Optimization Of A Computer Model Of A Grinding Process Using Genetic Algorithms

Grinding is a necessary component in the processing of a number of minerals, and improvements in this area could provide substantial cost savings for the minerals industry. Optimizing the process of grinding, however, is a difficult task because there are multiple objectives; i.e., several aspects of the process are important including the fineness of the ground product, the energy consumed in the process, and the viscosity characteristics of the final slurry. Genetic algorithms (GAs) are search algorithms based on the mechanics of natural genetics and have been proven capable of solving multiple objective optimization problems. These robust search techniques are able to rapidly locate near-optimum solutions across a broad spectrum of very difficult optimization problems. In this paper, two varieties of GA are used to optimize a grinding process as it is described by a computer model: (1) a simple genetic algorithm (SGA) and (2) a micro genetic algorithm (MGA). The performance of these techniques is compared to the performance of a more conventional search technique, a simplex algorithm. Results indicate that the GAs are efficient, versatile, quite capable of improving the grinding process, and easily incorporated into the toolbox of today's process engineer.