The coordinate transform was used firstly to establish a mathematical model for the vehicle ride comfort system. The input signals of the model were transformed into the acceleration signals of the front axle and the rear driving axle. Compared with the traditional vehicle ride comfort model, the calculation accuracy of the proposed model was improved and appeared more practical in engineering because of using the acceleration signal as input signal. The complicated tyres were separated by the equivalent transform which made the model more available. A vehicle ride comfort simulation model was established by means of software package matlab/simulink, and its complicated input parameters were discussed considering the stiffness of the flat spring and the nonlinearity of the damping. In order to validate the simulation model, the corresponding vehicle tests were performed. The comparison between the test data and the simulation curves showed that their agreement was quite good.

The differential Reynolds stress turbulence model(DRSM) with the second order moment closure was used to numerically simulate the external flow field around the Ahmed standard car model, and the aerodynamic drag coefficient and the velocity vector distributions in the 3 sections of the wake were obtained by comparison between the simulation result and the corresponding wind tunnel experimental data. The differences among the drag coeffcients from the RNG kε model simulation, the DRSM simulation and the experiment were discussed, which revealed that the DRSM model is characterized by higher computation accuracy than RNG kε model, can provide correctly the vehicle aerodynamic characteristics including the second flow in the vehicle wake.

An improved federated Kalman filter configuration was designed based on the analysis of the features of GPS/DR integrated positioning system. A conventional Kalman filter is used to process the information from the linear GPS subsystem, while an unscented Kalman filter is used to process the information from the nonlinear DR subsystem. The main filter makes the optimal estimation of the results from the subsystems using the dynamic informationsharing coefficient, and compensates the error of DR subsystem to reduce the accumulative error of DR. The test results showed that the proposed algorithm meets the positioning request of the onboard navigation system and keeps a good reliability and a high positioning accuracy when the GPS positioning is poor.

An integrated model based on the neural network and Markov chain was established to predict the real estate price along the urban rail transit under the condition of limited a priori data and a host of uncertainties. A BP neural network was used to fit roughly the real estate price curve along the urban rail transit based on a little sample data. Markov chain was applied to achieve the state transition probability matrix of the system to modify the predieted result. The model takes the randomness of the real estate price into account, and it stll available regardless of that the statistics of the real estate prices are scarce. The real estate price along the rail transit No.13 in Beijing was taken as an example to illustrate the application of the model. The calculation results showed that the model is characterized by good accuracy and reliability of the predicition.

In cosideration of cross contry environment that is often variable in illumination and complicated in ciecumstance, a crosscountry environmental perception technique on the basis of stereo vision system was presented. The ooriginal images were preprocessed by Gauss filter and CLAHE method and Harris corners were located with subpixel accuracy. Fundamental matrix was calculated using RANSAC theory. And epipolar restrain and continuity restrain were deeply studued. Finally, 3D visulization techniques confirmed the effectiveness and feasibility of the proposed method.

Waste rubber powder and thermoplastic elastomer SBS were used to modify heavy highway asphalt AH90. The physical properties of the modified asphalt were measured and its microstructure was analyzed. Results show that, after modification, the penetration obviously decreases; the ductility and elastic recovery significantly increase; the softening point and the equivalent softening point T800 increase while the equivalent brittle point T1, 2 decreases. The stability of the modified asphalt at high temperature and its deformability at low temperature are both greatly improved.

There are some limitations in studying the low temperature properties of the asphalt by means of the bending beam rheometer(BBR) and the dynamic shear rheometer(DSR), so the samples of asphalt were tested by DSR at normal temperature and the obtained dynamic rheological properties were converted to the viscoelastic properties at low temperature using the timetemperature transformation principle and the transformation relation between the static and dynamic viscoelastic functions. The results obtained by such a indirect way were compared with the low temperature properties obtained directly by the BBR, and their interrelationship was analyzed. The results showed that the DSR data at normal temperature can be used to predict the low temperature behavior of the asphalt. Therefore, it is possible to get the viscoelastic properties of the asphalt over a much larger time interval than those measured at low temperature.

