Double-Step Optimization Algorithm of PID Turning

A self-adaptive d-step optimization algorithm combined methods of the actual tuning process and on line waveform identification, including the first-coarse-tuning and second-fine-tuning, is proposed. Under the close-loop operation with coarse-tuning setting, real-time waveform is analyzed on line to provide an assessment of whether the current performance achieves the control goal. Otherwise it makes another judgment of whether a second-fine-regulation is adequate or should conduct a model identification firstly and a fine-tuning secondly, obtaining the most suitable PID-parameter at last. Simulation examples are provided to illustrate this algorithm can tune and optimize the PID-parameter on-line according to control goal, avoid the influences of disturbance or time-varying property, has high robustness and self-adaptation