Sensitivity analysis to reduce duplicated features in ANN training for district heat demand prediction

Abstract Artificial neural network (ANN) has become an important method to model the nonlinear relationships between weather conditions, building characteristics and its heat demand. Due to the large amount of training data required for ANN training, data reduction and feature selection are important to simplify the training. However, in building heat demand prediction, many weather-related input variables contain duplicated features. This paper develops a sensitivity analysis approach to analyse the correlation between input variables and to detect the variables that have high importance but contain duplicated features. The proposed approach is validated in a case study that predicts the heat demand of a district heating network containing tens of buildings at a university campus. The results show that the proposed approach detected and removed several unnecessary input variables and helped the ANN model to reduce approximately 20% training time compared with the traditional methods while maintaining the prediction accuracy. It indicates that the approach can be applied for analysing large number of input variables to help improving the training efficiency of ANN in district heat demand prediction and other applications.

[1]  Hadi Dowlatabadi,et al.  Sensitivity and Uncertainty Analysis of Complex Models of Disease Transmission: an HIV Model, as an Example , 1994 .

[2]  Gustavo Camps-Valls,et al.  Unbiased sensitivity analysis and pruning techniques in neural networks for surface ozone modelling , 2005 .

[3]  Melvin Robinson,et al.  Prediction of residential building energy consumption: A neural network approach , 2016 .

[4]  Douglas M. Hawkins,et al.  Plackett–Burman technique for sensitivity analysis of many-parametered models , 2001 .

[5]  Kamaruzzaman Sopian,et al.  Artificial neural network modeling and analysis of photovoltaic/thermal system based on the experimental study , 2019, Energy Conversion and Management.

[6]  Julian D. Olden,et al.  Illuminating the “black box”: a randomization approach for understanding variable contributions in artificial neural networks , 2002 .

[7]  Yixuan Wei,et al.  Prediction of occupancy level and energy consumption in office building using blind system identification and neural networks , 2019, Applied Energy.

[8]  Frédéric Magoulès,et al.  Data Mining and Machine Learning in Building Energy Analysis , 2016 .

[9]  Brian Vad Mathiesen,et al.  A review of computer tools for analysing the integration of renewable energy into various energy systems , 2010 .

[10]  Betul Bektas Ekici,et al.  Prediction of building energy consumption by using artificial neural networks , 2009, Adv. Eng. Softw..

[11]  Y. Noorollahi,et al.  Using artificial neural networks for temporal and spatial wind speed forecasting in Iran , 2016 .

[12]  Nora El-Gohary,et al.  A review of data-driven building energy consumption prediction studies , 2018 .

[13]  Jaydev Sharma,et al.  Comparison of feature selection techniques for ANN-based voltage estimation , 2000 .

[14]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[15]  A. T. C. Goh,et al.  Back-propagation neural networks for modeling complex systems , 1995, Artif. Intell. Eng..

[16]  Frédéric Magoulès,et al.  A review on the prediction of building energy consumption , 2012 .

[17]  Daniel Friedrich,et al.  Demand Forecasting for a Mixed-Use Building Using Agent-Schedule Information with a Data-Driven Model , 2020, Energies.

[18]  Ji-Hoon Jang,et al.  Prediction of optimum heating timing based on artificial neural network by utilizing BEMS data , 2019, Journal of Building Engineering.

[19]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[20]  Miroslav Kubat,et al.  An Introduction to Machine Learning , 2015, Springer International Publishing.

[21]  Chirag Deb,et al.  Forecasting diurnal cooling energy load for institutional buildings using Artificial Neural Networks , 2016 .

[22]  Moon Keun Kim,et al.  Predicting electricity consumption in a building using an optimized back-propagation and Levenberg–Marquardt back-propagation neural network: Case study of a shopping mall in China , 2018, Sustainable Cities and Society.

[23]  Yacine Rezgui,et al.  Operational supply and demand optimisation of a multi-vector district energy system using artificial neural networks and a genetic algorithm , 2019, Applied Energy.

[24]  A. J. Morris,et al.  Non-linear projection to latent structures revisited (the neural network PLS algorithm) , 1999 .