Using finite element analysis softerware ANSYS, the transient thermal and electrical coupling analysis is carried out on the temperature field inside the chamber in which diamond is synthesized using flaky catalyst. The influence of the anvil's temperature on the synthetic effect of diamond is taken into consideration, and a load step document is created to realize the anvil's temperature loading. The timehistory graph of the temperature in the centre of the chamber illumates that the higher the initial anvil's temperature is, the quicker the chamber temperature reaches thermal equilibrium state. This relationship directly affects the diamond's nucleation and growth. Numerical nodal temperature indicates the tendency that, inside the synthetic chamber, the centre temperature is higher than elsewhere; the axial temperature difference is about 80 ℃, and radial temperature difference is about 50 ℃. Numerical analysis of unit heat generation rate and thermal flux density is conducted. Results show that the chamber's boundary extermal thermal conduction is the primary factor that leads to temperature ununiformity distrubtion inside the chamber.

There are many factors involved in the quality of the gasassisted product. By the analysis of the special features of the gasassisted injection molding process, with the help of the orthogonal design of experiment, a fuzzy multiobject optimal decisionmaking model was built based on the theory of fuzzy mathematics and the programming method. A case study showed that the built model is effective to select quickly and correctly the optimized gasassisted.

There arise usually the serious stress concentration around the hole in the composite plate with circular holes under the external loads. A calculation model of stress field of the composite plate with circular holes was built according to the theory of complex variable function. The complex stress function were derived by the affine transformation to describe accurately the stress field. The stress analysis of the composite plate with circular holes were performed by the model simulations, revealing the effects of the fiber orientation on the stress concentration and the Young's module in the two main directions on the holeedge stresses.
Key words：

According to the forming principal of the spherical surface，both the inner and outer spherical surface of the spherical hinge were formed using the circular motion of the work piece and the milling cutter. The errors of the size, the shape and the position were analyzed and the measares to reduce them were proposed. A combined machine tool configuration was used lohich is charaterized by simple structure as well as easy to adjust and maintain. An elastic clamping head was fitted on the spindle which makes the work piece by installed and removed conveniently. The cutting parameters were optimized through the speed control of the cutter motor and the spindle motor by the power supply frequency converters and the highefficiency and highaccuracy machining of the spherical surface was achieved.

Based on piezoelectric precision drive technology, a hydropump using a piezoelectric stack as actuator and a flexible hinge as magnifieation mechanism was prototyped and tested. The hysteresis behavior of Tokin AE0505 D16 type piezoelectric stack was analyzed and its stiffness characteristic and dynamic response charateristic were studied theoretically. A flexible hinge mechanism to magnify the displacement was prototyped and the factor affecting its torsional deformation and torsional stiffness were analyzed, and its structual parameters were proposed. The dynamic response, the amplitudefrequency characteristic, the hysteresis behavior, and the force output versus displacement of the flexible hinge were tested under different power supply excitations, and the factor affecting the displacement magnification ratio and the output charateristics of the magnification mechanism were found. The flexible himge was also analyzed by a FEM simulation software and its feasibility and security were validated. The developed piezoelectric stack pump was tested under different input signal voltages and frequencies and the effects of the input on the output flow rate and pressure of the pump were obtained.

Combining the pault diagnosis technique and the embedded technology, a protable fault diagnosis instrument was developed for the miniature application. Its hardware configuration, dual CPU in special, and software structure, including the communication software, were described. The developed portable fault diagnosis instrument based on the dual CPU incorporates many functions, including the data acquisition, storage, and analysis as well as communication in a unit with low cost, can be widely used in the system monitor and diagnosis in various equipments and installations. The experiment results showed that its function is stable.

The nonlinearity error of the dualfrequency laser interferometer with two longitudinal modes and special subsequent electrical signal processing system was studied. The expressions for the output signal pair were desived based on the theoretical analysis when the nonlinearity exised. From the above it was given that the nonlinearity makes the frequencymixing output signals shift and undulate, and its main factors are the perfection degree of the optical system adjustment and the polarization splitting degree of the polanzation beam splitter. The maximum signal shift amplitude was close to that observed on oscillograph and it could be used to estimate the perfection degree of the optical system adjustment and determine the maximum interpolation number in the subsequent signal processing stage.

To get rid of the weaknesses of the traditional defect detection of the microgear, such as low accuracy, inconvenient online implementation, and liable to manmade interference, a new detection technique for microgear defect based on the image processing technology using the charge coupled device(CCD) as the image sensor and the support vector machine(SVM) was proposed. The brilliant back illumination was provided by a light emitling diode(LED) illuminating source, the digital image of the tested mierogear was collected by an A102FCCD digital camera and input into a computer by an image acquisition card. The original gray level digital image with noise was processed by the edge retention filter to reduce its noise, and segmented by the iteration threshold method and the Otsu dual threshold method, and the binary image was formed. The samples of the gear were collected and the sample fatures were extracted. A gear defect recognition model was built based on the SVM. The recognition accuracy is as high as 97.8%. The proposed technique is characterized by low cost, high relialility, good popularization, and easy online implementation.

