JOURNAL OF BEIJING INSTITUTE OF TECHNOLOGY

Model Recognition of Astronauts' Blood Flow Change Under Sustained + Gz Acceleration 

1999, 8(3): 225-231.

[Abstract](301)

Abstract:
Aim Using animals as object of experiment to acquire various patterns of low cerebral blood pressure and reduced blood capacity in cerebral tissues of astronauts due to the load of acceleration. Methods The isotope tracking technique was applied to mark the blood and record the dynamic curves of cerebral blood flow changes under various accelerations, and the relevant mathematical model was set up using the method of system recognition. Also the method of factor analyzing was used to select two out of the data collected by eight sensors as two factors. Results One of the two factors reflects the various patterns in the astronaut's upper body, the other for the lower body. Parameters of rise time, delay time and steady value reflect the results under different acceleration. Conclusion Whether for the upper body or the lower body, blood flow changes can be considered as a second order system model. This method provides a new technique and method of doing research on astronaut's endurance of acceleration and selecting astronauts.

Measuring the Condition of Parking Lot by Image Processing

Wu Dayong, Wei Ping, Hou Chaozhen

1999, 8(3): 232-237.

[Abstract](167)

Abstract:
Aim To study the parking management in the condition of vehicles' increasing. Methods The methods of pattern recognition and image processing were used to analyze the eigenvalues of parking lot images. Results The automatic identification of every parking place in the parking plot was realized. The automatic measuring of parked vehicle count and parking lot utilization was completed. Conclusion It can complete the real time recognition, and has some practicabilities.

Identification and Control of Dynamical Systems Using Modified Neural Networks 

Ren Xuemei, Chen Jie

1999, 8(3): 238-244.

[Abstract](188) [PDF 231KB](2)

Abstract:
Aim To study the identification and control of nonlinear systems using neural networks. Methods A new type of neural network in which the dynamical error feedback is used to modify the inputs of the network was employed to reduce the inherent network approximation error. Results A new identification model constructed by the proposed network and stable filters was derived for continuous time nonlinear systems, and a stable adaptive control scheme based on the proposed networks was developed. Conclusion Theory and simulation results show that the modified neural network is feasible to control a class of nonlinear systems.

Incremental Multi Step R Learning 

Hu Guanghua, Wu Cangpu

1999, 8(3): 245-250.

[Abstract](315)

Abstract:
Aim To investigate the model free multi step average reward reinforcement learning algorithm. Methods By combining the R learning algorithms with the temporal difference learning (TD( λ ) learning) algorithms for average reward problems, a novel incremental algorithm, called R( λ ) learning, was proposed. Results and Conclusion The proposed algorithm is a natural extension of the Q( λ) learning, the multi step discounted reward reinforcement learning algorithm, to the average reward cases. Simulation results show that the R( λ ) learning with intermediate λ values makes significant performance improvement over the simple R learning.

Efficient Temporal Difference Learning with Adaptive λ

Bi Jinbo, Wu Cangpu

1999, 8(3): 251-257.

[Abstract](221)

Abstract:
Aim To find a more efficient learning method based on temporal difference learning for delayed reinforcement learning tasks. Methods A kind of Q learning algorithm based on truncated TD( λ ) with adaptive schemes of λ value selection addressed to absorbing Markov decision processes was presented and implemented on computers. Results and Conclusion Simulations on the shortest path searching problems show that using adaptive λ in the Q learning based on TTD( λ ) can speed up its convergence.

Solvability Condition for Robust Stabilization Problem of Control Systems with Parameter Uncertainties 

Wu Qinghe

1999, 8(3): 258-263.

[Abstract](218) [PDF 204KB](2)

Abstract:
Aim The solvability condition for robust stabilization problem associated with a plant family P(s,δ) having parameter uncertainty δ was considered. Methods Using Youla parameterization of the stabilizers this problem was transformed into a strong stabilization problem associated with a related plant family G (s, δ). Results A necessary solvability condition was established in terms of the parity interlacing property of each element in G(s,δ). Another apparently necessary solvability condition is that every element in P(s,δ) must be stabilizable. Conclusion The two conditions will be compared with each other and it will be shown that every element in G(s,δ) possesses parity interlacing property if P(s,δ) is stabilizable.

A Kind of Fuzzy Causal Diagnosis Method 

Wang Qinglin, Lu Dong, Li Ning, Chen Jindi

1999, 8(3): 264-269.

