Abstract:
In order to achieve an acceptable final product quality in the dairy industry, there are strict regulations on monitoring and control of various aspects of milk such as temperature, pressure, flow, concentration, composition, hygiene, taste and smell. The target of this technology is the production of products which fulfil all qualitative requirements, together with decreased energy consumption, reduced waste and minimized costs, which can be achieved through automatic process control. Traditional process control typically uses invasive sensing techniques and is largely limited to single point measurements. However, when operating conditions change away from the initial norm, it is desirable to have a more distributed or multi-dimensional picture of the dynamic state of the process. Although instrumentation for single point measurement of physical and physical-chemical properties such as temperature, pressure, flow and levels in tanks are currently commercially available and applied in the industry, sensors to determine product concentration and composition are not yet fully developed and a need for multidimensional monitoring of these parameters exists. Electrical Resistance Tomography (ERT), one of the most common modalities of process tomography, has become a promising technique in monitoring industrial systems due to its various advantages such as high speed, low cost, no radiation hazard, and non-intrusiveness. The application of ERT to the milk processing industry has not been investigated comprehensively before and it may have numerous potential beneficial applications in monitoring and control of the various stages of industrial milk processing. The goal of this work was set to develop the required methodologies for the application of ERT to the milk processing industry and extract valuable knowledge and information of the process state. In order to achieve the stated goal the challenge was to develop the application methodologies and to relate the ERT measurements to accurate and constructive knowledge of the process, which would provide multidimensional insight into the state of the process and could then be used in a control scheme.