Optimal Sensor Deployment for Parameter Estimation Precision by Integrating Bayesian Networks in Wet-Grinding Systems

Accurately and efficiently determining a system’s physical variables is crucial for precise product-quality control. This study proposes novel method optimal sensor deployment to increase the accuracy of sensing data ensure timely detection product’s particle size in wet-grinding system. approach involves three steps. First, Bayesian network (BN) designed model cause–effect relationship between by applying path model. The detectability determined confirm that mean shifts all are identifiable using sets Second, location accelerometers mounted on chamber shell according coupled computational fluid dynamics–discrete element simulations. Third, shuffled frog leaping algorithm developed combining BN minimize maximum output deviation index among sensory costs; this achieved under constraints shift detectability, achieving optimum allocation. Subsequently, case performed zirconia powder production process demonstrate proposed minimizes requirements index, costs, detectability. systematic universal; it can be integrated into monitor architecture parameter estimation other complex systems.