Modelling of calibration steam flowmeter test bench

The present project develops simulation methods for the test bench that serves to calibrate the steam flowmeters by using steam as working fluid. The test bench contains a condenser, the purpose of which is to condense the steam. By maintaining a steady-state regime, the weight of condensate measured during a given period of the time is used to calibrate the steam flowmeter. Two mathematical models are developed in this project for the heat transfer process within the condenser, which is a shell-and-tube heat exchanger. The first model is based on neural network theory for predicting steam temperature, pressure, flow and the height of condensate. The input parameters to this model include the opened-closed position of the control valves and the inlet temperature of the cooling water. The second model (physical model) is based on the mass and energy conservation principles. For increasing the accuracy, the condenser is divided into control volumes to which the mass and energy balances are applied. The model describes the dynamic response of the system at different locations including the condensate level and the steam pressure within the heat exchanger. The input values are the opened-closed position of the valves, the flow rate and inlet temperature of cooling water. Both of these models give satisfactory results compared with experimental data. However, the physical model is preferred by its capability of explaining the relationships between parameters. It is suggested for a future work that physical model is applied to a multivariable control system for the regulation of the level of condensate and the steam pressure.