A PHASE PREDICTION ALGORITHM OF HIGH ENTROPY ALLOYS BASED ON META-LEARNING AND ENSEMBLE LEARNING

Accurate phase prediction of high entropy alloys is beneficial to reduce the workload of material design and development cycle, and improve the performance of materials. Therefore, a phase prediction algorithm of high entropy alloys based on meta-learning and ensemble learning is proposed. The algorithm consisted of relation mapping model and optimization model. Among them, the former established a mapping relationship the meta-features combined with material knowledge and the performance of the selective ensemble learning to recommend an appropriate ensemble algorithm. The latter adopted artificial bee colony algorithm based on single accuracy constraint to improve the accuracy of ensemble learning. The experimental results show that the prediction performance of this algorithm is better than that of other selective ensemble learning algorithms.