Stochastic chance-constrained optimization framework for the thickening-dewatering process with an uncertain feed quantity

Thickening-dewatering process is widely used for solid–liquid separation in mineral processing plants. Usually, the feed quantity of the thickener is an uncertain variable, so the underflow concentration of the thickener cannot be accurately predicted, which may result in improper operation and high security risk. This paper proposes a stochastic chance-constrained optimization framework for the thickening-dewatering process with the uncertain feed quantity. Firstly, a feed quantity predictive model is established by gray correlation analysis (GCA) and partial least squares (PLS) to reduce the uncertainty of the feed quantity. Then, the stochastic chance-constrained optimization model of the thickening-dewatering process is established to minimize the energy economic index (EEI). In order to solve the stochastic chance-constrained optimization model, the nonlinear time-of-use electricity price function in the objective is linearized, and then the stochastic chance-constrained optimization model is transformed into a deterministic model. The cases study illustrate that the optimization framework can reduce the EEI and the security risk of the thickening-dewatering process.