Design of Modified Predictive Active Disturbance Rejection Controller for Cross Direction Basis Weight Control Process of Paper Machine

Abstract

This paper focuses on the demand for accurate cross direction basis weight of paper and proposes a predictive active disturbance rejection controller (ADRC) with good real-time performance and strong robustness, which is effective for processes with large time delays and is applied to the cross direction basis weight of paper. To address the time-varying uncertainties caused by fluctuations in the speed of the paper machine, an ADRC with a feedback filter is utilized to actively estimate and compensate for disturbances. For the significant time delay characteristic of the cross direction basis weight loop, a predictive structure with a weighted average to mitigate the delay is adopted to compensate for the untimeliness caused by the significant delay. This leads to a modified predictive ADRC strategy that can both predict information and compensate for total disturbances, aiming to solve the control challenges in the cross direction basis weight of paper loop due to uncertainties and significant time delay characteristics. The paper first elaborates on the design process of the predictive ADRC, analyzes its stability, and then presents the rules for controller parameter tuning. Finally, it compares the proposed controller with three other controllers with good real-time performance through simulation. The research results indicate that the modified predictive ADRC proposed in this paper has excellent disturbance rejection capability and setpoint tracking performance, with the advantages of simple parameter tuning, good real-time performance, and smooth control process, providing an effective method for process control in the low-carbon papermaking industry.