Modeling and Optimization of an Intelligent Coagulant Dosing System for Slurry Dewatering Based on LQG Control

Coagulant dosing plays a critical role in slurry solid-liquid separation, directly affecting dewatering efficiency and effluent quality. Traditional dosing methods based on empirical rules or fixed ratios lack adaptability, leading to performance fluctuations and chemical waste. This paper presents an LQG-based intelligent dosing strategy using multi-parameter feedback. A dynamic state-space model integrates sensor data—including flow rate, density, concentration, moisture, and turbidity—for real-time state estimation and optimal control of dosing concentration and flow. A dual-stage dosing structure is adopted, combining estimation and feedback regulation. Field tests demonstrate that the proposed system improves dosing accuracy (±2.1%) and increases coagulant efficiency by 14.2%. The method is adaptable to complex scenarios such as municipal wastewater, tunneling slurry, and mine tailings, showing strong practical value.