P9
heating network means reducing the amount of heat that is lost to the ground. Heat losses up to 20% of the total produced heat are common in district heating networks, and there is a substantial economic and environmental potential in reducing heat losses. With computer simulations, I have demonstrated that it is possible to achieve small supply temperature reductions at several area substations in the Aarhus area. I have also found that the benefit of using ensemble forecasting greatly increases in area substations that are often operating close to their maximum pumping capacity. Value Creation – the production planner’s dilemma Counting cards in blackjack gives the player an edge, because it is a way of estimating his risk profile and use it to make better decisions. In the same way, knowing your risk profile and the probability of various scenarios can help you make smarter decisions in the production planning and operation of a district heating system. Imagine the following situation a production planner may face: The heat demand forecast shows that she can cover tomorrow’s heat demand using only the cheaper CHP production units – but just barely. If the forecast is too low, it will be necessary to turn on an electric boiler to cover the peak of the demand. Here comes the production planner’s dilemma. Should she play it safe and buy electricity for the boiler on the day-ahead market? This cause of action is the cheapest if the forecast turns out to be too low. Alternatively, she could choose to trust the forecast and not buy electricity for the boiler on the day-ahead market. This is the cheapest if the forecast holds, but if it turns out to be wrong, then she might need to buy electricity on the intra-day market, potentially at a much higher price. In situations like this, knowledge of how much the forecast can be trusted is highly valuable. If the production planner had been equipped with an ensemble forecast that showed very little spread, she could have chosen not to buy the day ahead without running an unnecessary risk. On the other hand, if the ensemble forecast had indicated a highly uncertain forecast, she would know to play it safe.
Figure 3: PhD fellow Magnus Dahl (on the left) from AffaldVarme Aarhus and Assistant Professor Gorm B. Andresen (on the right) in the control room of AffaldVarme Aarhus. Photo: Lars Kruse, Aarhus University.
This may sound complex, but the technique of ensemble forecasting is simple and any district heating provider can use it to gauge their weather-based risk profile. A district heating provider that wishes to begin doing ensemble forecasting will need to upgrade their weather forecast subscription to include ensemble weather forecasts. Once a data feed with an ensemble weather forecast is in place, it is easy to pass it through any weather-based model for heat load prediction to create an ensemble heat load forecast. The technique is simple, the data is available and the potential value creation is significant for modern green district heating systems. A SMART CITY SHOWCASE: In READY we demonstrate new smart city energy infrastructure and low-energy building renovations in Aarhus, Denmark and Växjö, Sweden. Both cities host some of the worlds most advanced large-scale district heating systems.
The project is funded by EU under the FP7 framework under grant agreement no 609127
Thanks to the 4DH Research Centre
Figure 2: Large district heating systems often consist of a transmission system and a distribution system. The transmission system is the production side and the distribution system is the consumer side. The transmission system and the distribution system are connected at a number of area substations with heat exchangers. Knowledge of the forecast uncertainty can be used to control the supply temperature in a smart way and reduce heat losses to the ground.
For further information please contact:
AffaldVarme Aarhus Att.: Magnus Dahl Bautavej 1 DK- 8210 Aarhus V
Phone: +45 4185 8669 magda@aarhus.dk
www.dbdh.dk
J O U R N A L N 0 . 1 / 2 0 1 7
Made with FlippingBook Ebook Creator