基于LSTM模型的蒸汽消耗预测  

Optimization of Boiler Scheduling Based on LSTM Model for Predicting Steam Consumption in the Cigarette Industry

作　　者：赵璟瑞 彭韵超 杨梅 ZHAO Jingrui;PENG Yunchao;YANG Mei

机构地区：[1]红塔烟草(集团)有限责任公司昭通卷烟厂,云南昭通657100

出　　处：《今日自动化》2024年第6期186-188,共3页Automation Today

摘　　要：在卷烟生产过程中,蒸汽属于二次生产加工能源,一般由动力车间根据生产需求计划进行生产,是不可或缺的能源之一。准确预测蒸汽消耗对于节约能源、优化锅炉调度、降低成本和提高效率至关重要。传统的预测方法通常存在精度不足或无法处理时间序列数据的复杂性等问题。文章提出了一种基于LSTM的模型来预测卷烟生产过程中的蒸汽消耗量,并探讨了如何利用该模型进行锅炉调度的优化。通过实际数据集的实验验证,证明了LSTM模型的优越性和其在卷烟工业中的应用潜力。In the production process of cigarettes,steam belongs to the secondary production and processing energy,which is generally produced by the power workshop according to the production demand plan and is one of the indispensable energy sources.Accurately predicting steam consumption is crucial for energy conservation,optimizing boiler scheduling,reducing costs,and improving efficiency.Traditional prediction methods often suffer from issues such as insufficient accuracy or inability to handle the complexity of time series data.This article proposes a model based on Long Short Term Memory(LSTM)to predict steam consumption in cigarette production processes,and explores how to use this model to optimize boiler scheduling.The superiority of the LSTM model and its potential application in the cigarette industry have been demonstrated through experimental verification on actual datasets.

关 键 词：蒸汽消耗 锅炉调度 LSTM模型 

分 类 号：TK228[动力工程及工程热物理—动力机械及工程]