Improved techniques for self-diagnosis of an ultrasonic flow meter

Sammanfattning: For both economical and environmental raisons, accurate measurements of energy consumptions are important. The District Heating Industry supplies buildings and private houses of a whole city in hot water through a network. The energy supplied to a customer is measured by thermometers and by a flow meter inside the district-heating substation present at the customer’s place. The annual cost of the total metering error is approximately estimated at 170 Millions of Swedish Crowns for the Swedish District Industry. Accurate measurements of the flow rate turn out to be essential to a correct estimation of the energy consumption. Ultrasonic flow meters are of particular interest because of their good accuracy for a low cost. The main features of ultrasonic flow meters are studied. The estimation of the mean flow velocity and the calibration of the meter are described. Once installed, any kind of meter provides larger errors than in the facility where it has been calibrated. It is particularly true for ultrasonic flow meters that are very sensitive to installations effects. Installations effects can either be static or dynamic. Special attention is paid to errors generated by velocity profile variations. A diagnosis of the sensitivity of the meter to installations effects would allow for compensations for the errors. It is outlined that the notion of diagnosis comes within the theory of detection. A self-diagnosis is performed by the meter itself. The decision rule has to be modelled and inserted into the meter. Signal analysis allows detection of specific noise characteristic of a particular type of installation effect. An example of self-diagnosis showing the detection of pulsations in a flow is described in details.