摘要
近年来,ABE(丙酮-丁醇-乙醇)已被证明是一种很有前景的“绿色”替代燃料,通常由生物发酵法产生。已有研究发现,增加ABE燃料中丁醇的占比能进一步提升ABE燃料性能,同时,提高丁醇产量也是ABE生物发酵工艺的研究重点。因此,基于优良菌种的发酵数据,利用BP人工神经网络结合遗传算法建立ABE发酵的预测模型,并进行优化发酵条件的探究。结果表明:所构建的BP和GA-BP模型均具有非常准确的拟合性和预测性,GA-BP模型的拟合预测性能较优于BP模型,能作为发酵工艺优化的有力工具使用。当葡萄糖浓度为61 g/L、初始pH为6.1、初始温度为34℃、接种量为13%时,ABE发酵具有最大的丁醇预测产量,为19.98 g/L,丁醇产量在ABE燃料中占比高达81.85%。
In recent years,ABE(acetone-butanol-ethanol)has proven to be a promising"green"alternative fuel,which is usually produced by biological fermentation.Research has found that increasing the ratio of butanol in ABE fuel can further improve the performance of ABE fuel.Simultaneously,improving the production of butanol is also the research focus of ABE biological fermentation process.Therefore,this paper builds a prediction model of ABE fermentation based on the fermentation data of excellent bacteria,using BP artificial neural network and genetic algorithm to establish a prediction model of ABE fermentation,and to explore the optimization of fermentation conditions.The results show that the constructed BP model and GA-BP model have very accurate fit and predictability.GA-BP model has better fitting prediction performance than BP model,and can be used as a powerful tool for fermentation process optimization.When the glucose concentration was 61 g/L,the initial pH was 6.1,the initial temperature was 34℃,and the inoculum size was 13%,the ABE fermentation had the largest predicted butanol yield of 19.98 g/L,butanol production accounts for as high as 81.85%in ABE fuel.
作者
耿振龙
孙天韵
李肖丹
刘昕
金超
刘海峰
Geng Zhenlong;Sun Tianyun;Li Xiaodan;Liu Xin;Jin Chao;Liu Haifeng(State Key Laboratory of Engines,Tianjin University,Tianjin 300072,China;School of Environmental Science and Engineering,Tianjin University,Tianjin 300072,China;Tianjin Key Laboratory of Biomass/Wastes Utilization,Tianjin University,Tianjin 300072,China)
出处
《环境工程》
CAS
CSCD
北大核心
2023年第S02期914-919,923,共7页
Environmental Engineering
基金
国家自然科学基金(52176125)
天津市杰出青年基金(20JCJQJC00160)