A novel gainswitching force feedback algorithm was proposed based on the common force reflect algorithm to make the masterend force truly reflect the force acting on the end actuator forkglove of the slave robot in the bidirectional force feedback servosystem of the masterslave system of the teleoperation engineering robot. A mathematical model for the relationship between the displacement and the driving force of the robot forkglove cylinder was built based on the kinematics of the forkglove to adjust the force feedback control gain of the masterend. The proposed gainswitching algorithm was compared experimentally with the common force refleet algorithm. It was proved that the novel algorithm accurately reflects the effects of the inertia force and the damp force of the robot forkglove during its motion process under the prerequisite for system stability, the operator can sense more truly the variation of the force acting on the staveend of the robot, which helps to improve the working capability of the teleoperation system.

In the bilateral hydraulic servo control system of a construction telerobot with insitu force sensing, the pf type force feed back architecture is liable to result in an impact on the operator hand, and its high amplitude will cause the control unstable. So, based on the TS type fuzzy reasoning logic, an improved force feedback control strategy with the feedback scaling which could be modified online nonlinearly and continuously was proposed. A RBFPID force controller with adjustable parameters was designed and a new bilateral hydraulie servo control system was developed. Its experimental results showed that the proposed force feedback control strategy improves the tracking precision of the slavehand to the masterhand, makes the masterhand follow well the force acting on the slavehand, reduces the impact of the feedback force, enhances the compliance and transparency of the teleoperation of the construction telerobot system.

The manipulation smoothness, the hydraulic pressure shock, the thermal loading, the unloading oil pressure, and the dynamic response of an electrohydraulic control system with the poppet valves for a tractor were studied indetail. The concept of the dynamic pressure feedback was proposed to solve the strong hydraulic pressure shock in the system and tractor pitch oscillation. The test of the new electrohydranlic control system led to following results: compared with the original control system, the tractor pitch speed caused by the lift of the farm implement reduced from 0.3 to 0.08 rad/s, the peak pressure shock in the hydraulic system decreased from 4.0 to 2.8 MPa, the unloading pressure loss in the integrated control valve block was 0.12 MPa, the hydraulic oil temperature at thermal balance was 38 ℃, the switching frequency of the control system was 5.5 Hz.

Taking Landakin stoliczkana as the reference animal, ternary bionic coupling samples were designed according to the orthogonal design of experiment involving the morphology, the surface structure, and the material as test pactors. The samples were made on the substrate of the hard aluminum alloy LY12 or the steeel 45# with wearresistant coating. The erosive wear resistances of the samples were investigated by the sand blast test using the Al2O3 sand with partical size of 1000 μm as the erodent. The results indicated that the erosive wear resistance of the coupling sample with pitting structure is 18.7% higher than that of the sample without pits under the following test conditions: erosive time duration 180 s，sand impact angle 30°， impact distance 200 mm， blast air pressure 0.4 MPa. The optimal combination of the coupling sample characteristic factors are:hard aluminium alloy LY12 as the substrate, the morphology of the nonsmooth units is the regularly distributed rotundity pit with diameter 3 mm and spacing 6 mm, the coating is AL2O3+13%TiO2 with thickness of 100 μm.

An arbitrary LagrangeEuler(ALE) finite element method was introduced to build a mathematical model to trace the sloshing of the lubricating oil in the micropits on the piston surface to help understanding the oil storage mechanism of the bionic nonsmooth surface. The free surface profile and the pressure cloud diagram of the lubricating oil sloshing were simulated by the nonlinear finite element solver LSDYNA, demonstrating intuitively the whole dynamic process of the lubricating oil sloshing in the nonsmooth micropit. The lubricating capability of the hemisphere, taper, and cuniform oil storage micropits were simulated compuratively and the result showed that the lubricating capaqility of the cuniform micropit is the best, being in agreement with the experiment.

