Load Forecasting of Supermarket Refrigeration | Lisa Buth Rasmussen
| | Abstract | The Danish power production coming from renewable energy, is increasing, therefore a flexible energy system is needed. In the present Thesis a refrigeration load forecasting method is developed, which is an important tool in a future flexible energy system.
Observed refrigeration load and local ambient temperature from a Danish supermarket, Fakta, are used. Numerical weather predictions (NWP) of ambient temperature, provided by Danish Meteorological Institute (DMI), from the model DMI-Hirlam-S05, is also used.
A multiplicative seasonal autoregressive model, only based on observations, is developed and found to be inadequate for refrigeration load forecasting. A previous developed adaptive linear time-series model is used as basic model, from that three other adaptive linear time-series models are developed. These are used for forecasting 1 to 42 hours horizon, which are evaluated and performance is compared to each other.
The models are fitted using k-step recursive least squares with forgetting, and includes regime switching, diurnal curve input and low-pass filtered ambient temperature, modelled with and without basis splines.
The achieved results clearly indicate that these methods are suitable for refrigeration load forecasting and that an enhancement of the non-linear effect from the ambient temperature in the models has occurred with spline fitted ambient temperature, seeing that the method for implementing B-splines is important for the performance of the model.
Finally ideas for further refinements, such as other local inputs i.e. relative humidity in the Supermarket are discussed. Future use of the method such as controlling load consumption from a single cabinet or from several Supermarkets are also discussed. | | Type | Master's thesis [Academic thesis] | | Year | 2013 | | Publisher | Technical University of Denmark, Department of Applied Mathematics and Computer Science / DTU Co | | Address | Matematiktorvet, Building 303B, DK-2800 Kgs. Lyngby, Denmark, compute@compute.dtu.dk | | Series | M.Sc.-2013-87 | | Note | DTU supervisors: Peder Bacher, pbac@dtu.dk, DTU Compute, and Henrik Madsen, hmad@dtu.dk, DTU Compute | | Electronic version(s) | [pdf] | | Publication link | http://www.compute.dtu.dk/English.aspx | | BibTeX data | [bibtex] | | IMM Group(s) | Mathematical Statistics |
|