A Comparison of Energy Consumption Prediction Models Based on Neural Networks of a Bioclimatic Building

Khosravani, H.; Castilla, M.; Berenguel, Manuel; Ruano, A. E.; Ferreira, Pedro

doi:10.3390/en9010057

Cited by 99 publications

(47 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Most of the studies using artificial neural networks for power demand forecasting have used data from residential buildings, commercial or office buildings. Experiments on solar powered buildings have also been conducted [12]. These experiments used power demand data from business days, non-business days, and seasonal data.…”

Section: Power Demand Forecasting Using Deep Learningmentioning

confidence: 99%

A Hybrid Neural Network Model for Power Demand Forecasting

Kim

Choi

Jeon³

et al. 2019

Energies

View full text Add to dashboard Cite

The problem of power demand forecasting for the effective planning and operation of smart grid, renewable energy and electricity market bidding systems is an open challenge. Numerous research efforts have been proposed for improving prediction performance in practical environments through statistical and artificial neural network approaches. Despite these efforts, power demand forecasting problems remain to be a grand challenge since existing methods are not sufficiently practical to be widely deployed due to their limited accuracy. To address this problem, we propose a hybrid power demand forecasting model, called (c, l)-Long Short-Term Memory (LSTM) + Convolution Neural Network (CNN). We consider the power demand as a key value, while we incorporate c different types of contextual information such as temperature, humidity and season as context values in order to preprocess datasets into bivariate sequences consisting of <Key, Context[1, c]> pairs. These c bivariate sequences are then input into c LSTM networks with l layers to extract feature sets. Using these feature sets, a CNN layer outputs a predicted profile of power demand. To assess the applicability of the proposed hybrid method, we conduct extensive experiments using real-world datasets. The results of the experiments indicate that the proposed (c, l)-LSTM+CNN hybrid model performs with higher accuracy than previous approaches.

show abstract

Section: Power Demand Forecasting Using Deep Learningmentioning

confidence: 99%

A Hybrid Neural Network Model for Power Demand Forecasting

Kim

Choi

Jeon³

et al. 2019

Energies

View full text Add to dashboard Cite

show abstract

“…The analyzed case-study corresponded to a micro-grid integrated in the center for solar energy research (CIESOL) building, a bioclimatic building designed to research renewable energies ( Figure 2a) located at the University of Almeria (Almeria, Spain) 36.83 N, 2.40 W. This placement, with an annual average sunshine duration of almost 3000 h, is very suitable for solar energy applications and the CIESOL building counts with both active solar heating and cooling [17]. In addition to on-going efforts in analyzing and characterizing the overall power consumption of the building [18,19], this work aimed to control and optimize the building parts power management by a conscious design and operation of micro-grids to be scaled-up to an integrated building scheme. To achieve this, first, a laboratory on the first floor of the building was selected (Figure 2b) as the reference case for operating a micro-grid consisting of a dedicated photovoltaic and storage facility where the EV was also considered as part of the proposed configuration.…”

Section: The Attached Problemmentioning

confidence: 99%

Energy Management Strategy for Micro-Grids with PV-Battery Systems and Electric Vehicles

et al. 2018

View full text Add to dashboard Cite

This paper analyzes the impact of photovoltaic (PV) systems on storage and electric vehicles in micro-grids. As these kinds of systems are becoming increasingly popular in the residential sector, the development of a new generation of equipment, such as more efficient batteries or solar panels, makes further study necessary. These systems are especially interesting in commercial or office buildings, since they have a more repetitive daily pattern of electricity consumption, which usually occurs within the maximum solar radiation hours. Based on this need, a novel control strategy aimed at efficiently managing this kind of micro-grid is proposed. The core of this strategy is a rule-based controller managing the power flows between the grid and the batteries of both the PV system and the electric vehicle. Through experimental data and simulations, this strategy was tested under different scenarios. The selected testbed consisted of the laboratory of a research center, which could be easily scalable to the entire building. Results showed the benefits of using an electric vehicle as an active agent in energy balance, leading to a reduction of the energetic costs of a micro-grid.Energies 2018, 11, 522 2 of 13 residential buildings is presented in [2], where energy storage and load management were presented as the main strategies for increasing self-consumption. The latest developments in battery technology have enabled their use in residential houses. In [3], self-consumption was quantified for a PV system with a home battery. Due to different electricity taxes, market tariffs, and stochastic consumption and production in Europe, that paper did not ensure that the battery was to be profitable in today's market. However, increasing electricity prices and the expected decrease in battery prices suggest that energy storage could be a profitable option in the near future, even in Northern Europe. A methodology for sizing residential PV battery systems was proposed in [4], and a control algorithm for these kinds of systems were proposed in [5,6]. Both works dealt with the limitations in the power flows between the PV system, the public electrical gird, and batteries. In combination with energy storage, intelligent load management may address important improvements in the self-consumption of buildings [7,8]. This strategy basically consists of scheduling the controllable loads to match the hours with the maximum production of solar energy. In commercial and office buildings where the loads and solar energy production occur at the same time, it is likely to reach high self-consumption [9]. However, in houses whose occupants spend the central part of the day outside, a mismatch between production and consumption occurs. In this scenario, active demand-side management (DSM) allows the improvement of self-consumption by selecting when a water heater, a washing machine, or other programmable appliances should start working.Taking into account the concepts of energy storage and DSM, the plug-in electric vehicle (EV) arises as a c...

show abstract

“…Artificial neural networks (ANNs) are one of the most widely used black box models, and they are achieving good results in a great variety of problems, including the prediction of energy consumption of a building. Some of the problems of energy estimation solved with ANN are energy test bench in buildings [17][18][19][20], electric power prediction [21][22][23][24], and heating/cooling consumption prediction [25][26][27][28].…”

Section: Introductionmentioning

confidence: 99%

Modeling of Energy Efficiency for Residential Buildings Using Artificial Neuronal Networks

Álvarez

Rabuñal

García-Vidaurrázaga

et al. 2018

Advances in Civil Engineering

View full text Add to dashboard Cite

Increasing the energy efficiency of buildings is a strategic objective in the European Union, and it is the main reason why numerous studies have been carried out to evaluate and reduce energy consumption in the residential sector. The process of evaluation and qualification of the energy efficiency in existing buildings should contain an analysis of the thermal behavior of the building envelope. To determine this thermal behavior and its representative parameters, we usually have to use destructive auscultation techniques in order to determine the composition of the different layers of the envelope. In this work, we present a nondestructive, fast, and cheap technique based on artificial neural network (ANN) models that predict the energy performance of a house, given some of its characteristics. The models were created using a dataset of buildings of different typologies and uses, located in the northern area of Spain. In this dataset, the models are able to predict the U-opaque value of a building with a correlation coefficient of 0.967 with the real U-opaque measured value for the same building.

show abstract

A Comparison of Energy Consumption Prediction Models Based on Neural Networks of a Bioclimatic Building

Cited by 99 publications

References 54 publications

A Hybrid Neural Network Model for Power Demand Forecasting

A Hybrid Neural Network Model for Power Demand Forecasting

Energy Management Strategy for Micro-Grids with PV-Battery Systems and Electric Vehicles

Modeling of Energy Efficiency for Residential Buildings Using Artificial Neuronal Networks

Contact Info

Product

Resources

About