Review Fuzzy logic control and soft sensing applications in food and beverage processes S. Birle * , M. A. Hussein, T. Becker Technische Universität München, Center of Life and Food Sciences Weihenstephan, (Bio-)Process Technology and Process Analysis, Weihenstephaner Steig 20, 85354 Freising, Germany article info Article history: Received 27 December 2011 Received in revised form 30 May 2012 Accepted 5 June 2012 Keywords: Fuzzy logic Soft sensing On-line monitoring Prediction Food and beverage processes abstract Biotechnological processes e particularly fermentation processes - play a very important technological an economical role for the production steps in the food and beverage sector. In order to ensure constantly high product quality combined with efficient manufacturing, intelligent control systems and strategies are required. However, biosystems contain living organisms and therefore underlie particular process dynamics such as nonlinear and time-varying behavior. Furthermore, initial process conditions cannot be kept constant and therefore precise process reproducibility hardly can be ach- ieved. On that account these multivariate systems put high requirements to the practical on-line observation, control, monitoring and prediction of significant process key parameters whose acquirement is of crucial importance for a comprehensive understanding and control of the process. During the last decades great efforts have been undertaken to cope with those challenges by means of intelligent soft computing and reveal great opportunities to integrate human expertise and learning procedures for improved process control strategies of biological systems. Particularly fuzzy logic based control systems show high potential to manage the complex production processes and to deal with fragmental process information. This review critically presents the chances as well as the limitations of fuzzy and hybrid expert system approaches in food and beverage process control from a theoretical and application based point of view. Ó 2012 Elsevier Ltd. All rights reserved. Contents 1. Introduction ...................................................................................................................... 254 2. Theory of fuzzy logic and fuzzy-based expert systems ................................................................................. 255 3. Fuzzy reasoning, sensing and control .................................................................................................257 3.1. Quality evaluation ............................................................................................................ 257 3.1.1. Fuzzy symbolic approach ............................................................................................... 257 3.2. Sensing and control .......................................................................................................... 259 3.2.1. Fuzzy based soft-sensing of process parameters and phase recognition ...................................................... 259 4. Hybrid fuzzy systems ........................................................... .................................................. 260 4.1. Introduction to neural network techniques ...................................................................................... 260 4.2. Soft sensing via artificial neural networks ....................................................................................... 261 4.3. Hybrid systems .............................................................................................................. 263 5. Conclusion and outlook ............................................................................................................ 266 References ......................................................................................................................... 267 1. Introduction Due to their inherent complexity and abundance of uncertainty factors biotechnological systems, especially fermentation processes, are very difficult to describe. The quality of the product is decisively determined by its taste which is extremely difficult to * Corresponding author. Tel.: þ49 8161 71 2623; fax: þ49 8161 71 3883. E-mail address: sbirle@wzw.tum.de (S. Birle). Contents lists available at SciVerse ScienceDirect Food Control journal homepage: www.elsevier.com/locate/foodcont 0956-7135/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodcont.2012.06.011 Food Control 29 (2013) 254e269