论文部分内容阅读
乙烯装置作为石化行业能耗大户,在乙烯装置能量优化过程中,炉群系统能耗优化起到至关重要的作用。在保证工业装置产品收率不变的情况下,本文采用调整操作变量裂解炉出口温度,达到炉群整体燃料气消耗降低的目的。本文采用K均值聚类算法结合即时学习局部建模方法,建立了精确的燃料气消耗预测模型,模型平均绝对百分比误差为0.0626%,相对误差在5%以内,满足实际工业过程对预测模型精度的要求。以某组工业数据为例,通过差分进化算法,炉群整体燃料气消耗量降低2.5%,有效的通过操作变量优化达到整体乙烯装置经济效益提高。
Ethylene plant As a big energy consumer in the petrochemical industry, energy optimization of the plant system plays a crucial role in the energy optimization of ethylene plant. In order to keep the yield of industrial products unchanged, this paper adjusts the operating temperature of the cracker outlet temperature to achieve the purpose of reducing the overall fuel gas consumption of the furnace group. In this paper, an accurate prediction model of fuel gas consumption is established by K-means clustering algorithm combined with instant learning local modeling method. The average absolute percentage error of the model is 0.0626% and the relative error is less than 5%, which meets the requirements of the actual industrial process to predict model accuracy Claim. Taking a group of industrial data as an example, through the differential evolution algorithm, the overall fuel gas consumption of the furnace group is reduced by 2.5%, and the economic benefit of the overall ethylene plant is effectively improved through the operation variable optimization.