controller algorithm. The results show that the proposed method is effective.

A simulation method based on Kalman filter for nonlinear continuous systems fulfilled by trigger subsystem block is proposed. In this method the form of Simulink blocks is combined with M-functions to simulate the process of periodic sampling and processing by micro-computers, accordingly the functions of D/A and A/D and the online filter processing are achieved perfectly. The implementation process of the proposed method is introduced. Simulation of the Continuous Stirred Tank Reactor (CSTR) system is carried out to demonstrate the implementation of the proposed method in detail. Results show the efficiency of the proposed method.

The problem of reliable control of discrete-time T-S fuzzy systems with actuator faults is investigated and the design scheme of a reliable controller is proposed. Compared with the existing reliable controllers, different control structures are adopted in the proposed reliable controller, which can introduce more control gain matrices, and a method for introducing additional variable is also applied in the proof of the main result. Therefore, the conservatism of the existing results could be obviously reduced on account of the underlying measures. Furthermore, the obtained reliable control conditions are given in terms of linear matrix inequalities, which could be easily solved via standard numerical software. Finally, the effectiveness of the proposed approach is illustrated by means of numerical experiments.

In order to improve movement performance and control efficiency, and to save the cost of the control system of the three-DOF permanent magnet spherical motor, this paper studies a three-DOF permanent magnet spherical motor with three-dimensional position detection and its control method. The magnetic field model, the dynamic model, the torque model and the back-EMF model of permanent magnet spherical motor were established. The motion control mechanism was analyzed. Based on these models, the voltage balance equation and kinetic equation of the spherical motor were put forward and the control model was established. Applying the PID control method, the closed-loop control method of the spherical motor was obtained. Simulation results show that, compared with the current torque method, the control method proposed in this paper can greatly improve the movement performance.

In order to rapidly estimate extracellular polysaccharide content in submerged fermentation process of Morchella Esculenta, relativities between the fermentation parameters and extracellular polysaccharide were analyzed. There is little relativity between viscidity or conductance in fermentation broth and extracellular polysaccharide. At 0.01 level of significance, the relationships of extracellular polysaccharide and fermentation time, turbidity in fermentation broth, residual reducing sugar are of perfect association that all the correlation coefficients are over 0.7. After false associations among the parameters are removed, only partial correlation coefficient of turbidity in fermentation broth and extracellular polysaccharide is above 0.8. So, turbidity in fermentation broth can been considered as the optimized characteristic parameter for extracellular polysaccharide. The three mathematic models by linear fitting, binomial regression, BP neural network were tested by estimation experiment of the extracellular polysaccharide content. The results show that the estimation by BP neural network is the best that the variance is only 2.37×10^-6.

Pueraria starch of different concentration was added to myofibril protein to form mixed gels by high intensity pulsed electric fields and water bath treatment. Gelling properties were determined. The effects of the electric intensity, pulse number and mass fraction of pueraria starch were studied. It was demonstrated by single factor experiment and orthogonal analysis experiment, that mix solutions of myofibril protein (5%) and pueraria starch (5%) could acquire favorable mixed gels by the high intensity pulsed electric fields pretreatment and water bath of 80℃ for 30 min when electric intensity was 30 kV/cm and pulse number was 6. The influence sequence of factors on mixed gels was electric intensity, pulse number and mass fraction of pueraria starch. Study results indicate that the high intensity pulsed electric field pretreatment could promote myofibril protein and pueraria starch to form favorable mixed gels.

In order to explore an effective combination drying method for grains, maize was dehydrated by a combination of hot-air and microwave drying.In the first drying stage, maize was dried by hot-air of 60℃.When maize moisture content was reduced to 20%,microwave drying was applied until corn moisture content fell to 12%~14%.Determinations of moisture and temperature of maize during the combination drying process were carried out and related mathematical models were established.Compared to hot-air drying, electrical energy consumption for combination drying was reduced by 50.6%.Furthermore, mould numbers from maize dried by different methods were counted; results show that microwave radiation has a significant germicidal efficacy.