Just-in-Time Kernel Learning with Adaptive Parameter Selection for Soft Sensor Modeling of Batch Processes

Autoencoder-based nonlinear Bayesian locally weighted regression for soft sensor development

The framework of locally weighted learning (LWL) has established itself as a popular tool for developing nonlinear soft sensors in process industries. For LWL-based soft sensors, the key factor for achieving high performance is to construct accurate localized models. To this end, in this paper a nonlinear local model training algorithm called nonlinear Bayesian weighted regression (NBWR) is proposed. In the NBWR, the nonlinear features of process data are first extracted by the autoencoder; then, given a query sample a local dataset is selected on the feature space and a fully Bayesian regression model with differentiated sample weights is developed. The benefits of this approach, which include better consistency of correlation, stronger abilities to deal with process nonlinearities and uncertainties, overfitting and numerical issues, lead to superior performance. The NBWR is used for developing a soft sensor under the LWL framework, and a real-world industrial process is used to evaluate the performance of the NBWR-based soft sensor. The experimental results demonstrate that the proposed method outperforms several benchmarking soft sensing approaches.

Semi-Supervised Hybrid Local Kernel Regression for Soft Sensor Modelling of Rubber-Mixing Process

Soft sensor techniques have been widely adopted in chemical industry to estimate important indices that cannot be online measured by hardware sensors. Unfortunately, due to the instinct time-variation, the small-sample condition and the uncertainty caused by the drifting of raw materials, it is exceedingly difficult to model the fed-batch processes, for instance, rubber internal mixing processing. Meanwhile, traditional global learning algorithms suffer from the outdated samples while online learning algorithms lack practicality since too many labelled samples of current batch are required to build the soft sensor. In this paper, semi-supervised hybrid local kernel regression (SHLKR) is presented to leverage both historical and online samples to semi-supervised model the soft sensor using proposed time-windows series. Moreover, the recursive formulas are deduced to improve its adaptability and feasibility. Additionally, the rubber Mooney soft sensor of internal mixing processing is implemented using real onsite data to validate proposed method. Compared with classical algorithms, the performance of SHLKR is evaluated and the contribution of unlabelled samples is discussed.

Evaluating the Performance of Newly Integrated Model in Nonlinear Chemical Process Against Missing Measurements

Application of data-driven soft sensors in manufacturing fields, for instance, chemical, pharmaceutical, and bioprocess have rapidly grown. The issue of missing measurements is common in chemical processing industries that involve data-driven soft sensors. Locally weighted Kernel partial least squares (LW-KPLS) algorithm has recently been proposed to develop adaptive soft sensors for nonlinear processes. This algorithm generally works well for complete datasets; however, it is unable to cope well with any datasets comprising missing measurements. Despite the above issue, limited studies can be found in assessing the effects of incomplete data and their treatment method on the predictive performances of LW-KPLS. To address these research gaps, therefore, a trimmed scores regression (TSR) based missing data imputation method was integrated to LW-KPLS to formulate trimmed scores regression assisted locally weighted Kernel partial least squares (TSR-LW-KPLS) model. In this study, this proposed TSR-LW-KPLS was employed to deal with missing measurements in nonlinear chemical process data. The performances of TSR-LW-KPLS were evaluated using three case studies having different percentages of missing measurements varying from 5 % to 40 %. The obtained results were then compared to the results from singular value decomposition assisted locally weighted Kernel partial least squares (SVD-LW-KPLS) model. SVD-LW-KPLS was also proposed by incorporating a singular value decomposition (SVD) based missing data treatment method into LW-KPLS. From the comparative studies, it is evident that the predictive accuracies of TSR-LW-KPLS are superior compared to the ones from SVD-LW-KPLS.

A robust localized soft sensor for particulate matter modeling in Seoul metro systems

Developing accurate soft sensors to predict and monitor the indoor air quality (IAQ) of hazardous pollutants that accumulate in underground metro systems is of key importance. The just-in-time (JIT) learning technique possesses a local feature that can track the variations in the dynamic process more effectively, which is different from the traditional soft sensor modeling methods, such as partial least squares (PLS), which models the process in an offline and global way. In this study, a robust soft sensor that combined the JIT learning technique with a least squares support vector regression (LSSVR) method, named JIT-LSSVR, was derived in order to improve the prediction performance of a PM2.5 soft sensor in a subway station. Additionally, in order to eliminate the adverse effects caused by the outliers in the process variables, an outlier detection step was integrated into the JIT-LSSVR modeling procedure. The performance evaluation results demonstrated that the proposed robust JIT-LSSVR soft sensor has the capability to model nonlinear and dynamic subway systems. The root mean square error of the JIT-LSSVR soft sensor was improved by 55% in comparison with that of the LSSVR soft sensor.

A soft sensor for bioprocess control based on sequential filtering of metabolic heat signals

Soft sensors are the combination of robust on-line sensor signals with mathematical models for deriving additional process information. Here, we apply this principle to a microbial recombinant protein production process in a bioreactor by exploiting bio-calorimetric methodology. Temperature sensor signals from the cooling system of the bioreactor were used for estimating the metabolic heat of the microbial culture and from that the specific growth rate and active biomass concentration were derived. By applying sequential digital signal filtering, the soft sensor was made more robust for industrial practice with cultures generating low metabolic heat in environments with high noise level. The estimated specific growth rate signal obtained from the three stage sequential filter allowed controlled feeding of substrate during the fed-batch phase of the production process. The biomass and growth rate estimates from the soft sensor were also compared with an alternative sensor probe and a capacitance on-line sensor, for the same variables. The comparison showed similar or better sensitivity and lower variability for the metabolic heat soft sensor suggesting that using permanent temperature sensors of a bioreactor is a realistic and inexpensive alternative for monitoring and control. However, both alternatives are easy to implement in a soft sensor, alone or in parallel.