2012 4 30 2 Journal of Northwestern Polytechnical University Apr. Vol. 30 2012 No. 2 Neal-Smith 710072 CAP Neal-Smith PIO Neal-Smith V249 A 1000-2758 2012 02-0279-07 Neal-Smith CAP Neal-Smith Neal-Smith 1 STT Step Target Tracking Neal-Smith Time-Domain Neal-Smith TDNS PIO 2 3 Ⅰ Ⅱ PIO 4 TDNS 1 PIO 85. 87% ISE ITAE 5 6 8 PIO 7 CAP 2011-05-12 20090753008 1985
280 30 ITAE 2 8 1 CAP - C * B 1 - C * 1 Neal-Smith Neal-Smith Neal-Smith θ e t ITAE ITAE TDNS T T ITAE = t y t - y ref t dt 1 0 y t y ref t
2 Neal-Smith 281 2 ω n1 1 ω n1 TDNS TDNS 3 5 s ω n1 = 7 CAP ξ n1 ξ n1 = 0. 707 ξ n1 = 0. 707 ω n1 T θ2 = 6 2 2. 1 CAP T θ2 = 7 /6 GJB185-86 2 θ = 7 2 S + 7 6 3 F z S 2 + 2 0. 707 7S + 7 2 K θ ( s + 1 θ F z s = T θ ) 2 s 2 + 2ζ sp ω sp + ω 2 sp e -τ e0s 2 3 ζ sp ω sp T θ2 τ e0 θ = 7 2 S + 7 6 F z S 2 + 2 0. 707 7S + 7 2 1 4 S 2. 2 Neal-Smith 2 ξ n1 CAP TDNS ω n1 T θ2 ω n1 8 n /α T θ2 T θ2 n /α 1 /T θ2 ω n1 n /α ω n1 T θ2 CAP C-H ω n1 T θ2 2. 2. 1 Neal-Smith ξ n1 9 0. 25 s A θ c 2 5 0. 25 s D 1 /40 3 2 A 3
282 30 TDNS ω n1 < 10 rad /s D D τ T N θ RMS D τ P = τ + T N 7 θ RMS 6 1 D 2 θ RMS Neal-Smith TDNS D Neal-Smith 1 8 ω BW - D ω BW - ω BW - 1 = D - 0. 25 ln ( 1 40) 5 t P2 = T I = 1 ( - Tw ω BW ) 9 0. 25 s 1 t P1 = T L = t P2 ω 2 BW 10 1 /40 2. 2. 2 T w 8 k P τ 0. 25 θ RMS - PC PC = 57. 3tan -1 t P1 ω BW - 57. 3tan -1 t P2 ω BW 11 θ RMS PC D θ RMS 4 1 W P s = k p T L s + 1 e -τs T I s + 1 T N s + 1 6 k P 1 ~ 100 T L 0 ~ 1. 0 s T I 4 0 ~ 1. 0 s T N 0. 1 ± 20% s τ D 0. 2 ± 20% s 0. 2 s 0. 6 s 4 W P s = k p T L s + 1 T I s + 1 e -τ P s 8 τ P Step1
2 Neal-Smith 283 2 D 2. 7 s θ RMS k P k 0 100 - t P1 t P2 t P1 0 1 t P2 0 1 Step2 ± 100 /s ± 25 PC Step3 δ P = K θ θ + K q q 12 ITAE 1 CAP TDNS 1 1 12 K θ K q k P - T L T N K θ K q 1 20 1 3 3. 1 ITAE 2. 2. 2 4 NSGA-II 14 100 Pareto Pareto 1 100 5 1 s 5 s D 5 5 s θ RMS TDNS 8 5 5 Pareto 1 CAP TDNS 5 Neal-Smith CAP 3 0. 8 s 2. 7 s 4. 5 s 3. 2 1 PID NSGA-II 10 Pareto 6 TDNS 7 2. 7 s 7 8 TDNS 1 1
284 30 6 7 TDNS 8 3. 3 PIO 4 Ⅰ PIO Ⅱ PIO 1 PIO 11 2 D 1. 3 ~ 2 s θ RMS D d 2 θ RMS /dd 2 100 PIO 3 CAP θ RMS D 0. 006 5 100 0. 8 s 2. 5 s θ RMS D 56. 48 PIO 100 Ⅱ PIO TDNS PIO 1 Anderson M R. Unified Pilot-Induced Oscillation Theory Volume 3. PIO Analysis Using Multivariable Methods. Virginia Polytechnic Inst and State Univ Blacksburg Dept of Aerospace and Ocean Engineering 1995 2. Neal-Smith PID.. 2005 4 407-411 Ning Guodong Fang Zhenping. PID Multi-Objective Optimization Design Based on Pareto Optimality. Journal of Beijing University of Aeronautics and Astronautics 2005 4 407-411 in Chinses 3 Zhao Z Liu Y Yan S W et al. Pilot-Induced Oscillations PIO Susceptibility Evaluation Approach Based on Sequential Quadratic Programming. IEEE 2010 468-471 4 Bailey R E Bidlack T J. A Quantitative Criterion for Pilot-Induced Oscillations-Time Domain Neal-Smith Criterion 1996 598-610 5 Avanzini G Minisci E A. Evolutionary Design of a Full-Envelope Flight Control System for an Unstable Fighter Aircraft. IEEE 2010 1-8 6. Pareto PID. 2010 4 385-390
