Tuning hyperparameters of a SVM-based water demand forecasting system through parallel global optimization

Abstract Recently, the number of machine learning based water demand forecasting solutions has been significantly increasing. Different case studies have already reported practical results proving that accurate forecasts may support optimization of operations in Water Distribution Networks (WDN). However, tuning the hyper-parameters of machine leaning algorithms is still an open problem. This paper proposes a parallel global optimization model to optimize the hyperparameters of Support Vector Machine (SVM) regression trained to provide accurate water demand forecasts in the short-time horizon (i.e. 24 h). Every SVM has the first 6 hourly water consumptions as input features and a specific hourly water demand as target to be predicted, among the remaining 18. The Mean Average Percentage Error (MAPE), computed on leave-one-out validation, is the black-box objective function optimized. Moreover, a preliminary time-series clustering has been applied in order to evaluate if this can improve the accuracy of the forecasting mechanism. Time-series clustering implies that the overall number of SVMs, whose hyperparameters are optimized through parallel global optimization, increases, with a SVM trained for each cluster identified and for each hourly water demand to be predicted, making even more critical a quick tuning of the hyperparameters. Results on the urban water demand data in Milan prove that forecasting error is significantly low and that preliminary clustering allows for further reducing error while also improving computational performances.

[1]  Isam Shahrour,et al.  Urban water demand forecasting using the stochastic nature of short term historical water demand and supply pattern , 2013 .

[2]  Francesco Archetti,et al.  Short-term forecasting of hourly water consumption by using automatic metering readers data , 2015 .

[3]  Panos M. Pardalos,et al.  Stochastic subgradient descent method for large-scale robust chance-constrained support vector machines , 2017, Optim. Lett..

[4]  Iosif Meyerov,et al.  SVM Regression Parameters Optimization Using Parallel Global Search Algorithm , 2013, PaCT.

[5]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[6]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[7]  Bernd Bischl,et al.  Effectiveness of Random Search in SVM hyper-parameter tuning , 2015, 2015 International Joint Conference on Neural Networks (IJCNN).

[8]  Yin Wang,et al.  Dynamic Forecast of Daily Urban Water Consumption Using a Variable-Structure Support Vector Regression Model , 2015 .

[9]  Y. D. Sergeyev,et al.  Global Optimization with Non-Convex Constraints - Sequential and Parallel Algorithms (Nonconvex Optimization and its Applications Volume 45) (Nonconvex Optimization and Its Applications) , 2000 .

[10]  Alexander J. Smola,et al.  Learning with Kernels: support vector machines, regularization, optimization, and beyond , 2001, Adaptive computation and machine learning series.

[11]  Hyun-Han Kwon,et al.  Development of Ensemble Model Based Water Demand Forecasting Model , 2014 .

[12]  Victor P. Gergel,et al.  Parallel global optimization on GPU , 2016, Journal of Global Optimization.

[13]  Maamar Sebri,et al.  Forecasting urban water demand: A meta-regression analysis. , 2016, Journal of environmental management.

[14]  Francesco Archetti,et al.  Automatic Configuration of Kernel-Based Clustering: An Optimization Approach , 2017, LION.

[15]  Aaron Klein,et al.  Efficient and Robust Automated Machine Learning , 2015, NIPS.

[16]  Alberto Bemporad,et al.  Water demand forecasting for the optimal operation of large-scale drinking water networks: The Barcelona Case Study. , 2014 .

[17]  Antonio Candelieri.  Clustering and Support Vector Regression for Water Demand Forecasting and Anomaly Detection , 2017 .

[18]  T. Mazzuchi,et al.  Urban Water Demand Forecasting: Review of Methods and Models , 2014 .

[19]  Jiaqi Liu,et al.  A novel clustering method on time series data , 2011, Expert Syst. Appl..

[20]  Jasper Snoek,et al.  Practical Bayesian Optimization of Machine Learning Algorithms , 2012, NIPS.

[21]  Nando de Freitas,et al.  Taking the Human Out of the Loop: A Review of Bayesian Optimization , 2016, Proceedings of the IEEE.

[22]  Gwo-Fong Lin,et al.  An Hourly Streamflow Forecasting Model Coupled with an Enforced Learning Strategy , 2015 .

[23]  Joaquín Izquierdo,et al.  Predictive models for forecasting hourly urban water demand , 2010 .

[24]  T. Warren Liao,et al.  Clustering of time series data - a survey , 2005, Pattern Recognit..

[25]  Yuan Ren,et al.  Determination of Optimal SVM Parameters by Using GA/PSO , 2010, J. Comput..

[26]  Francesco Archetti,et al.  Identifying Typical Urban Water Demand Patterns for a Reliable Short-term Forecasting – The Icewater Project Approach , 2014 .

[27]  Panos M. Pardalos,et al.  Robust chance-constrained support vector machines with second-order moment information , 2018, Ann. Oper. Res..

[28]  Helena Mala-Jetmarova,et al.  Lost in optimisation of water distribution systems? A literature review of system operation , 2017, Environ. Model. Softw..

[29]  Luuk C. Rietveld,et al.  Improving the Performance of Water Demand Forecasting Models by Using Weather Input , 2014 .

[30]  V. Kavitha,et al.  Clustering Time Series Data Stream - A Literature Survey , 2010, ArXiv.

[31]  Jingcheng Wang,et al.  Urban water demand forecasting by LS-SVM with tuning based on elitist teaching-learning-based optimization , 2014, The 26th Chinese Control and Decision Conference (2014 CCDC).