No abstract

No abstract

In this work, control strategies for aggregation of a portfolio of supermarkets towards the electricity balancing market, is investigated. The supermarkets are able to shift the power consumption in time by pre-cooling the contained foodstuff. It is shown how the flexibility of an individual supermarket can be modeled and how this model can be used by an aggregator to manage the portfolio to deliver upward and downward regulation. Two control strategies for managing the portfolio to follow a power reference are presented and compared. The first strategy is a non-convex predictive control strategy while the second strategy consists of a PI controller and a dispatch algorithm. The predictive controller has a high performance but is computationally heavy. In contrast the PI/dispatch strategy has lower performance, but requires little computational effort and scales well with the number of supermarkets. Two simulations are conducted based on high-fidelity supermarket models: a small-scale simulation with 20 supermarkets where the performance of the two strategies are compared and a largescale simulation with 400 supermarkets which only the PI/dispatch controller is able to handle. The large-scale simulation shows how a portfolio of 400 supermarkets successfully can be used for upward regulation of 900 kW for a two hour period.

No abstract

This paper presents a study of models for forecasting the electrical load for supermarket refrigeration. The data used for building the models consists of load measurements, local climate measurements and weather forecasts. The load measurements are from a supermarket located in a village in Denmark. Every hour the hourly electrical load for refrigeration is forecasted for the following 42 hours. The forecast models are adaptive linear time series models. The model has two regimes; one for opening hours and one for closing hours, this is modelled by a regime switching model and two different methods for predicting the regimes are tested. The dynamic relation between the weather and the load is modelled by simple transfer functions and the non-linearities are described using spline functions. The results are thoroughly evaluated and it is shown that the spline functions are suitable for handling the nonlinear relations and that after applying an auto-regressive noise model the one-step ahead residuals do not contain further significant information.