A Sensor Based Control Method for Turkish Coffee Cooking Process

Abstract

Due to the recent technological advancements, autonomous cookers are being developed for heterogeneous food mixtures. Advanced sensing systems play a critical role in such appliances to ensure a precise control of the cooking process and a successful operation. Such devices are of high demand among the consumers and they are built upon various systems for sensing. However, the sensing systems might encounter major difficulties if exogenous disturbances such as steam, heat or condensation on the sensor exist. This situation may lead to false executions and problems in providing sustainable performance. In this study, a state observer model is developed for an automatic Turkish coffee machine possessing an induction heater. This model can be used to estimate the temperature of the mixture with only one temperature sensor placed outside the cooking chamber. In addition, a dynamic model is constructed utilizing the data acquired by real time coffee making experiments. The proposed structure is tested by placing a temperature sensor inside the cooking chamber. The constructed model estimates the actual temperature of the mixture within an interval of 25℃ and 90℃ with %96.63 accuracy. According to these results, Turkish coffee with the desired taste and consistency can be cooked using the recommended sensing method.