Search for: All records

Pressure swing adsorption (PSA) is a widely used technology to separate a gas product from impurities in a variety of fields. Due to the complexity of PSA operations, process and instrument faults can occur at different parts and/or steps of the process. Thus, effective process monitoring is critical for ensuring efficient and safe operations of PSA systems. However, multi-bed PSA processes present several major challenges to process monitoring. First, a PSA process is operated in a periodic or cyclic fashion and never reaches a steady state; Second, the duration of different operation cycles is dynamically controlled in response to various disturbances, which results in a wide range of normal operation trajectories. Third, there is limited data for process monitoring, and bed pressure is usually the only measured variable for process monitoring. These key characteristics of the PSA operation make process monitoring, especially early fault detection, significantly more challenging than that for a continuous process operated at a steady state. To address these challenges, we propose a feature-based statistical process monitoring (SPM) framework for PSA processes, namely feature space monitoring (FSM). Through feature engineering and feature selection, we show that FSM can naturally handle the key challenges in PSA process monitoring and achieve early detection of subtle faults from a wide range of normal operating conditions. The performance of FSM is compared to the conventional SPM methods using both simulated and real faults from an industrial PSA process. The results demonstrate FSM’s superior performance in fault detection and fault diagnosis compared to the traditional SPM methods. In particular, the robust monitoring performance from FSM is achieved without any data preprocessing, trajectory alignment or synchronization required by the conventional SPM methods. 

In the last few decades, various spectroscopic soft sensors that predict sample properties from its spectroscopic readings have been reported. To improve prediction performance, variable selection that aims to eliminate irrelevant wavelengths is often performed prior to soft sensor model building. However, due to the data-driven nature of many variable selection methods, they can be sensitive to the choice of the training data, and oftentimes the selected wavelengths show little connection to the underlying chemical bonds or function groups that determine the property of the sample. To address these limitations, we proposed a new variable selection method, namely consistency enhanced evolution for variable selection (CEEVS), which focuses on identifying the variables that are consistently selected from different training dataset. To demonstrate the effectiveness and robustness of CEEVS, we compared it with three representative variable selection methods using two published NIR datasets. We show that by identifying variables with high selection consistency, CEEVS not only achieves improved soft sensor performance, but also identifies key chemical information from spectroscopic data.