Chiller Load Forecasting Using Hyper-Gaussian Nets

Energy load forecasting for optimization of chiller operation is a topic that has been receiving increasing attention in recent years. From an engineering perspective, the methodology for designing and deploying a forecasting system for chiller operation should take into account several issues regarding prediction horizon, available data, selection of variables, model selection and adaptation. In this paper these issues are parsed to develop a neural forecaster. The method combines previous ideas such as basis expansions and local models. In particular, hyper-gaussians are proposed to provide spatial support (in input space) to models that can use auto-regressive, exogenous and past errors as variables, constituting thus a particular case of NARMAX modelling. Tests using real data from different world locations are given showing the expected performance of the proposal with respect to the objectives and allowing a comparison with other approaches.

[1]  Mohammad Taghi Hamidi Beheshti,et al.  A local linear radial basis function neural network for financial time-series forecasting , 2010, Applied Intelligence.

[2]  Manuel R. Arahal,et al.  A prediction model based on neural networks for the energy consumption of a bioclimatic building , 2014 .

[3]  Fred Collopy,et al.  Automatic Identification of Time Series Features for Rule-Based Forecasting , 2001 .

[4]  Maria del Carmen Pegalajar Jiménez,et al.  An Application of Non-Linear Autoregressive Neural Networks to Predict Energy Consumption in Public Buildings , 2016 .

[5]  Pedro Antonio Gutiérrez,et al.  MELM-GRBF: A modified version of the extreme learning machine for generalized radial basis function neural networks , 2011, Neurocomputing.

[6]  Jukka Heikkonen,et al.  Time Series Predicition using Recurrent SOM with Local Linear Models , 1997 .

[7]  Ashwin M. Khambadkone,et al.  Energy optimization methodology of multi-chiller plant in commercial buildings , 2017 .

[8]  Eduardo Dulce-Chamorro,et al.  Parsimonious Modelling for Estimating Hospital Cooling Demand to Improve Energy Efficiency , 2021, Log. J. IGPL.

[9]  Jiejin Cai,et al.  Applying support vector machine to predict hourly cooling load in the building , 2009 .

[10]  Yu Gong,et al.  Online Modeling With Tunable RBF Network , 2013, IEEE Transactions on Cybernetics.

[11]  Narasimhan Sundararajan,et al.  An efficient sequential learning algorithm for growing and pruning RBF (GAP-RBF) networks , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[12]  C. L. Philip Chen,et al.  Exploiting the interpretability and forecasting ability of the RBF-AR model for nonlinear time series , 2016, Int. J. Syst. Sci..

[13]  Abbas Khosravi,et al.  A review on artificial intelligence based load demand forecasting techniques for smart grid and buildings , 2015 .

[14]  Ruey S. Tsay,et al.  Functional-Coefficient Autoregressive Models , 1993 .

[15]  Qiuming Zhu,et al.  A global learning algorithm for a RBF network , 1999, Neural Networks.

[16]  E. Nadaraya.  On Estimating Regression , 1964 .

[17]  Jiwen Dong,et al.  Time-series prediction using a local linear wavelet neural network , 2006, Neurocomputing.

[18]  Abdullatif Ben-Nakhi,et al.  Cooling load prediction for buildings using general regression neural networks , 2004 .

[19]  Eric C. Rouchka,et al.  Reduced HyperBF Networks: Regularization by Explicit Complexity Reduction and Scaled Rprop-Based Training , 2011, IEEE Transactions on Neural Networks.

[20]  Manas Ranjan Senapati,et al.  An adaptive local linear optimized radial basis functional neural network model for financial time series prediction , 2015, Neural Computing and Applications.

[21]  Guillermo Bejarano,et al.  Suboptimal hierarchical control strategy to improve energy efficiency of vapour-compression refrigeration systems , 2017 .

[22]  Luis Fontan,et al.  Very Short-Term Load Forecaster Based on a Neural Network Technique for Smart Grid Control , 2020, Energies.

[23]  M Martell,et al.  Multiobjective control architecture to estimate optimal set points for user comfort and energy saving in buildings. , 2020, ISA transactions.

[24]  João M. Lemos,et al.  Modelling and cooling power control of a TES-backed-up vapour-compression refrigeration system , 2020 .

[25]  John C. Platt.  A Resource-Allocating Network for Function Interpolation , 1991, Neural Computation.

[26]  Sung-Bae Cho,et al.  Time Series Forecasting with Multi-Headed Attention-Based Deep Learning for Residential Energy Consumption , 2020, Energies.

[27]  Min Gan,et al.  A locally linear RBF network-based state-dependent AR model for nonlinear time series modeling , 2010, Inf. Sci..

[28]  M. J. D. Powell,et al.  Radial basis functions for multivariable interpolation: a review , 1987 .

[29]  Donald F. Specht,et al.  A general regression neural network , 1991, IEEE Trans. Neural Networks.

[30]  Jee-Heon Kim,et al.  Modeling and Optimizing a Chiller System Using a Machine Learning Algorithm , 2019, Energies.

[31]  Sheng Chen,et al.  Practical identification of NARMAX models using radial basis functions , 1990 .

[32]  F. Girosi,et al.  Networks for approximation and learning , 1990, Proc. IEEE.

[33]  Jason Runge,et al.  Forecasting Energy Use in Buildings Using Artificial Neural Networks: A Review , 2019, Energies.