Adaptive neural network control of pneumatic servo system considering state constraints
Hai-Peng Ren,Shan-Shan Jiao,Jie Li,Yi Deng
DOI: https://doi.org/10.1016/j.ymssp.2021.107979
IF: 8.4
2022-01-01
Mechanical Systems and Signal Processing
Abstract:<p>At present, there are many efforts put into high performance tracking control of the pneumatic position servo system that have achieved considerable results, but the state constraints are rarely considered in the existing methods. The controller designed without considering the state constraints might lead to the damage of the device or the degradation of the expected performance. With the development of smart mechatronics, it is expected that the field engineers just connect the system together and put the controller into effect, which needs the controller have ability to deal with unknown model, unknown valve zero point, unknown control direction, etc, automatically. This paper aims at this complicated goal as well as considering the state constraints to improve the tracking accuracy. To fulfill the comprehensive goal, radial basis function neural network (RBFNN) is employed to treat the unknown model, unknown disturbance, unknown valve zero point, Nussbaum function is used to deal with unknown control direction, barrier Lyapunov function (BLF) is used to tackle the stability problem considering state constraints. The stability of the controlled system is proved by using Lyapunov theory and Young's inequality. Experiment results prove the effectiveness of the method, and the comparative experiment results show the superiority of the proposed method.</p>
engineering, mechanical