Ensemble‐based adaptive soft sensor for fault‐tolerant biomass monitoring

Abstract

The accuracy and precision of soft sensors depend strongly on the reliability of underlying model inputs. These inputs (particularly readings of hardware sensors) are frequently subject to faults. This study aims to develop an adaptive soft sensor capable of reliable and robust biomass concentration predictions in the presence of faulty model inputs for a Pichia pastoris bioprocess. Hence, three soft sensor submodels were developed based on three independent model inputs (base addition, CO 2 production, and mid-infrared spectrum). An ensemble-based algorithm combined the submodels to form an ensemble model, that is, an adaptive soft sensor, to achieve fault-tolerant prediction. The algorithm's basic steps are as follows: the initial determination of submodel reliability is followed by selecting appropriate submodels to generate a reliable prediction via variance-based weighting of the submodels. The adaptive soft sensor demonstrated high robustness and accuracy in biomass prediction in the presence of multiple simulated sensor faults (RMSE = 0.43 g L −1 ) and multiple real sensor faults (RMSE = 0.70 g L −1 ).

Data synchronization techniques and their impact on the prediction performance of automated recalibrated soft sensors in bioprocesses

“…Additional methods to compensate for sensor faults should also be incorporated to optimize the prediction models' robustness further. Besides the fault-tolerant fusion of redundant soft sensor models [41], the detection of sensor faults utilizing pattern recognition [42,43], symptom signal methods [36], or multivariate statistical process control [35] would be thinkable.…”

An Adaptive Soft Sensor for On‐Line Monitoring the Mass Conversion in the Emulsion Copolymerization of the Continuous SBR Process

“…Currently, data‐driven SS have gained great relevance due to significant advances in data acquisition systems and machine‐learning techniques. [ 5–12 ]…”

“…Currently, data-driven SS have gained great relevance due to significant advances in data acquisition systems and machine-learning techniques. [5][6][7][8][9][10][11][12] Unfortunately, SS estimates are always affected by unavoidable errors caused by inaccurate mathematical models, uncertain model parameters, sensor drifts, measurement noises, etc. To compensate for estimation errors, several adaptive techniques have been developed to automatically update an SS.…”

“…Many other adaptive SS have been proposed in the literature. [7,[10][11][12] Several mathematical models related to the SBR process have been published. Broadhead and Hamielec developed a nonlinear dynamic model for the production of SBR in a CSTR.…”