摘要
一种新型Al-Si—Cu-Mg铸造合金研制中，积累实验数据，用模糊建模与遗传算法优化热处理工艺。改进模糊聚类算法，获得清晰可解释的Takagi-Sugeno模糊系统，建立工艺参数与力学性能关系。用十进制编码，简化遗传操作，模糊调节交叉率和变异率，改进遗传算法。耦合模糊模型与遗传算法，将抗拉强度与延伸率分别稳定在400MPa和5％左右。实验证实上述方法是有效的。
In development of a new AI-Si-Cu-Mg casting alloy, based on a series experiments, data-driven fuzzy modeling and genetic algorithm （GA） have been used to optimize heat treatment process. With improved fuzzy clustering, Takagi-Sugeno type fuzzy inference system, which consisted of reduced interpretable rules, was used to correlate the process parameters to ultimate tensile strength and elongation. Instead of binary string encoding, more quickly base-10 genetic algorithms with simplified genetic operators were developed, in which digits vary over the numbers 0, l, 2 9 and fuzzy controllers were used to adapt the crossover and mutation. The GA and fuzzy model were incorporated to find the optimal process conditions for ultimate tensile strength and elongation about 400MPa and 5%. By comparing the experimental results, it is demonstrated that the proposed method is practical and efficient.
出处
《教学与科技》
2017年第4期7-20,共14页
Teaching and Science Technology
基金
中国工程物理研究院科学技术基金项目（编号：20010668及20000329）