[25]  Jacek M. Zurada,et al.  Sensitivity analysis for minimization of input data dimension for feedforward neural network , 1994, Proceedings of IEEE International Symposium on Circuits and Systems - ISCAS '94.

[26]  Nii O. Attoh-Okine,et al.  Analysis of learning rate and momentum term in backpropagation neural network algorithm trained to predict pavement performance , 1999 .

[27]  Mohammad Yusri Hassan,et al.  A review on applications of ANN and SVM for building electrical energy consumption forecasting , 2014 .

[28]  Yannis Dimopoulos,et al.  Use of some sensitivity criteria for choosing networks with good generalization ability , 1995, Neural Processing Letters.

[29]  M. Gevrey,et al.  Two-way interaction of input variables in the sensitivity analysis of neural network models , 2006 .

[30]  Francesco Martellotta,et al.  On the use of artificial neural networks to model household energy consumptions , 2017 .

[31]  M. Scardi.  Artificial neural networks as empirical models for estimating phytoplankton production , 1996 .

[32]  Nelson Fumo,et al.  A review on the basics of building energy estimation , 2014 .

[33]  Nikolaos Kourentzes,et al.  Feature selection for time series prediction - A combined filter and wrapper approach for neural networks , 2010, Neurocomputing.

[34]  Hideo Tanaka,et al.  Interval regression analysis by quadratic programming approach , 1998, IEEE Trans. Fuzzy Syst..

[35]  Francesco Grimaccia,et al.  Analysis and validation of 24 hours ahead neural network forecasting of photovoltaic output power , 2017, Math. Comput. Simul..

[36]  Luca A. Tagliafico,et al.  Heating and cooling building energy demand evaluation; a simplified model and a modified degree days approach , 2014 .

[37]  Monto Mani,et al.  Dynamic thermal performance of conventional and alternative building wall envelopes , 2019, Journal of Building Engineering.

[38]  Yeonsook Heo,et al.  Calibration of building energy models for retrofit analysis under uncertainty , 2012 .

[39]  Chen Liu,et al.  Artificial neural network aided real-time simulation of electric traction system , 2020 .

[40]  James Yu,et al.  District Heating Network Demand Prediction Using a Physics-Based Energy Model with a Bayesian Approach for Parameter Calibration , 2019, Energies.

[41]  Francesco Grimaccia,et al.  A Physical Hybrid Artificial Neural Network for Short Term Forecasting of PV Plant Power Output , 2015 .

[42]  H. T. Ozkahraman,et al.  The use of tuff stone cladding in buildings for energy conservation , 2006 .

[43]  Sylvain Robert,et al.  State of the art in building modelling and energy performances prediction: A review , 2013 .

[44]  I. Dimopoulos,et al.  Application of neural networks to modelling nonlinear relationships in ecology , 1996 .

[45]  Andrew H. Sung,et al.  Ranking importance of input parameters of neural networks , 1998 .

[46]  A. Regattieri,et al.  Artificial neural network optimisation for monthly average daily global solar radiation prediction , 2016 .

[47]  E. Martin,et al.  Non-linear projection to latent structures revisited: the quadratic PLS algorithm , 1999 .

[48]  Brian Henderson-Sellers,et al.  Sensitivity evaluation of environmental models using fractional factorial experimentation , 1996 .

[49]  Huidong Gu,et al.  Selecting the correct weighting factors for linear and quadratic calibration curves with least-squares regression algorithm in bioanalytical LC-MS/MS assays and impacts of using incorrect weighting factors on curve stability, data quality, and assay performance. , 2014, Analytical chemistry.

[50]  Youngdeok Hwang,et al.  Artificial neural network model for forecasting sub-hourly electricity usage in commercial buildings , 2016 .

[51]  Rajesh Kumar,et al.  Energy analysis of a building using artificial neural network: A review , 2013 .

[52]  Neal Wade,et al.  Design and analysis of electrical energy storage demonstration projects on UK distribution networks , 2015 .

[53]  Andrew Hunter,et al.  Application of neural networks and sensitivity analysis to improved prediction of trauma survival , 2000, Comput. Methods Programs Biomed..