To improve the reliability and efficiency of the design of stubble crushing blade roller, the dynamic gap angle was visualized by drawing the trajectory of the blade cuttingedge using UG-NX3 software. New dynamic gap angle can be obtained by changing the kinematic parameters of the stubble chopper. 3D models of stubble crushing blade and blade disc were built to obtain the dynamic relationships among related dimensions. The 3D models were assembled and their interpositions were checked. Kinematic simulation analysis was conducted and results show that the original design requirement was satisfied. This provide a new design and analysis method for stubble roller design.

The gas sensor arrays were composed of different separate sensors. The separate sensors composing the sensor array were tested in order to form an optimal sensor array. The distilled spirit(Erguoton brand) produced by Beijing Jinggong Wine Insdustry Technology Development Corporation was used as experimental reference specimen. The distillated water was blended into the spirits in the ratios of 1∶3, 1∶2 and 1∶1 respectively to prepare the working specimens. The working specimens were used to test the performance of the sensor array and the results were compared with that of the reference specimen. An artificial neural network was utilizid to analyze the collected data and estimate the performance of the sensor arrays. It was proved that the array performance can be considerably by changing the sensor.

Wireless sensor network technology can provide optimal and integrated solution to distributed data collection, delivery and analysis in farmland. An infield soil moisture and temperature monitoring system was developed which meets the application requirement in farmland environment. This system consists of the soil monitoring wireless sensor network and remote data center. In the wireless sensor network, the sensor node is developed using JN5121 module, an IEEE 802.15.4/ZigBee wireless microcontroller. The sink nodes for aggregating and dilevering network data is based on ARM9 processor platform in order to meet the requirements of highperformance. A GPRS module is integrated into the sink node for long distance communication. In the remote data center, the management software running on the host computer is developed for realtime data receiving and logging based on database management method. It also uses ArcEngine, an embedded GIS developer kit to realize online spatial analysis of infield data. This monitoring system may provide an effective research tool for spatial analysis and for irrigation decision making in precision algriculture.

The stiffness of a griddle body was enhanced using dynamic topological optimization under multiple frequencies constraints. In order to avoid the tedium solution of mapping and ill loading cases, the element (or element group) sensitivity of eigenvalue was presented. The nondimensional measure used for identification of the element (or element group) under multiple frequencies constraints to be changed is given. The optimal procedure is described. In the optimal process, available stiffeners were added to the structure as to form the base structure, and the added elements are divided into groups base on the stiffeners. The present procedure was implemented for the griddle body, and effective results were obtained.

Based on stochastic finite element method and reliability theory, a reliability analysis model for structures with multicrack was presented. Firstly, the expression of stress intensity factors was proposed using the least squares method. Secondly, uncertainties in material parameters, crack length and loading which affect fracture of structures were described as random variables, the reliability index for each crack was obtained by first order reliability method and stochastic finite element method. Finally, the structure with multicracks was taken as a series system composed of all cracks, and the correlation coefficients between failure models were evaluated. The failure probability of the structure was computed by Ditlevsen bounds method. A numerical example was performed for a finite plate with multiple holes and cracks. The result shows that the suggested method is feasible and applicable.

A multigrid method with a simple adaptive time stepping technique was proposed to solve the structure dynamic response in the finite element analysis(FEA). By the method the results of the initial mesh were utilized sufficienlty, the new approximate displacement vectors in the changed mesh were obtained by the bilinear interpolation technique, and the solution for the structure dynamic response problem was accomplished by the multigrid iteration procedure. The conjugate gradient method was applied in the prior smoothing process of the multigrid iteration procedure to improve the convergence rate. An equation to estimate approximately the error of the displacement vector was given in the correction of the coarse mesk. In the postsmoothing process of the multigrid iteration procedure, Jacobi relaxation iterations were chosen to improve the evenness of the solution. Combining the discretization of mesh with the numerical solution, a fast reanalysis procedure after the mesh refinement was established. Compared with the traditional FEA, the proposed method is characterized by simple calculation, fast iteration and timesavingness. It can be used as an effective tool for solving the structure dynamic response problem by the adaptive FEA.

In order to compare the influence of weight function on computational precision,the quartic spline function and exponential function were chosen as the weight function.Two interpolation functions were used as field functions and Meshless dynamic equations were built.The penalty method was utilized to enforce essential boundary conditions and frequency and modal vector were computed.Two Meshless results of free vibration of beam and their different radius were compared with analytical results.The numerical example of beam structrue validated the method are reliable.

