基于DE优化的炼焦能耗神经网络模型

Coking-optimized energy consumption neural network model based on DE optimization

现今焦化行业耗能高、污染严重的问题,引起社会广泛关注,使得降低炼焦能耗,保护环境成了焦化行业亟待解决的问题。针对炼焦生产过程中炼焦能耗数学模型难以建立、统计量大等问题,分析了影响炼焦能耗的主要因素,提出了神经网络数学模型。由于差分进化算法收敛速度快、不易陷入极值等优点,建立基于差分进化算法神经网络数学模型。通过MATLAB仿真实验结果表明,模型误差波动小,寻优速度快,模型精度达到95%以上。该研究为建立炼焦能耗数学模型方面,提供了新的思路,为炼焦行业高效低耗生产指明了新方向。

High energy consumption and serious pollution in the coking industry have aroused widespread concern in the society,making it an urgent problem to reduce coking energy consumption and protect the environment. In view of the problems such as difficulty in establishing mathematical model of coking energy consumption in coking production and large statistical quantity,the main factors affecting coking energy consumption were analyzed,and a neural network mathematical model was proposed. Due to the advantages of the differential evolution algorithm,such as fast convergence speed and difficulty in getting into the extreme value,the mathematical model of neural network based on the differential evolution algorithm is established. The MATLAB simulation results show that the model error fluctuation is small,the optimization speed is fast,and the model accuracy is above 95%. This study provides a new idea for the establishment of mathematical model of coking energy consumption,and points out a new direction for the efficient and low consumption production in coking industry.