Modeling a Virtual Flow Sensor in a Sugar-Energy Plant using Artificial Neural Network

In the search for increased productivity, the industry developed technological strategies to achieve this goal, a strategy called Industry 4.0. In the sugarcane industry, industrial plants are looking for tools capable of optimizing processes and reducing the time interval for unscheduled downtime, allied to a low maintenance cost. This work proposes to carry out the modeling of a soft sensor for measuring the flow of the broth out of a decanter (calead broth), using the technique of artificial neural networks. The flow of calead juice is an important variable in the sugar and ethanol manufacturing process, as it directly influences the plant's thermal balance, in addition to determining the amount of inputs needed to guarantee the quality of the sugar. In this approach, data from a sugarcane plant located in the interior of Pernambuco, Camutanga, are used to create a knowledge bank for the system through the history of the juice treatment system, and thus build a virtual sensor capable of measuring the flow of calead broth, to ensure measurement efficiency in cases of physical equipment failure. The results presented by the model from tests, in three different scenarios, showed the robustness of the proposed model, and in all scenarios the standard deviation was below 3%.