Abstract:Simulated moving bed (SMB) technology is considered one of the most successful techniques in chromatographic separation. However, due to the nonlinearity caused by discrete events and sensitivity to numerous separation performance parameters, purity control in SMB systems has been a challenging issue. Fuzzy controllers are increasingly popular in industrial environments due to their simplicity and effectiveness in handling nonlinearity. However, traditional fuzzy controllers used in industry often overlook considerations of error acceleration, resulting in slight deviations from target values under steady-state conditions and oscillatory behavior when system parameters change. This study proposes an advanced fuzzy controller, where in a series of experiments, the purity control targets for component B are set at 94% and 96%, and for component A are set at 96% and 96%, respectively. Experimental results indicate that the advanced fuzzy controller achieves higher precision, with an average deviation of around 0.1%, for both components B and A. Importantly, under variations in adsorbent parameter(from 0.01 to 0.03), feed concentration(from 4.5 to 5.2), and switching time(from 178 to 182), the experimental results demonstrate smoother control with the advanced controller, particularly when oscillations occur with conventional fuzzy controllers due to switching time variations, indicating robust control with the advanced fuzzy controller.
What problem does this paper attempt to address?
The paper aims to address the purity control issue in Simulated Moving Bed (SMB) technology. SMB technology is highly successful in chromatographic separation, but due to its nonlinear characteristics and sensitivity to multiple separation performance parameters, purity control in SMB systems becomes extremely challenging. Traditional fuzzy controllers, although simple and effective, often overlook the consideration of error acceleration when dealing with nonlinear problems, leading to slight deviations under steady-state conditions and oscillatory behavior when system parameters change.
To solve these problems, the study proposes an advanced fuzzy controller. Through a series of experiments, the controller sets the purity target for component B at 94% and 96%, and the purity target for component A at 96% and 96%. Experimental results show that this advanced fuzzy controller can achieve higher accuracy, with an average deviation of about 0.1% for components B and A. Additionally, under changes in adsorbent parameters, feed concentration, and switching time, the advanced fuzzy controller exhibits smoother control effects, especially when the traditional fuzzy controller shows oscillations due to switching time changes, the advanced controller demonstrates stronger robustness. In summary, although the advanced fuzzy controller exhibits higher fluctuations in some cases, it is more stable than the traditional fuzzy controller.