Online modeling and prediction of a hydraulic force-acting system using neural networks

Investigates the experimental modeling of the dynamic behavior of a force-acting industrial hydraulic actuator using a neural network (NN). Due to variable environmental stiffness as well as the characteristics of hydraulic components, the dynamics of the system is time-varying and highly nonlinear. It is therefore desirable to develop a nonlinear modeling scheme, based on NNs, to estimate and predict the output of the system online. In this paper, the predictability of an online-trained NN modeling a hydraulic force-acting system is first compared to a linear model. The result demonstrates that the NN outperforms its linear counterpart in terms of multi-step prediction. Then, a more detailed discussion of the online training of the NN is provided. The related aspects include the choice of the window length, the NN's structure and the criterion for terminating the training. The work studied in this paper should help in the design of appropriate force-control law and/or fault diagnosis algorithms.

[1]  N. Sepehri,et al.  Design of a hydraulic force control system using a generalised predictive control algorithm , 1998 .

[2]  R. Burton,et al.  Experimental Identification of a Flow Orifice Using a Neural Network and the Conjugate Gradient Method , 1996 .

[3]  Tariq Samad,et al.  Back propagation with expected source values , 1991, Neural Networks.

[4]  Nariman Sepehri,et al.  Modeling and prediction of hydraulic servo actuators with neural networks , 1999, Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251).

[5]  Warren P. Seering,et al.  Understanding bandwidth limitations in robot force control , 1987, Proceedings. 1987 IEEE International Conference on Robotics and Automation.

[6]  James E. Bobrow,et al.  Adaptive, High Bandwidth Control of a Hydraulic Actuator , 1996 .

[7]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[8]  Rolf Isermann,et al.  Identification of nonlinear dynamic systems classical methods versus radial basis function networks , 1995, Proceedings of 1995 American Control Conference - ACC'95.

[9]  Rambod Rastegari,et al.  Force tracking in multiple impedance control of space free-flyers , 2000, Proceedings. 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2000) (Cat. No.00CH37113).

[10]  Pradeep K. Khosla,et al.  Analysis and experimental verification of a forth order plant model for manipulator force control , 1994, IEEE Robotics & Automation Magazine.

[11]  Pentti Vähä.  Application of parameter adaptive approach to servo control of a hydraulic manipulator , 1988 .

[12]  Rolf Isermann.  Digital Control Systems , 1981 .

[13]  Lennart Ljung,et al.  Neural Networks in System Identification , 1994 .

[14]  Nariman Sepehri,et al.  Modifying Weights Layer-By-Layer with Levenberg-Marquardt Backpropagation Algorithm , 2001, Intell. Autom. Soft Comput..

[15]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.