Recommendation Method of Ore Blending Based on Thermodynamic Principle and Adaptive Step Size
In the modern metallurgical enterprises, it is necessary to design ore concentrates and production parameters according to the production, which is called ore blending. This paper presents a method of ore blending recommendation using thermodynamic principle and adaptive step size. The whole ore blending process is divided into three parts: blanking recommendation, parameter calculation and result correction. In the part of blanking recommendation, a case-based recommendation system is used to recommend the concentrate list. In the parameter calculation section, the SOMA is used to calculate the concentrate phase. And combined with the method of concentrate phase adjustment, an intelligent algorithm for rapid and accurate calculation of concentrate phase is proposed. Then the smelting parameters are calculated based on thermodynamic principle. In the result correction part, the high accuracy ore blending recommendation is achieved by combining adaptive step size with thermodynamic calculation software. Through the combination of three parts, a fast recommendation method for intelligent ore blending is proposed. The rationality and effectiveness of the method are verified by experiments.
KeywordsOre blending recommendation Concentrate phase calculation Swarm intelligence algorithm SOMA Adaptive step size
This work was supported in part by The National Key Research and Development Program of China (Grant No. 2016YFB0700502, 2016YF- B1001404) and National Natural Science Foundation of China (No. 61572075).
- 1.Li, Z., Cui, Z.: Multi-objective optimization of ore blending based on genetic algorithm. J. Guangxi Univ. (National Science Edition) 05, 1230–1238 (2013)Google Scholar
- 2.Kai-Lin, H.U., Ping, L.I.: The optimized scheduling for iron-making bulk ore blending process based on improved ant colony optimization. J. Shanghai Jiaotong Univ. 45(8), 1105–1112 (2011)Google Scholar
- 4.Ke, L., He, X., Ye, Y., He, H.: Current research situation and development trend of ore blending optimization technology. China Min. 26(01), 77–82 (2017)Google Scholar
- 13.Deep, K., Dipti: A new hybrid Self Organizing Migrating Genetic Algorithm for function optimization. In: IEEE Congress on Evolutionary Computation, pp. 2796–2803 (2007)Google Scholar
- 14.Kadlec, P., Raida, Z.: Self-organizing migrating algorithm for optimization with general number of objectives. In: Radioelektronika, pp. 1–5. IEEE (2012)Google Scholar