Because the planesection assumption is not applicable to tendon in unbonded prestressed concrete beams and slabs under loading, the constant stiffness method and momentcurvature nonlinear analysis method were used to compile the calculation programs of the law with serviceability and ultimate carrying capacity. Based on results of the model experiment and a large number of simulations analysis, the law of some basic factors such as nonprestressed reinforcement index βs, prestressing reinforcement index βp, ratio of span to depth l/h, loading type, distribution of prestressed reinforcements and depth from concrete extreme fiber to centroid of the prestressing steels hp affecting stress increment of tendons in simple beams and slabs were presented, and then calculation formulus of stress increment of tendons at serviceability and ultimate carrying capacity state in unbonded prestressed concrete beams and slabs were proposed.

A truss concrete composite freesupported beam prestressed encased with 4 circular steel tubes was tested and its curve of loaddeflection at mid span, the bending loading capacity, the distribution and development of the cracks were obtained. The results showed that the pure bending region of the beam accords with the plane section assumption from loading to failure, and the average crack spacing depends on the external diameter of the bottom chord tube and the ratio of the sum of the section areas of the two bottom chord tubes to the effective tensile concrete area. The formulas for the bending loading capacity, the stiffness, and the crack width which are in good agreement with the test results of the beam were presented.

The calculating method of strengthened CFRP area was deduced, and the optimization model of strengthened CFRP area for RC beams was built and the confinement conditions was determined. The optimized CFRP area and the corresponding ultimate bearing force was solved with the computational program compiled by optimum toolbox in MATLAB. 4 CFRP strengthened RC beams were fabricated and the theoretical calculated value has been proved by the measured results of the flexural tests.

On the basis of previous studies by other researchers, elasticplastic finite element method is employed to investigate the complicated tunneling problem. Using selfcoded program, the uniform finite element model is developed to study the characteristic of soiltunnelling interaction. The influences of forcecontolled finite element method and displacementcontrolled finite element method on soil free displacement field are compared. Results show that forcecontrolled method is better in simulating the enginnering process, while the displacementcontrolled method is simply available and the results are closer to real measurement. Both methods can be used in engineering projects.

The hydraulic ash transportation gaius increasingly wide application in thermal power plant. On the basis of the fundamental equations of the sturry waterhammer transient process, a mathematical model for the process of the pseudohomogeneous airentrained slurry waterhanvmer at the pump stopping was proposed using the characteristic line method. The transcent processes of the relative water pump outlet pressure, the dimensionless pump flow rate, the dimensionless pump rotational speed, the relative pressure head at the highest point of the pipeline were calculated by the model for the compound ash transportation pipeline in a certain thermal power plant. Based on the numerical calculation and the analysis of the calculated results, a solution to alleviate the harm of the slurry waterhammer in the pipeline was proposed, i.e.,installing a surge tank at the highest point of the pipeline and increasing the moment of inertia of the water pump motor.

Focused on the structure and parameter identificantions in TS fuzzy system modelling, a hybrid clustering method SOMK, which combines the SelfOrganizing Map (SOM) and Kmeans approaches, is used to cluster the input samples, and the DaviesBouldin (DB) validity criterion is used to determine the optimal number of clusters, which is taken as the optimal number of rules of the TS fuzzy system. Then the premise parameters are determined by the number of rules and the cluster centers. Furthermore, a least square algorithm is applied to obtain the consequent part of each rule. A case simulation is used to illustrate the proposed medolling method and the algorithm. Simulation results prove that the method has high modelling precision and high ability of power generalization, and it is valuable for practical application.

Higheffect coagulating pilotscale experiment on Hun River water source, contaminated by industrial sewage, was carried out in low temperature and low tuibidity period. The optimum dosage was determined as: Polyaluminum ferric chloride (PAFC) 30 mg/L and Polyacrylamide (PAM) 0.2 mg/L. Furthermore the pilotscale experiment results show that in the loading range from 2 to 3 m3/h, the processed water can reach the standard for industry utilization. It is also shown that increasing the water temperature can enhance the effect of coagulation.

The existing notions and techniques of consistency algorithm were further investigated. Then a new algorithm——BiSAC-2 was proposed to enforce the bidirectional singleton arc consistency in constraint network. The soundness and completeness of the algorithm were proved. BiSAC-2 has the same ability to simplify the problem as that of the conventional BiSAC-1. However the number for maintaining the arc consistency is reduced and the fixpoint can be reached much quickly, hence the efficiency can be improved. Experiments on random CSPs show that the new algorithm, BiSAC-2, needs much less time compared with the conventional algorithm BiSAC-1.

