Soft Computing Methods for Control and Instrumentation

摘要

In this work, the existing soft computing techniques have been enhanced, and211u001eapplied to control and instrumentation areas. First, new soft computing methods 211u001eare proposed. A Modified Elman Neural Network (MENN) is introduced to provide 211u001efast convergence speed. Based on Muller's method, the authors propose a new 211u001ereinforcement learning method, which can converge faster than the original 211u001ealgorithm. Second, the authors study the MENN-based identification and control 211u001eproblems. A dynamical system identification scheme as well as a trajectory 211u001etracking configuration using the MENNs are discussed, respectively. Third, the 211u001eapplications of soft computing methods in velocity and acceleration acquisition 211u001ein motion control systems are discussed. The authors construct a neural network-211u001ebased acceleration acquisition scheme to obtain clean and delayless acceleration 211u001esignals.