Energy Flexibility Prediction for Data Center Engagement in Demand Response Programs

In this paper, we address the problem of the efficient and sustainable operation of data centers (DCs) from the perspective of their optimal integration with the local energy grid through active participation in demand response (DR) programs. For DCs’ successful participation in such programs and for minimizing the risks for their core business processes, their energy demand and potential flexibility must be accurately forecasted in advance. Therefore, in this paper, we propose an energy prediction model that uses a genetic heuristic to determine the optimal ensemble of a set of neural network prediction models to minimize the prediction error and the uncertainty concerning DR participation. The model considers short term time horizons (i.e., day-ahead and 4-h-ahead refinements) and different aspects such as the energy demand and potential energy flexibility (the latter being defined in relation with the baseline energy consumption). The obtained results, considering the hardware characteristics as well as the historical energy consumption data of a medium scale DC, show that the genetic-based heuristic improves the energy demand prediction accuracy while the intra-day prediction refinements further reduce the day-ahead prediction error. In relation to flexibility, the prediction of both above and below baseline energy flexibility curves provides good results for the mean absolute percentage error (MAPE), which is just above 6%, allowing for safe DC participation in DR programs.

[1]  Yi Wang,et al.  Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges , 2018, IEEE Transactions on Smart Grid.

[2]  Yi Liang,et al.  Short term load forecasting based on feature extraction and improved general regression neural network model , 2019, Energy.

[3]  Ping-Huan Kuo,et al.  A Short-Term Wind Speed Forecasting Model by Using Artificial Neural Networks with Stochastic Optimization for Renewable Energy Systems , 2018, Energies.

[4]  Kuo-Ping Lin,et al.  Short-Term Wind Power Prediction Based on Improved Chicken Algorithm Optimization Support Vector Machine , 2019, Sustainability.

[5]  Jing Ma,et al.  Decomposition-Based Dynamic Adaptive Combination Forecasting for Monthly Electricity Demand , 2019, Sustainability.

[6]  Iniyan Selvarasan,et al.  Short-Term Forecasting of Total Energy Consumption for India-A Black Box Based Approach , 2018, Energies.

[7]  Gustavo Rau de Almeida Callou,et al.  An Artificial Neural Network Approach to Forecast the Environmental Impact of Data Centers , 2019, Inf..

[8]  Ilari Alapera,et al.  Data centers as a source of dynamic flexibility in smart girds , 2018, Applied Energy.

[9]  Ling Liu,et al.  A Hybrid Neural Network Model for Power Demand Forecasting , 2019, Energies.

[10]  Georges Da Costa,et al.  Green IT scheduling for data center powered with renewable energy , 2018, Future Gener. Comput. Syst..

[11]  Gabriela Hug,et al.  Forecasting of Smart Meter Time Series Based on Neural Networks , 2016, DARE@PKDD/ECML.

[12]  Stefan Feuerriegel,et al.  Integration scenarios of Demand Response into electricity markets: Load shifting, financial savings and policy implications , 2016 .

[13]  Hongseok Kim,et al.  Data-Driven Baseline Estimation of Residential Buildings for Demand Response , 2015 .

[14]  Jun Zhang,et al.  Learning-based power prediction for data centre operations via deep neural networks , 2016, E2DC@e-Energy.

[15]  Xiaofei Wang,et al.  Dynamic Resource Prediction and Allocation for Cloud Data Center Using the Multiobjective Genetic Algorithm , 2018, IEEE Systems Journal.

[16]  Do-Hyeun Kim,et al.  A Prediction Methodology of Energy Consumption Based on Deep Extreme Learning Machine and Comparative Analysis in Residential Buildings , 2018, Electronics.

[17]  Ping-Huan Kuo,et al.  An Electricity Price Forecasting Model by Hybrid Structured Deep Neural Networks , 2018 .

[18]  Joseph Rynkiewicz,et al.  Asymptotic statistics for multilayer perceptron with ReLu hidden units , 2019, ESANN.

[19]  Dan Andersson,et al.  A review of data centers as prosumers in district energy systems: Renewable energy integration and waste heat reuse for district heating , 2020 .

[20]  Marcel Antal,et al.  Data center optimization methodology to maximize the usage of locally produced renewable energy , 2015, 2015 Sustainable Internet and ICT for Sustainability (SustainIT).

[21]  Yang Zhao,et al.  Deep learning-based feature engineering methods for improved building energy prediction , 2019, Applied Energy.

[22]  Sungwon Jung,et al.  Prediction Model Based on an Artificial Neural Network for User-Based Building Energy Consumption in South Korea , 2019, Energies.

[23]  Kostas Kalaitzakis,et al.  Development of Demand Response Energy Management Optimization at Building and District Levels Using Genetic Algorithm and Artificial Neural Network Modelling Power Predictions , 2018, Energies.

[24]  Evangelos Spiliotis,et al.  Statistical and Machine Learning forecasting methods: Concerns and ways forward , 2018, PloS one.

[25]  Birgitte Bak-Jensen,et al.  Opportunities and challenges of demand response in active distribution networks , 2018 .

[26]  L. Nilsson,et al.  Data centres in future European energy systems—energy efficiency, integration and policy , 2019, Energy Efficiency.

[27]  Mauro Conti,et al.  Computational intelligence approaches for energy load forecasting in smart energy management grids: state of the art, future challenges, and research directionsand Research Directions , 2018 .

[28]  Ali Ouni,et al.  Optimal Deep Learning LSTM Model for Electric Load Forecasting using Feature Selection and Genetic Algorithm: Comparison with Machine Learning Approaches † , 2018, Energies.

[29]  Nicolò Rossetto.  Measuring the intangible : an overview of the methodologies for calculating customer baseline load in PJM , 2018 .

[30]  Emmanuel O. Ogedengbe,et al.  Optimization of energy performance with renewable energy project sizing using multiple objective functions , 2019 .

[31]  Jun Hu,et al.  Learning-based data analytics: Moving towards transparent power grids , 2018 .

[32]  Nicholas Good,et al.  Review and classification of barriers and enablers of demand response in the smart grid , 2017 .

[33]  Marcel Antal,et al.  Optimized flexibility management enacting Data Centres participation in Smart Demand Response programs , 2018, Future Gener. Comput. Syst..

[34]  Marcel Antal,et al.  Transforming Data Centers in Active Thermal Energy Players in Nearby Neighborhoods , 2018 .

[35]  Yibo Chen,et al.  Short-term prediction of electric demand in building sector via hybrid support vector regression , 2017 .

[36]  Giha Lee,et al.  Application of Long Short-Term Memory (LSTM) Neural Network for Flood Forecasting , 2019, Water.

[37]  Nadeem Javaid,et al.  Electricity Price and Load Forecasting using Enhanced Convolutional Neural Network and Enhanced Support Vector Regression in Smart Grids , 2019, Electronics.

[38]  Wan He,et al.  Load Forecasting via Deep Neural Networks , 2017, ITQM.

[39]  Florin Pop,et al.  Exploiting data centres energy flexibility in smart cities: Business scenarios , 2019, Inf. Sci..

[40]  Xi Zhang,et al.  Forecasting the Energy Embodied in Construction Services Based on a Combination of Static and Dynamic Hybrid Input-Output Models , 2019, Energies.