Based on facetbased approach, description logics were introduced as the formal method of knowledge representation. Description logics can be used to represent users' reuse requirements and components, so that they can match on the semantic level. A component retrieval and matching algorithm based on description logics was proposed. The algorithm calculates a real number to denote the matching degree of reuse requirement and any component. Thus the matching degree can help the user to find an appropriate component from the component library.

To increase the running speed of the collision detection algorithm, a new approach is proposed, which uses convex Bounding Volume Hierarchies (BVHs) to express convex polyhedron. Thus the distance problem of two convex polyhedrons is come down to a nonlinear programming problem with constraints. The nonlinear programming problem can be solved by simulated annealing genetic algorithm. Applying the acceptance criterion of the simulated annealing genetic algorithm, intersection and variation are carried out to optimize the time complexity. Results show that the efficiency of the simulated annealing genetic algorithm is high and the computing speed is faster.

Structural test is one of the basic approaches for test case identification. Because of the complexity of programming languages and the variety of applications, an efficient approach for automatic generation of structural test case is by iteration which is guided by program execution results. To examine the feasibility of full path coverage test, a R_N(K) path coverage criterion is presented. First, the test is treated as multiobjective optimization. To enhance the efficiency improvements are made on the encoding, the fittness function and the annealing gene of the genetic annealing algorithm. Then the optimization results and instumentation theory are used to generate test data. Experimental results show that this approach is better than the random method and the genetic method.

An approach of mapping objectoriented design changes to program changes are proposed, so that the prediction of possible source code change effort can be made. This approach focus on the definition and taxonomy of atomic changes and composite changes at both level, according to which change impact analysis algorithms are proposed and applied to conclude the change mapping rules from UML based design model to Java source code. A supporting tool implementing the heuristics is designed and a case study on an open source project is conducted to validate our method.

Robustness and smallworld property of scalefree network are the desirable features in worm network. Taking advantage of worm propagation properties, this paper proposed three worm propagation models which are employed to build scalefree worm network. In these models, a compromised host is connected to the worm network by linking with suitable worm nodes. The nodes are chosen according to the node properties, such as the degree and fitness of the node. Theoretical analysis and simulation were conducted and the results demonstrate that the worm networks generated by the proposed models are scalefree.

Using a three dimensional Markov chain, a nonsaturation throughput performance model was proposed for IEEE 802.11 Distributed Coordination Function (DCF) in fading channel. The impacts of different factors, including bit error ratio, traffic loads and packet length, on the throughput of the system were analyzed using this model. Based on the model, a simple formula was derived to calculate the optimized packet length for maximizing the nonsaturation throughput. Simulation results show that this model can effectively predict the nonsaturation throughput of 802.11 DCF in fading channel, and by employing the optimum packet length the system throughput can be enhanced.

According to the definition of 4D matrix DCT, the transfer core matrix can be considered as twodimensional unitary matrix. Float 4DMDCT inversible factorization was conducted and integer to integer inversible transform metrix was obtained. Then this implementation of 4DMDCT was applied to video sequnces and the integer trasform coefficients were quantized by the vector quantizer based on robust statistics. Experiment results indicate that the PSNR of the reconstructed image through this scheme improves more than 1dB than that of the reconstructed image of float scheme and the subject quality of the image is also improved.

An adaptive interleaving model for MIMO-OFDM systems was proposed. This model can better exploit instantaneous channel variation through rearranging transmitted symbols according to quasi-instantaneous channel SNR of the OFDM frame, which was to break up long burst channel error so that overall bit error probability could be reduced. An upper bound of pairwise error probability (PEP) of the systems was investigated. The code gain and diversity gain of the proposed arithmetic were given. Theoretical results show that the major factor affecting the performance of the proposed scheme is SNR, channel order and the number of receive attenna. Simulations indicate that significant SNR can be obtained and the system performance is better than that of single antenna case. Compared with block interleaving in the case of 2Tx2Rx at a 10-2 BER, this adaptive interleaving scheme offers 5dB increase.

