An infrared optical tomography based on independent component analysis for measurement in turbid liquid

摘要

Process tomography is a tool that provides an unperturbed way to examine and investigate the internal behaviour of flow process. The tool has successfully made a beneficial contribution in measuring parameters such as mass flow rate, concentration profile and particle sizing. Among the parameters that can be inspected using tomography technique is the turbidity level of water. Most of the turbidimeters in the market are in the form of point sensor, which means the meter needs to be put on the water sample. This kind of measurement is unsuitable for industrial flow since it disturbs the flow. The investigation in distinguishing the spatial distribution in two-phase flow attracts a major interest in industry. In beverage industry, the existing of gas bubble in opaque liquid such as milk has degraded the quality of product. This thesis presents an investigation into the application of optical tomography through the use of Independent Component Analysis (ICA) method to estimate turbidity level of water and explore the presence of gas bubble in contaminated water. The system consists of eighteen infrared transmitters and eighteen receivers in which the light projection is designed in fan beam mode. An ICA algorithm has been implemented to analyse the data and the LabVIEW software was used to construct 18 x 18 pixels of concentration profile. In water turbidity experiment, several volumes of green colour ingredients were mixed together with pure water for varying the turbidity level of the water. For gas bubble's investigation, three types of flow conditions were studied: low bubble flow, medium bubble flow and high bubble flow. The behaviour of gas bubbles was investigated in contaminated water in which the water sample was prepared by adding 25 ml of colour ingredients into 3 liters of pure water. The result shows that the application of ICA has enabled the system to estimate the turbidity level and detect the presence of gas bubbles in contaminated water. This information is expected to provide vital information on the flow inside the pipe and hence, could be very significant in increasing the accuracy of the process industries.