金属矿山充填浓度多目标优化控制

Multi-objective Optimization Control of Filling Concentration in Metal Mines

针对常规PID控制时充填浓度存在滞后性、时变性等特点,且易受尾砂性质和粒径等因素影响,提出了一种充填浓度的多目标优化控制策略。通过对充填工艺流程进行分析,构建充填浓度的多目标控制模型,设定多目标决策变量,建立对应的目标函数和约束条件,从底流浓度、充填质量、充填成本方面对充填浓度进行多目标优化,采用遗传算法进行求解,并在MATLAB中进行仿真。通过仿真曲线可知,多目标优化算法在评价指标方面满足质量合格、低成本的控制需求,与传统PID控制算法相比,多目标优化控制算法响应速度更快,鲁棒性更强,该充填系统运行以来,充填浓度稳定在70%±1%左右,成本降低15%,满足工业生产要求。

Slurry concentration has characteristics of hysteresis and time-varying in conventional PID control,which is easily affected by the properties of tailings and particle size.Therefore,a multi-objective optimization control strategy for slurry concentration is proposed.By analyzing the filling process flow,a multi-objective control model for slurry concentration is established according to setting multi-objective decision variables and establishing corresponding objective functions and constraints.The slurry concentration is optimized by underflow concentration,filling quality and filling cost,the model was solved by genetic algorithm and simulated in MATLAB.Through the simulation curve,the multi-objective optimization algorithm meets the control requirements of qualified quality and low cost in terms of evaluation indicators.Compared with the traditional PID control algorithm,the multi-objective optimization control algorithm has faster response speed and stronger robustness.Since the operation of the filling system,the slurry concentration has stabilized at 70%±1%,and the filling cost has decreased by 15%,which meets the requirements of industrial production.