Dynamik i fjärrvärmesystem
Doctoral thesis, 1999
The purpose of this study is to gain understanding of the dynamic phenomena, which exist in district heating systems. Particularly, the importance of the network is noticed, where pressure waves and temperature waves play a major role. Most of the presented results are obtained by simulations. In many cases these are verified with measurements.
District heating systems are by nature unsteady. Control activities in production units and substations are going on continuously. The exchange of information of pressure, flow and temperature, between the production and the heat load, is transferred through the network by pressure and temperature waves.
Rapid events are calculated according to an elastic theory. At a study of the dynamic course of events after a pump trip, this theory was used. Strong and unsymmetrical pressurerizing creates major pressure waves after a pump trip, but strategically located pressure vessels in the network efficiently reduce these pressure waves. With the same theory, transient flow during valves manoeuvres with intermittent actuators was studied. The use of intermittent actuators for sectioning valves is in principle an incorrect method, which creates considerable and unnecessary mechanical stresses of the pipe network.
Heat load models, based on primary data from production units, are presented. In addition to the time of day, weekday/weekend and time of the year, only information about the outdoor temperature is required for a good prediction of the heat load and the return temperature. Furthermore, the temperature dynamics in the network should not be neglected when predicting the heat power and the flow at the production unit. The temperature dynamics are described by temperature waves.
Smooth variations in delivered heat power are often desirable. However, the pipe network as a heat storage offers no possibility to level out the daily variation of the heat power at the production unit. Nevertheless, with a correct prediction of the outdoor temperature, it is possible to reduce short peaks of delivered heat power. To even out the flow is easier, which makes it possible to keep the supply temperature low. Besides reduced heat losses, this also leads to more efficient use of the production units. The classical control curve of the supply temperature is an easy way to reduce flow variations. A more sophisticated control strategy to even out the flow is also presented.
Network models are often made very detailed to fit all kind of calculations. However, in many cases much more simple models are more suitable to use. A method to simplify network models is presented. In this procedure the "point volumes" play a major role.
district heating networks
heat load models
district heating systems