Journal of Energy Storage · 2023 · 31 citations · 28 references
ComputingBuilding ScienceComputer ScienceArtificial IntelligenceEnergy ManagementMachine LearningBuilding Energy StorageDeep LearningEnergy StorageNeural NetworksReinforcement LearningRenewable EnergyAgent LearningDeep Learning For ForecastingDeep Learning-based ForecastingDeep Reinforcement LearningElectricity Consumption ForecastingEnergy PredictionEnergy Prediction ModelsPhotovoltaic Systems
The control of the electrical power supply is one of the key bases to reach the sustainable development goals set by United Nations. The achievement of these objectives encourages a dual strategy of creation and diffusion of renewable energies and other technologies of zero emission. Thus, meet the emerging necessities require, inevitably, a significant transformation of the building sector to improve the design of the electrical infrastructure. This improvement should be linked to advanced techniques that allows the identification of complex patterns in large amount of data, such as Deep Learning ones, in order to mitigate potential uncertainties. Accurate electricity and energy supply prediction models, in combination with storage systems will be reflected directly in efficiency improvements in buildings. In this paper, a branch of Deep Learning models, known as Standard Neural Networks, are used to predict electricity consumption and photovoltaic generation with the purpose of reduce the energy wasted, by managing the storage system using Reinforcement Learning technique. Specifically, Deep Reinforcement Learning is applied using the Deep Q-Learning agent. Furthermore, the accuracy of the predicted variables is measured by means of normalized Mean Bias Error (nMBE), and normalized Root Mean Squared Error (nRMSE). The methodologies developed are validated in an existing building, the School of Mining and Energy Engineering located on the Campus of the University of Vigo.
28
Deep Reinforcement Learning: A Brief Survey
Kai Arulkumaran, Marc Peter Deisenroth, Miles Brundage et al. · IEEE Signal Processing Magazine · 2017 · 4.3K citations
Efficient Processing of Deep Neural Networks: A Tutorial and Survey
Vivienne Sze, Yu‐Hsin Chen, Tien-Ju Yang et al. · Proceedings of the IEEE · 2017 · 4K citations
A State-of-the-Art Survey on Deep Learning Theory and Architectures
Md Zahangir Alom, Tarek M. Taha, Chris Yakopcic et al. · Electronics · 2019 · 1.6K citations · Full text