Zinc flotation working condition judgment method based on improved self-adaptive multi-population genetic algorithm

Abstract

The invention discloses a zinc flotation condition judgment method based on an improved self-adaptive multi-population genetic algorithm which is applied to a foam flotation process. The method comprises the following steps of using a foam image feature data set obtained in real time as an initial training population; adding selections of an experiential worker, and on the basis of the selection, considering subjectivity of manual selection and influence degree of the environment, adding a subjective factor coefficient, namely subjectively judging an offset interval, evaluating and judging whether the current selection is within an error allowable range or not, thereby controlling the subjective influence of the human to be within a certain range. In the whole process, the current good working condition is fed back in real time. Population adjustment is conducted through the working condition returned in real time. Real-time refreshing of the optimal solution is achieved. The influence of grade value changes of different raw ore piles is reduced, and finally effective control over the grade value and the recovery rate is achieved.

Description

technical field [0001] The invention relates to a zinc flotation working condition discrimination method based on an adaptive multi-population genetic algorithm for selecting the corresponding optimal solution under the optimal working condition in the froth flotation process. Background technique [0002] In the foam flotation process, the genetic algorithm can effectively solve the optimal value problem or the optimal interval problem, and then effectively discriminate the working conditions. In the initial traditional genetic algorithm, the initial population is randomly generated, and the search rate is reduced. It is easy to fall into a local optimal state, and secondly, only the difference in individual fitness is considered, but the difference in the evolutionary state of the population is not considered. [0003] Apply the multi-population genetic algorithm to the foam flotation process, set the immigration operator, successively migrate from the previous population ...

AI Technical Summary

Problems solved by technology

Among the existing algorithms, some ignore the subjective initiative of human beings and the importance of artificial experience, and some make full use of human experience, but ignore the influence of human subjectivity and the large fluctuations in human judgment. Inevitably wrong decisions, without finding effective ways to solve the impact of human subjectivity, or without taking any measures to evaluate and compensate

[0004] Secondly, in the existing algorithm, the initial population has not changed, and the optimal value can only be obtained for objects with stable raw ore grade values. For objects with fluctuating raw ore grade values, the calculated optimal value is far from the real The offset error of the optimal value is large

Images