论文部分内容阅读
针对烧结配矿中传统试错法难以求出最优解的问题,综合考虑品位、成本、库存、国外矿比例等多种影响因素,建立其优化配比的线性规划模型,用带罚函数的一阶梯度法完成模型求解。并在构建知识库和模型库的基础上,引入专家系统的解释器和推理机制,建立一套基于专家系统的多角色决策支持系统。实验结果表明,该系统在满足烧结配矿工艺要求的基础上,能降低生产成本7.97%。
In view of the difficulty of finding the optimal solution for the traditional trial-and-error method in sintering and ore blending, a linear programming model of optimal ratio is established considering various factors such as grade, cost, inventory, foreign ore ratio, First Order Gradient Model Finding Solution. Based on the construction of knowledge base and model base, this paper introduces the interpreter and reasoning mechanism of expert system and establishes a set of multi-character decision support system based on expert system. The experimental results show that the system can reduce the production cost by 7.97% on the basis of meeting the requirements of sintering and ore blending.