Optimization of SAG Control Rule Based on Clustering Learning Algorithm

SAG machine is an important part of the SABC process. In view of the characteristics of many variables, nonlinearity, strong coupling, large hysteresis, time-varying, the traditional control methods are difficult to achieve better results, and the introduction of clustering learning algorithm is proposed. The normal operation data of SAG machine is subjected to cluster learning analysis, and compared with the existing operating experience, and self-learning affects the control relationship between the process parameters of its operating conditions, and forming the optimized control rule set, which can be used as the rule base of expert system, instead of the original control method, its operation state is improved, that the cost is reduced, and the energy consumption is saved, and the economic benefit of concentrator is improved.