Domain Adaptation and Federated Learning for Ultrasonic Monitoring of Beer Fermentation
Domain Adaptation and Federated Learning for Ultrasonic Monitoring of Beer Fermentation
复制标题
DOI：
10.3390/fermentation7040253
复制
发表时间：
2021-12-01
影响因子：
3.7
通讯作者：
Watson, Nicholas J.
中科院分区：
文献类型：
--
作者：
Bowler, Alexander L.;Pound, Michael P.;Watson, Nicholas J.
啤酒发酵过程传统上是通过采样和离线麦芽汁密度测量来监测的。在线和在线传感器将提供发酵进度的实时数据，同时最大限度地减少人为参与，从而能够识别滞后发酵或预测乙醇生产终点。超声波传感器以前用于在线发酵监测，并且越来越多地与机器学习模型相结合来解释传感器测量结果。然而，发酵过程通常会持续很多天，因此需要投入大量时间来从足够数量的批次中收集数据以进行机器学习模型训练。如果必须监控不同的发酵过程，例如精酿啤酒厂中不同的配方，则这种工作量的支出必须成倍增加。在这项工作中，评估了三种方法，以使用先前从实验室规模发酵收集的超声波传感器数据来提高工业规模发酵过程中的机器学习模型的准确性。这些方法包括同时在两个领域上训练模型、在联邦学习策略中训练模型以保护数据隐私，以及在工业规模数据上微调性能最佳的模型。与仅基于工业发酵数据的训练相比，所有方法都提供了更高的预测准确性。联邦学习方法表现最佳，与基本案例模型相比，16 项机器学习任务中的 14 项达到了更高的准确性。
Beer fermentation processes are traditionally monitored through sampling and off-line wort density measurements. In-line and on-line sensors would provide real-time data on the fermentation progress whilst minimising human involvement, enabling identification of lagging fermentations or prediction of ethanol production end points. Ultrasonic sensors have previously been used for in-line and on-line fermentation monitoring and are increasingly being combined with machine learning models to interpret the sensor measurements. However, fermentation processes typically last many days and so impose a significant time investment to collect data from a sufficient number of batches for machine learning model training. This expenditure of effort must be multiplied if different fermentation processes must be monitored, such as varying formulations in craft breweries. In this work, three methodologies are evaluated to use previously collected ultrasonic sensor data from laboratory scale fermentations to improve machine learning model accuracy on an industrial scale fermentation process. These methodologies include training models on both domains simultaneously, training models in a federated learning strategy to preserve data privacy, and fine-tuning the best performing models on the industrial scale data. All methodologies provided increased prediction accuracy compared with training based solely on the industrial fermentation data. The federated learning methodology performed best, achieving higher accuracy for 14 out of 16 machine learning tasks compared with the base case model.