An adaptive allocation algorithm on the basis of Genetic strategy for subcarrier and power to maximize the sum capacity of multiuser orthogonal frequency division multiplexing (MU-OFDM) systems in a downlink transmission while maintaining proportional fairness is developed. It is a lowcomplexity suboptimal algorithm which separates subchannel allocation and power allocation. Subchannel allocation is first performed under the condition of an equal power distribution. A genetic strategy then fulfills power allocation among users while maintaining proportional fairness. After the power allocation, each user transmit power is allocated equally to maximize the sum capacity. Simulation results show that the proposed algorithm can offer balance and tradeoff between sum capacity and fairness. Furthermore, it can also satisfy differently required data rate compared to the maxmin problem method and the sum capacity is distributed more fairly among users than the sum capacity maximization method.

Considering that traditional UWB-IR system repeats and transmits very short pulses for each bit, an internally adaptive LDPC coded UWB-IR system to improve the repetition code was proposed. In the system LDPC code rate and the number of the repetition code are adaptively chosen. Simulation and performance analysis were conducted under AWGN channel. It is shown that the performance of the proposed system is greatly improved compared with that of the traditional UWB-IR system and at high code rate it is also better than that of Turbo coded UWB-IR system.

With the goal of minimizing the overall transmit power while satisfying the required data rate, an adaptive algorithm combined exclusive manner with shared manner was proposed for multiuser MIMOOFDM systems in downlink scenario. Subcarriers allocation to users was conducted according to user spatial correlations in this algorithm. The users swith high spatial correlations were palced in same group, and on the basis of rule of propotion fairness, subcarriers were allocated in an exclusive manner in the same group. While in different groups, subcarriers were allocated in the shared manner using grouping zero forcing method. Then greedy method was used to allocate bits and power to different users. Simulation results show that the overall power of the proposed algorithm is less than both the exclusive manner and the shared manner, the spectrum efficiency is improved compared to the exclusive manner, and the number of antennas is less than that of the shared manner as well.

The traditional power transformer fault diagnosis method was bettered by an improved gray clustering algorithm based on the theory of gray system. A concept of influence gene was introduced in the optimization of the power transformer fault standard model. Two ends of the trapezium whiting function were selected to analyze and judge the fault pattern and location of the transformer by the trapezium gray clustering algorithm. The results of the proposed diagnosis method were in good agreement with those of actual hanging hood detections, illustrating the new method is effeitive and practical, and characterized by the quickness and the realtime possibility.

Through the solution of inverse kinematics, a classified neural network was trained by genetic arithmetic combined with LevenbergMarquardt method to recognize the relationship between joint space and Cartesian space of parallel robot and the evaluation of forward kinematics(FKs) in a range with a little error was provided. Using QuasiNewton iteration method, accurate solution can be obtained. Applied the proposed solution in 6PRRS parallel robot, the result shows that this method can get high precision value and spends little time. Therefore it can be applied in real time task space control or in getting work space of parallel robot.

For benchmark translational oscillator with rotational actuator system, an output feedback hybrid dynamical impulse global regulator with bounded inputs was presented which was based on energy shaping plus damping injection methodology of EulerLagrangian system and recent result on invariant set stability theorems for nonlinear impulse hybrid systems. And from the perspective of energy dissipation, the effectiveness of the proposed regulator was studied. Experiment results for the benchmark system show that system performance under the proposed strategy is better than classical continuous passivity based controller given by Ortega et al.

A Halo orbit of lagrange liberation point was designed with different correction method and Richardson threeorder approximation solution as initial conditions. Then on the basis of invariant manifold theory and Poincaré section, a transfer trajetory between L1 and L2 of the EarthMoon system was given. Simulation shows that a low energy trasfer trajetory can be achieved with this prposed method.

In order to reduce response time of weapon systems and increase speed of information processing in mission planning system, a recursive route planning algorithm in the point of view of engineering application and on the idea of back reasoning from target position to warship position iteself was presented. Under some given condition, on the rule of the least number of navigation points which could complete the desired attack angle aviation, this algorithm can deduce next navigation point from target to the launcher in adverse direction of attack angle and get the planning route. Based on the planned route and in consideration of threats existed in the battle area, a tangetial route planning algorithm was proposed. According to the shortest planned route rule and the least number of adjusting planned routes, and by adding new navigation point or modifying old navigation point, this algorithm can adjust the planned old route to the shortest tangent of the threat area to avoid enemy. Simulation results show the validity and efectriveness of the proposed algorithms.