[Abstract](198)

Abstract:
Aim To improve the causal diagnosis method presented by Bandekar and propose a new method of finding the root fault order according to the fault possibility by means of numerical calculation. Methods Based on the causal graph, by utilization of fuzzified threshold value and fuzzy discrimination matrix, a kind of fuzzy causal diagnosis method was given and the fault possibility of each elements in the root fault candidate set (RFCS) was obtained. Results and Conclusion The order of each element in the RFCS can be obtained by the fault possibility, which makes the location of fault much easier. The diagnosis speed of this method is quite high, and by means of the fuzzified threshold value and fuzzy discrimination matrix, the result is more robust to noises and bad parameter's choice.

Predictive Control of Optoelectronic Tracking System

Zhang Hongye, Wei Wuzhong, Cai Yingzhen

1999, 8(3): 270-275.

[Abstract](190)

Abstract:
Aim To solve the time delay problem in the optoelectronic tracking system, improving the tracking accuracy. Methods The discount least square algorithm was applied to forecast the tracking error caused by the 40?ms delay, and the predicting algorithm was improved by the adaptive discount method.Results The tracking errors of the two methods were compared, and an optimal controller with the improved adaptive discount predicting algorithm was adopted for simulation. Conclusion The predicting algorithms, especially the adaptive discount predicting algorithm, can decrease the tracking error greatly, and the desired tracking prediction can be achieved both in the transient state and in the steady state.

Improving IAC Algorithm Based on Range Rate Measurement

Zhang Hongye, Wang Yueyong, Cai Yingzhen

1999, 8(3): 276-281.

[Abstract](210)

Abstract:
Aim Interactive multiple model(IMM) algorithm was introduced into two? stage estimation to improve the estimating accuracy for system position and velocity.Methods The state estimation was carried out in mixed coordinates according to the nonlinear measure equation, a generalized interactive acceleration compensation(IAC) algorithm in mixed coordinate was presented. Results Simulation result shows the estimation accuracy is improved through changing measure equation in polar coordinates. Conclusion The estimation accuracy for position and velocity estimation, has been improved greatly, and the proposed algorithm has the advantage of less calculating time comparing with other multiple model methods.

Self Adjusting Feedforward Compensation Tracking Control for Proportional Valve Controlled Motor

Peng Xiwei, Wang Yu, Wang Xiangzhou

1999, 8(3): 282-287.

[Abstract](224)

Abstract:
Aim Aiming at the position tracking control for valve controlled motor electrohydraulic proportional servo systems mainly driving the static load torque, the tracking performance was studied in the presence of the variable gain and deadzone. Methods On the basis of conventional composite control with the deadzone compensation method, a comprehensive control approach with the deadzone and self adjusting feedforward compensation was proposed. Results Experimental results showed that the good tracking performance was achieved for the sinusoidal and constant velocity position tracking under a wide variations of load torque. Conclusion The position tracking accuracy for valve controlled motor electrohydraulic proportional servo systems has been solved by using the comprehensive control approach with the deadzone and self adjusting feedforward compensation.

Precision Point to Point Control of Proportional Valve Controlled Motor with a Time Varying Load

Peng Xiwei, Wang Yu, Wang Xiangzhou

1999, 8(3): 288-293.

[Abstract](267)

Abstract:
Aim In accordance with the positioning control for valve controlled motor electrohydraulic proportional servo systems driving the static load torque, the positioning performance was studied in the presence of the time? varying deadzone and gain. Methods The large positioning errors caused by the time varying deadzone were significantly reduced by using the dynamic compensation method for the deadzone; and the large overshoot caused by the time varying gain were dramatically reduced by using the three section intelligent control schemes. Results Experimental results demonstrated that the positioning performance of rapid response, high accuracy and smaller or even no overshoot was achieved under a wide variations of load torque. Conclusion The good positioning performance for valve controlled motor servo systems has been achieved in the presence of the time varying deadzone and gain.

Design and Research of Information Processing System Based on Mix Platform

Dong Ning, Xu Lixin, He Chao, Zhang Yuhe

1999, 8(3): 294-299.

[Abstract](166)

Abstract:
Aim To develop an information processing system with real time processing capability and artistic user interface for the optoelectronic antagonism general measuring system. Methods The A/D board and the multifunctional board communicating with every instruments were designed, data collecting and processing were realized by selecting appropriate software platform. Results Simulating results show the information processing system can operate correctly and dependably, the measuring rules, interactive interface and data handling method were all accepted by the user. Conclusion The designing approach based on the mix platform takes advantages of the two operating systems, the desired performances are acquired both in the real time processing and with the friendly artistic user interface.

