Methods for planning and operating decentralized combined heat and power plants

In recentyears,the numberof decentralizedcombinedheat and power (DCHP) plants, which are typically located in small communities, has grown rapidly.These relativelysmallplantsarebased on Danishenergyresources,mainly natural gas, and constitute an increasingpart of the total energy production in Denmark. The topic of this thesisis the analysisof DCHP plants, with the purpose to optimize the operation of such plants. This involvesthe modeling of district heating systems, which are frequently connected to DCHP plants, as well as the use of heat storage for balancing between heat and power production. Furthermore,the accumulated effect from increasing number of DCHP plants on the total power production is considered. Methods for calculating dynamic temperatureresponsein districtheating (DH) pipes have been reviewed and analyzed numerically.Furthermore, it has been shown that a tree-structuredDH network consistingof about one thousand pipes can be reduced to a simplechain structureof ten equivalentpipes without loosing much accuracy when temperaturedynamics are calculated. A computationally eilicientoptimization method based on stochastic dynamic programming has been designed to find an optimum start-stop strategy for a DCHP plant with a heat storage. The method focuses on how to utilize heat storage in connection with CHP production. A model for the total power production in Eastern Denmark has been applied to the accumulated DCHP production. Probability production simulationshave been extended from the traditional power-only analysisto include one or several heat supply areas. ISBN 87-550-2709-1 ISSN 0106-2840 Information Service Department . Rls@.2000