Main Data History
Show Index Toggle 0 comments
  •  Quick Edit
  • District heating literature

    This document contains mini-reviews of scientific papers related to district heating.

    Predicting load in district heating systems

    (Dotzauer 2002) develops a simple model with a piece-wise linear temperature component and a stochastic component with 1 parameter for each hour of the week. This model is fitted to actual load data (least square) and used to predict the load. The conclusion is that a simple model like this provides results that are as good as more elaborate commercial tools, but the results are sensitive to the input temperature. Therefor good temperature forecasting is key to good load prediction.

    This can be a motivation for using an ensemble of forecasts when planning the load.

    Estimating heat demand

    Cox et al. writes about the importance of using either annual, monthly or hourly time resolution for temperature data when estimating the annual heating demand. They find that temporal resolution is not super important. (Cox 2015)

    Operational optimization in a district heating system

    In (Benonysson 1995) a model of a small district heating system (Ishøj, 1 plant 17 nodes) is investigated. The supply temperature is optimized with respect to the operational cost, with electricity prices and heat consumption varying over the day. They find that “minimization of district heating cost calls for a very active control of the supply temperatures from the plant”. It is not clear how applicable their results are to a large district heating system like the one in Aarhus, where the terrains is also more hilly.

    The time delays are handled in this model.

    Machine Learning in District Heating System Energy Optimization

    (Idowu 2014) is an early account of a very ambitious work in progress. The idea is to minimize the use of energy in the DH system by controlling heat storage (accumulators) in an optimal way. They want to achieve this by using supervised learning to predict heat consumption on an individual house basis and then using the results as input to a reinforcement learning control strategy. This requires very good data on house to house level and the installation of additional sensors. Once they get some results this could be interesting. Or it may be interesting for Theis, since he works more on an individual house level.

    In any case I suggest you read this one as inspiration, Gorm.

    Forecasting Heat Load for Smart District Heating Systems: A Machine Learning Approach

    (Idowu 2014) This second article is a publication of the first part of the work outlined above. Four different supervised learning algorithms are explored in their applicability for predicting the heat load of individual substations. This is a bottoms up approach to heat load prediction, and the greatest relevance to my work is in the very nice literature review on related work in heat load prediction (on both production and consumer side).