Abstract:This paper deals with the problems of estimating state variable for dynamic System (either linear or non-linear).If We limit our problems in linear estimating algorithm，then the minimum variance estimator is sufficiently optimum. But in general，the optimal estimator may not be the linear case. For this purpose，we discuss the approximations of the non-linear. filtering problems. we present the methods for two types of approximations，the first，Gram-Charlier approximation; and the second，the statistical second order approximation. In order to facilitate calculation，in this paper the approximate solution in recurrence form is given. The generalized Kalman filtering method and statistical lineanzation method are used only as the particular cases.

Abstract:After an analysis by means of expressing the supersymmetric charge in terms of particle's creation and destruction operators，we find the conditions，under which supersymmetry on the bound state of a Supersymmetric field-Theoretic Bag model represents finite dimensions. From these conditions，we get the rules of choosing wave functions for particles inside bound state. uhich will largely simplify the task of finding bound state of the model. Examples are given to show the new features of such notation in comparison with "off such" notation. The possibility of describing leptons and quarks by a Supersymmetric Field-Theoretic Bag model is also discussed.

Abstract:The external gravitational potential of the earth is shown by the spherical harmonics model. In general，the parameters of the model are γ、θ、λand the polar is the North Pole. When the model is app1ied directly to the calculation of the free f1ight trajectory，since the recurrent calculation can not be avoided，an increase in orders of model wil1 result in a decrease in the speed of calculation. In this paper，the transitional method of the spherical harmonics[2] is applied and a new model is gained. The parameters of the new model are γ(geocentric radius)，β(range)，ζ(cross-deviation)and the polar is decided on after choice-making. With some approximations the new model is now being applied in the calculation of trajectory. With the recurrent calculation got rid of，the speed of calculation is increased. Practical calculations have shown that when the orders of the model are 30，the application of the new model enables us to enhance the speed of calculation 6 or 7 times and the error will be no more than several meters.

Abstract:In this paper，two problems have been studied，i.e，(a) Under the condition of operate priori estimates，error coefficient of guidance tool of special Trajectory is separated by Bayes'method. Calculate formula and variance matrix of estiminates of error coefficient are also given. (b) On the basis of (a) the method of calculating guidance tool error for normal trajectory is introduced.

Abstract:An improved Barrar-type intermediate orbit is proposed in this paper and an analytic solution is reached. When the special C values are applied by means of different combinations，we can obtain the analytic solution due to the J₂ or J₃ oblateness perturbation for the long-range ballistic vehicle，as well as the solution of Kepler's ellipse. This method is worth to popularize，it's application to dozens of simulative trajectories have shown that this method not only helps to simplify the whole process of calculations but also results in smaller truncation error. the range error and lateral error for the J₂ perturbation are approximately 10m while those for the J₃ are about 10mm.

Abstract:The fatigue crack propagation rate at different points of aluminium alloy LD10cs butt-welded joints, tested by method of surface flaw has been discussed in this paper. The results show that the fatigue crack propagation rate at different points of the welded joints regularly varies with metal structure change resulting form the heatcycling of welding. The fatigue crack propagation rate of the fusion line and weld toe are greater than that of the others. The existence of the weld reinforcement will cause stress concentration at weld toe, and give worse infuence on the fatigue properties of this point, Hence, the fatigue properties of the weld toe will be the worst at welded joints. The various factors that affect the fatigue crack propagation rate at different points of the welded joints have been investigated and examined using optical metallography and scanning electronic microscope tochniques in detail. The suggestion to improve the quality of the welded joint has been presented.

Abstract:The method of parallel computation of Two-Dimensional Tensor Code has been developed in the paper [3].Some improvement has been made in the treatment of the points of boundaries，so that the acceleration both in and on the boundaries can be conputed with unique formula.

Abstract:The transient exciting test is one of the important methods to get the dynamic parameters of mechanical systems. Adopting the transient exciting methods，the dynamic exciting test has been carvied out for the main spindle system of machine tool Sl-235 with super-accuracy in this paper. The test technique the foundamental principle and the characteristic of signal processing of the transient exciting methods are presented here Final1y the natural friqnences，the damping ratio and the dynamic stiffness have been determined. The transfer function of system has been obtained by Levy's direct curve fitting methods.

Abstract:The paper describes basic principles of adaptive signal processing and its some new fast algorithms. Three examples of its application to radar systems are given: radar sidelobe canceller，high resolution direction finding and adaptive radar signal detection.

Abstract:This article is a comprehensive report about Free Electron Lasers (FEL). An argumentation on physical fundamental，progressing situations，and working features of FEL is presented，and many fields in which FEL can probably be used are mentioned as well.

Abstract:This paper adopts the free energy minimigation technigue for The calculation of theoretical rocket performance under equilibrium flow condition. According to the analysis of optimization theory，the direct search method is employed in combination with computer program. A solid propellant ingredient is obtained for maximum specific impulse and maximum characteristic velocity. The numerical results obtained agree fairly well with the results obtained by graphical method for calculating the performance from the literature [5].It is userful for determining the optimum composition of solid propellant systems.

Abstract:Microprocessor application development based upon an existing microcomputer can be effective and very economic. This paper takes hardware and software into consideration in implementing such a system.

Abstract:In this paper we consider a new kind of optimal stopping problems. Let {x_n,y_n,ζ_n}^∞_n=1 be an integrable and adapted stochastic process. We will find an optimal stopping rule for {x_n, ζ_n}^∞_n=1 in the class of stopping rule D such that for any t∈D，Ey_t≥V_y-a， where α is a constant and V_y is the value of {y_n, ζ_n}^∞_n=1 such that V_y＜∞.Adopting the Lagrange's method and the generalized Snell's method we obtain some existence theorems respectively，Finally，we discuss the two methods and apply them to solve the optimal stopping problem of the stochastic sequence of randon vectors.