Multi-Agent Systems for District Heating Management

This thesis investigates the applicability of multi-agent systems as a control approach for district heating systems. The consumers, i.e., the heat exchange systems, in current district heating systems are purely reactive devices and have typically no communication capabilities. They are only able to make local decisions without taking into account the global situation in the system. In this work, the possibilities of a new type of heat exchanger systems that has an open software environment and communication capabilities are explored. Operators of district heating systems have several, often conflicting, goals, e.g., to satisfy the demand of the customers and to minimize production costs. Major concerns are how to cope with the uncertainty caused by discrepancies between the estimated and actual customer demand, and the temporal constraints imposed by the relatively long production and distribution times (up to 24 hours). The approach studied in this thesis is to equip each consumer with an agent that makes predictions of future needs and to form clusters of consumers within which it is possible to redistribute resources fast and at a low cost. The agents have two fundamental goals: the local goal, which is to satisfy the customers’ needs, and the global goal, which is to improve the overall performance of the system by cooperating with the other agents in the cluster. Results from a simulation study indicate that the suggested approach makes it possible to reduce production while maintaining the quality of service. The study also show that it is possible to control the trade-off between quality-of-service and degree of surplus production. In another study, a smallscale experiment in a controlled physical environment, two agent-based approaches are evaluated and compared to existing technologies. The experiment shows that it is possible to automatically load balance a small district heating network using agent technology. Finally, a generalized formal characterization of the problem space under investigation is provided, i.e., production and logistics network management, together with a preliminary evaluation of the applicability of the suggested multi-agent system approach for this general problem area