Backlash Nonlinear Compensation of Servo Systems Using Backpropagation Neural Networks

He Chao, Xu Lixin, Zhang Yuhe

1999, 8(3): 300-305.

[Abstract](202)

Abstract:
Aim To eliminate the influences of backlash nonlinear characteristics generally existing in servo systems, a nonlinear compensation method using backpropagation neural networks(BPNN) is presented. Methods Based on some weapon tracking servo system, a three layer BPNN was used to off line identify the backlash characteristics, then a nonlinear compensator was designed according to the identification results. Results The simulation results show that the method can effectively get rid of the sustained oscillation(limit cycle) of the system caused by the backlash characteristics, and can improve the system accuracy. Conclusion The method is effective on sloving the problems produced by the backlash characteristics in servo systems, and it can be easily accomplished in engineering.

Equilibrium Control Theory and Direct Method of Feedback Linearization for Nonlinear Time Varying Systems 

Wang Qinglin, Chen Yue

1999, 8(3): 306-311.

[Abstract](240) [PDF 192KB](4)

Abstract:
Aim To present a simple and effective method for the design of nonlinear and time varying control system. Methods A new concept of dynamic equilibrium of a system and its stability were presented first. It was pointed out that what is controlled directly by the input of a control system is the system's dynamic equilibrium rather than the states. Based on it, a new feedback linearization method for nonlinear system based on the Lyapunov direct method was given. Simulation studies were also carried out. Results The example and simulation show that by use of the method, the controller design becomes very simple and the control effect is quite satisfying. Conclusion The new method unifies the stabilizing problem(regulating problem) with the tracking problem. It is a very simple and effective method for the design of nonlinear and time varying control system.

Neural Network for Prediction of Thermal Elastomer Quality

Liu Nianquan, Wu Cangpu, Liu Qing

1999, 8(3): 312-317.

[Abstract](186)

Abstract:
Aim To predict the indexes of quality of the thermal elastomer by polymerization process data. Methods Neural networks were used for learning the relationship between the product quality and the polymerization process condition variables in an industrial scale batch polymerization reactor. Results The indexes of quality of the product were inferred with acceptable accuracy from easy to measure reaction process condition variables. Conclusion The method proposed in this paper provides on line soft sensors of the indexes of quality of the thermal elastomal.

Adaptive Active Control of Low Frequency Noise in a Duct 

Li Chuanguang, Li Wu, Han Xiuling

1999, 8(3): 318-323.

[Abstract](197)

Abstract:
Aim To study a method realizing noise control for a physical model of progressive wave in a duct. Methods A mathematical model was constructed and a transfer function of the adaptive system for noise control was also worked out; moreover, the effects of some algorithms such as RLS,LMS and LSL on noise control were analyzed and compared. Results Without the feedback of sound, the mean noise reduce value(MNRV) of 27 5 dB for broadband noise from 0 to 500?Hz in frequency were achieved. When acoustic feedback took place and an air stream loudspeaker was used, the MNRV was only about 4 9?dB. But if the loudspeaker had a plain frequency feature, MNRV was improved by 10 2?dB. Conclusion The technique is applied to ruducing the noise from engines' exhausted gas pipes. It is, in principle, used for noise cancelling in a closed three dimensional space.

Reliability Analysis of a Mixed Redundant Model in C³I System 

Xu Yong, Hou Chaozhen

1999, 8(3): 324-330.

[Abstract](211)

Abstract:
Aim To define a mixed redundant model(MRM), improving the reliability of C³I system. Methods The model combined the technology characters of two? unit system with one warm stand by unit and function substitute system. The reliability and availability equations of MRM were deduced. Results and Conclusion Compared with several other reliability models, it has obvious effect upon improving the system reliability. The effect? cost rate is very high among these models. The model can be used in reliability design, evaluation and check of C 3I system. Only a little attached cost is needed to improve C³I system reliability effectively.

Error Analysis of a Dynamically Tuned Gyro Strapdown Northfinder 

Qi Yutong, Chen Fengyu, Su Haibin

1999, 8(3): 331-336.

[Abstract](499)

Abstract:
Aim To analyze the mathematical error model of a dynamically tuned gyro (DTG) strapdown northfinder in detail, guide the process of design, manufacture and adjustment of northfinder. Methods Each error source of this type of northfinder was determined, and the influence of each source on northfinding result was formulated. Results and Conclusion Under the guidance of the analysis, select relevant method for each source which has different effect on result to reduce northfinding error, a type of northfinder meeting the practical requirements of user was developed.

1999 Vol. 8, No. 3