黄酒酿造是多菌种混合发酵,具有产物多样的特点,已有的黄酒发酵过程模型是建立在主要生化反应基础上的发酵动力学模型,模型的精度和泛化能力尚不能满足工业需求。针对黄酒醪液中生成产物多样的特征,该文利用模糊系统的建模策略,将自适应神经模糊推理系统的单维度输出扩展到多维度输出,提出了多输出自适应神经模糊推理系统模型;然后针对该模型参数量大的特点,该文将莱维飞行和层次学习策略融入随机分形搜索算法,提出了层次学习随机分形搜索算法,用于模型参数的辨识与优化。仿真结果表明,该算法提升了模型的精度和泛化能力,实现了不同生产批次黄酒发酵状态的良好预测。
Chinese rice wine making is a mixed fermentation of multiple bacterial strains, which has the characteristics of diverse metabolites. The existed modeling approaches are fermentation kinetic models built upon the majority biochemical reactions in the fermentation process. However, these models cannot meet the industrial needs because they are weak in generality and low in accuracy. According to the diversity of products produced in the rice wine fermentation process, this paper proposes the multi-output adaptive network-based fuzzy inference system (MOANFIS) by using the modeling strategy of fuzzy system. MOANFIS extends the output of adaptive network-based fuzzy inference system(ANFIS) from one dimension to multiple dimensions. Then, in view of the large number of MOANFIS model parameters, this paper integrates Levy flight and level-based learning strategy into the stochastic fractal search (SFS) algorithm, and proposes the level-based learning stochastic fractal search (LLSFS) for the identification and optimization of model parameters. The experimental results showed that the LLSFS improved the accuracy and generalization ability of MOANFIS, rice wine fermentation status of different production batches can be accurately predicted.
[1] 吴宗文, 孙军勇, 吴殿辉,等.绍兴黄酒发酵过程中有机酸及产酸细菌的初步研究[J].食品与发酵工业, 2016, 42(5):12-18.
WU Z W, SUN J Y, WU D H, et al.Preliminary study on organic acid and acid-producing bacterial during Chinese rice wine fermentation[J].Food and Fermentation Industries, 2016, 42(5):12-18.
[2] 胡武瑶, 杨昳津, 窦慧, 等.不同麦曲酿造黄酒中挥发性风味物质的代谢差异[J].食品与发酵工业, 2020, 46(8):226-233.
HU W Y,YANG Y J,DOU H,et al.Metabolic differences of volatile flavor compounds in Huangjiu fermented with different wheat Qu[J].Food and Fermentation Industries, 2020, 46(8):226-233.
[3] 谢广发. 黄酒酿造技术[M].第三版.北京:中国轻工业出版社, 2020.
XIE G F.Brewing Technology of Huangjiu[M].3th ed.Beijing:China Light Industry Press, 2020.
[4] 徐玲, 刘登峰, 徐保国.基于数字式Smith预估增量式PID黄酒发酵温度控制[J].制造业自动化, 2014,36(23):33-36;43.
XU L, LIU D F, XU B G.The temperature control of Chinese rice wine based on the digital smith predictor incremental PID[J].Manufacturing Automation, 2014,36(23):33-36;43.
[5] LIU D F, ZHANG H T, XU B G, et al.Influence of fermentation temperature and source of enzymes on enological characteristics of rice wine[J].Journal of the Institute of Brewing, 2014, 120(3):231-237.
[6] LIU D F, ZHANG H T, XU B G, et al.Development of a kinetic model structure for simultaneous saccharification and fermentation in rice wine production[J].Journal of the Institute of Brewing, 2015, 121(4):589-596.
[7] 刘登峰, 熊伟丽, 徐玲, 等.黄酒双边发酵过程的建模[J].系统仿真学报, 2014, 26(3):626-630;674.
LIU D F, XIONG W L, XU L, et al.Modeling and simulation of Nverhong rice wine pre-fermentation[J].Journal of System Simulation, 2014, 26(3):626-630;674.
[8] 宗原, 刘登峰, 刘以安.基于改进蚁狮优化算法的黄酒发酵过程模型的参数辨识[J].食品与发酵工业, 2021, 47(2):153-159.
ZONG Y, LIU D F, LIU Y A.Model parameter identification of rice wine fermentation process based on an improved ant lion algorithm[J].Food and Fermentation Industries, 2021, 47(2):153-159.
[9] JANG J S.ANFIS:Adaptive-network-based fuzzy inference system[J].IEEE Transactions on Systems, Man, and Cybernetics, 1993, 23(3):665-685.
[10] ZOUNEMAT-KERMANI M, TESHNEHLAB M.Using adaptive neuro-fuzzy inference system for hydrological time series prediction[J].Applied Soft Computing, 2008, 8(2):928-936.
[11] SHIHABUDHEEN K V, PILLAI G N.Recent advances in neuro-fuzzy system:A survey[J].Knowledge-Based Systems, 2018, 152:136-162.
[12] GOLDBERG D E.Genetic Algorithms[M].Pearson Education India, 2013.
[13] DAS S, SUGANTHAN P N.Differential evolution:A survey of the state-of-the-art[J].IEEE Transactions on Evolutionary Computation, 2010, 15(1):4-31.
[14] TANABE R, FUKUNAGA A.Success-history based parameter adaptation for differential evolution[C].2013 IEEE Congress on Evolutionary Computation.IEEE, 2013:71-78.
[15] EBERHART, SHI Y H.Particle swarm optimization:Developments, applications and resources[C].Proceedings of the 2001 Congress on Evolutionary Computation (IEEE Cat.No.01TH8546).IEEE, 2022, 1:81-86.
[16] SALIMI H.Stochastic fractal search:A powerful metaheuristic algorithm[J].Knowledge-Based Systems, 2015, 75:1-18.
[17] MIRJALILI S, LEWIS A.The whale optimization algorithm[J].Advances in Engineering Software, 2016, 95:51-67
[18] JAIN M, SINGH V, RANI A.A novel nature-inspired algorithm for optimization:Squirrel search algorithm[J].Swarm and Evolutionary Computation, 2019, 44:148-175.
[19] ABDOLLAHZADEH B, GHAREHCHOPOGH F S, MIRJALILI S.African vultures optimization algorithm:A new nature-inspired metaheuristic algorithm for global optimization problems[J].Computers & Industrial Engineering, 2021, 158:107408.
[20] YANG Q, CHEN W N, DA DENG J, et al.A level-based learning swarm optimizer for large-scale optimization[J].IEEE Transactions on Evolutionary Computation, 2017, 22(4):578-594.
[21] TAKAGI T, SUGENO M.Fuzzy identification of systems and its applications to modeling and control[J].IEEE Transactions on Systems, Man, and Cybernetics, 1985, (1):116-132.
[22] CHEGINI S N, BAGHERI A, NAJAFI F.PSOSCALF:A new hybrid PSO based on sine cosine algorithm and levy flight for solving optimization problems[J].Applied Soft Computing, 2018, 73:697-726.
