Driven by the opportunity to harvest the flexibility related to building climate control for demand response applications, this work presents a data-driven control approach building upon recent advancements in reinforcement learning. More specifically, modelassisted batch reinforcement learning is applied to the setting of building climate control subjected to a dynamic pricing. The underlying sequential decision making problem is cast on a Markov decision problem, after which the control algorithm is detailed. In this work, fitted Q-iteration is used to construct a policy from a batch of experimental tuples. In those regions of the state space where the experimental sample density is low, virtual support tuples are added using an artificial neural network. Finally, the resulting policy is shaped using domain knowledge. The control approach has been evaluated quantitatively using a simulation and qualitatively in a living lab. From the quantitative analysis it has been found that the control approach converges in approximately 20 days to obtain a control policy with a performance within 90% of the mathematical optimum. The experimental analysis confirms that within 10 to 20 days sensible policies are obtained that can be used for different outside temperature regimes.
Abstract-This paper presents the design of a control strategy for the energy management of a grid-connected microgrid with local distributed energy resources as: 10-kW photovoltaic plant, 11-kW wind turbine, and 15-kW-190-kWh vanadium-based electric storage system. According to future regulations, the renewable energy producers will also have to provide a day-ahead hourly production plan. The overall idea is, by knowing the meteorological forecasts for the next 24 h, to dispatch the microgrid in order to be able to grant the scheduled hourly production by means of proper management of the storage system. The usage of the storage system is, however, minimized by the energy management strategy. The system design is validated by experimental testing carried out in SYSLAB, a distributed power system test facility at Risø Campus, Technical University of Denmark.
Abstract-The increasing share of distributed energy resources and renewable energy in power systems results in a highly variable and less controllable energy production. Therefore, in order to ensure stability and to reduce the infrastructure and operation cost of the power grid, flexible and controllable demand is needed. The research area of demand side management is still very much in flux and several options are being presented which can all be used to manage loads in order to achieve a flexible and more responsive demand. These different control schemes are developed with different organization of the power sector in mind and thus can differ significantly in their architecture, their integration into the various markets, their integration into distribution network operation and several other aspects. This paper proposes a classification of load control policies for demand side management in smart buildings, based on external behavior: direct, indirect, transactional and autonomous control; internal operation: decision support system scope, control strategy, failure handling and architecture. This classification assists in providing an overview of the control schemes as well as different ways of representing a building.
This paper describes the application of stochastic grey-box modeling to identify electrical power consumption-to-temperature models of a domestic freezer using experimental measurements. The models are formulated using stochastic differential equations (SDEs), estimated by maximum likelihood estimation (MLE), validated through the model residuals analysis and cross-validated to detect model over-fitting. A nonlinear model based on the reversed Carnot cycle is also presented and included in the modeling performance analysis. As an application of the models, we apply model predictive control (MPC) to shift the electricity consumption of a freezer in demand response experiments, thereby addressing the model selection problem also from the application point of view and showing in an experimental context the ability of MPC to exploit the freezer as a demand side resource (DSR).
Abstract-This paper presents the grey-box modeling of a vapor-compression refrigeration system for residential applications based on maximum likelihood estimation of parameters in stochastic differential equations. Models obtained are useful in the view of controlling refrigerators as flexible consumption units, which operation can be shifted within temperature and operational constraints. Even if the refrigerators are not intended to be used as smart loads, validated models are useful in predicting units consumption. This information can increase the optimality of the management of other flexible units, such as heat pumps for space heating, in order to smooth the load factor during peak hours, enhance reliability and efficiency in power networks and reduce operational costs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.