2 Neal-Smith 285 Liu Nannan Shi Yu Fan Shenghui. PID Multi-Objective Optimization Design Based on Pareto Optimality. Information and Control 2010 4 385-390 in Chinese 7. PID. 2009 21 003 61-65 Li Xuebin. Multi-Objective Optimization and Multi-Attribute Decision Making for PID Controller Parameters Tuning in Ship Diesel Design. Journal of Naval University of Engineering 2009 21 003 61-65 in Chinese 8.. 2003 Gao Jinyuan Li Luyu Feng Yachang. Aircraft Handling Qualities. Beijing National Denfence Industry Press 2003 in Chinese 9 Bailey R E Bidlack T J. Unified Pilot-Induced Oscillation Theory. Volume 4. Time-Domain Neal-Smith Criterion. Calspan Advanced Technology Cenier Buffalo NY 1995 8-10 10 K D S A A P. A Fast Elitist Non-Domina2ted Sorting Genetic Algorithm for Multi-Objective Optimization NSGA2 II. Paris France 2000 1-11 11. PIO.. 2010 4 1-4 Zhao Zhizhong Liu Yan Gao Zhenghong. PID Susceptibility Evaluation Approach Based on Sequential Quadratic Programming. Flight Dynamics 2010 4 1-4 in Chinese A Fairly Successful Exploration of Multi-Objective Optimization of Aircraft Flight Control System Design Using Time-Domain Neal-Smith Criterion Nie Rui Zhang Weiguo Li Guangwen Liu Xiaoxiong Department of Automatic Control Northwestern Polytechnical University Xi'an 710072 China Abstract Sections 1 and 2 of the full paper explain what we believe to be a fairly successful exploration. Their core consists of 1 to match the pilot and the aircraft with the man-vehicle closed loop reference model we introduce the flight quality requirements into the optimization algorithm to design the aircraft flight control system 2 we establish the equivalent model of the aircraft with the CAP criterion 3 using the time-domain Neal-Smith criterion we perform the multi-objective optimization to obtain the optimal pilot reference models under different tasks 4 we select the optimal pilot reference model according to the flight quality requirements and combine it with the aircraft equivalent model to form the man-vehicle closed loop reference model 5 with the man-vehicle closed loop reference model we design the aircraft flight control system by using the multi-objective optimization algorithm. Section 3 simulates the multi-objective optimization of the pilot reference model and the man-vehicle closed loop reference model the simulation results given in Figs. 5 through 8 and their analysis show preliminarily that our multi-objective optimization algorithm can not only tune the parameters of the controller of the aircraft flight control system but also select the optimal pilot reference model under different tasks and use the optimization results to evaluate the PIO pilot-induced oscillation susceptibility. Key words aircraft algorithms analysis closed loop control systems design evaluation man machine systems models numerical control optimization parameter optimization requirements engineering tracking position sensitivity analysis simulation standards time domain analysis flight control systems multiobjective optimization flight quality reference model time-domain Neal-Smith criterion