Research on multi-objective ore allocation optimization based on improved differential evolutionary algorithm

The optimization problem of ore allocation is of great significance in the metallurgical industry, and its objective is to simultaneously satisfy the minimum deviation of ore grade and the maximum grinding capacity. However, due to the complex nonlinear relationship between various properties of the ore and the objective function, it is difficult for traditional optimization algorithms to solve it effectively. The study proposes a multi-objective optimization method based on the optimized radial basis neural network and improved differential evolutionary algorithm. First, the radial basis neural network is utilized to fit the objective function to capture its complex nonlinear relationship. Then, an improved differential evolutionary algorithm is used to solve it to optimize the proportion of ore at each discharge point while satisfying the constraints. The actual operation of the mine plant is optimized using the improved differential evolutionary algorithm and compared with the traditional algorithm. The experimental results show that the improved differential evolutionary algorithm is able to find a solution set close to Pareto's cutting edge within 50 generations, which has a significant advantage in terms of convergence speed and quality of solution compared to the traditional differential algorithm. This study provides an effective solution to the optimization problem of ore allocation, which has important application value.