Predictive controller with estimation of measurable disturbances. Application to an olive oil mill C. Bordons*, J.R. Cueli Escuela Superior de Ingenieros. Universidad de Sevilla, Camino de los Descubrimientos, s/n. 41092 Sevilla, Spain Received 11 November 2002; received in revised form 6 June 2003; accepted 13 June 2003 Abstract This paper describes the application of a predictive controller that deals with measurable disturbances in the extraction process in an olive oil mill. The work focuses on the thermal part of the process, where the raw material is prepared for the mechanical separation. The system under consideration can be viewed as composed of several changing-level stirred tanks. The paper shows the development of the controller based upon a model obtained from first principles combined with experimental results and validated with real data. Strong disturbances and large time delays appear in the process, so predictive control strategies have been tested under linear and nonlinear simulation. Finally, they have been implemented on the real plant. A study about the consideration of different models for the estimation of measurable disturbances along the prediction horizon has been carried out, showing that a good performance can be obtained by the use of an appropriate model. A new idea that can improve periodic disturbance rejection in Model Based Predictive controllers is presented. # 2003 Elsevier Ltd. All rights reserved. Keywords: Process control; Predictive control; Agricultural processes 1. Introduction The automatic control of the extraction of oil out of olives is still an open field, since many installations are usually operated in manual mode [10]. As olive oil mills are becoming bigger the opportunities for automation are increasing, therefore it is important to acquire the necessary knowledge of the process behaviour in order to design the appropriate control strategies. The process is composed of several operations: recol- lection and reception of raw material (olives), washing, preparation, extraction, and storage of the produced oil [5]. Fig. 1 shows the most important phases of the pro- cess: preparation and extraction. The preparation phase is crucial for the whole pro- cess; it consists of two sub-processes. The first one is olive crushing by a special mill, whose objective is to destroy the olive cells where oil is stored. The second one aims at homogenizing the paste by stirring it while its temperature is kept constant at a specific value (around 35  C). This is performed by a machine called a thermomixer, which homogenizes the three phases of the paste (oil, water and a by-product) while it exchanges energy with surrounding pipes of hot water. This is done in order to facilitate oil extraction in the following process: mechanical separation in the decanter. This paper is focused on thermomixer control since homo- genization is crucial in the entire process. Bad opera- tional conditions in the thermomixer can dramatically reduce the quality and quantity of the final product. There are three main obstacles that appear when try- ing to maintain the optimal operating conditions in the thermomixer. The first one is the existence of large delays (around one and a half hours) because of the thermal nature of the process. The predictive controller treats the delays in a convenient way. The second obstacle is caused by the on–off mechanism of feeding the paste, so the inlet paste flow does not reach a con- stant value. These changes introduce periodic variations in level and therefore temperature changes since the quantity of product inside the machine varies. As the level can be easily measured, it can be considered as a measurable disturbance and hence can be taken into account by the predictive algorithm as a feedforward 0959-1524/$ - see front matter # 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0959-1524(03)00043-X Journal of Process Control 14 (2004) 305–315 www.elsevier.com/locate/jprocont * Corresponding author. Tel.: +34-954487348; fax: +34- 954487340. E-mail addresses: bordons@cartuja.us.es (C. Bordons), cueli@