Advertisement

A Forecasting Framework Based on Kalman Filter Integrated Multivariate Local Polynomial Regression: Application to Urban Water Demand

  • Guoqiang Chen
  • Tianyu Long
  • Yun BaiEmail author
  • Jin Zhang
Article
  • 31 Downloads

Abstract

In this study, a forecasting framework for daily urban water demand has been proposed. It was developed based on the extended Kalman filter (EKF) which consists of state estimation, forecasting and error correction. The forecasting and error correction models can be substituted. As an example, a multivariate local polynomial regression (MLPR) was used to linearize the complex system which is essential for EKF. A correctional prediction of residual based on relevance vector regression was employed to update and substitute error estimation value in the EKF. To improve the precision of the forecasts, the historical data series was decomposed into low- and high-frequency subseries using discrete wavelet transformation. Five category forecasts with the lead time of 1-day were assessed in comparison of the proposed model: MLPR, multi-scale relevance vector regression, autoregressive moving average, Back Propagation neural network and multiple linear regression. According to the performance criteria, the MLPR is slightly beneficial in capturing the basic dynamics of the daily urban water demand in the short term, but the state estimation and error correction can greatly improve the results. The proposed model obtains better forecasting performances than existing models, which is attributed to good state estimation from the Kalman transmission gain and favorable feature learning performance using MLPR.

Keywords

Kalman filter Multivariate local polynomial regression Water demand Forecast Relevance vector regression 

Notes

Acknowledgements

This work is supported by the Natural Science Foundation of China (71801044), the Humanities and Social Science Foundation of Ministry of Education of China (17YJC630003), the Natural Science Foundation of Chongqing (cstc2018jcyjAX0436), and the Research Start-Up Funds of CTBU (1756012). In addition, the authors would also like to thank the editors/reviewers for their valuable suggestions and comments that have helped improve this paper.

Author Contributions

TL and YB conceived and designed the experiments; GC performed the experiments, analyzed the data and wrote the paper; JZ contributed to the scientific discussion and editing. The authors thank JX for supplying the dataset.

Compliance with Ethical Standards

Conflicts of interest

The authors declare no conflict of interest.

References

  1. 1.
    Adamowski J, Chan FH, Prasher SO, Ozga-Zielinski B, Sliusarieva A (2012) Comparison of multiple linear and nonlinear regression, autoregressive integrated moving average, artificial neural network, and wavelet artificial neural network methods for urban water demand forecasting in Montreal, Canada. Water Resour Res 48:W01528.  https://doi.org/10.1029/2010WR009945 CrossRefGoogle Scholar
  2. 2.
    Bai Y, Wang P, Li C, Xie J, Wang Y (2014) A multi-scale relevance vector regression approach for daily urban water demand forecasting. J Hydrol 517:236–245.  https://doi.org/10.1016/j.jhydrol.2014.05.033 CrossRefGoogle Scholar
  3. 3.
    Bai Y, Wang P, Li C, Xie J, Wang Y (2015) Dynamic forecast of daily urban water consumption using a variable-structure support vector regression model. J Water Resour Plann Manage 141:04014058.  https://doi.org/10.1061/(ASCE)WR.1943-5452.0000457 CrossRefGoogle Scholar
  4. 4.
    Chen G, Long T, Xiong J, Bai Y (2017) Multiple random forests modelling for urban water consumption forecasting. Water Resour Manage.  https://doi.org/10.1007/s11269-017-1774-7 Google Scholar
  5. 5.
    Ding F, Liu X, Ma X (2016) Kalman state filtering based least squares iterative parameter estimation for observer canonical state space systems using decomposition. J Comput Appl Math 301:135–143.  https://doi.org/10.1016/j.cam.2016.01.042 MathSciNetCrossRefzbMATHGoogle Scholar
  6. 6.
    Emmanuel AD, Thomas AM, Refik S, Roberson JA (2014) Urban water demand forecasting: review of methods and models. J Water Resour Plann Manage 140:146–159.  https://doi.org/10.1061/(ASCE)WR.1943-5452.0000314 CrossRefGoogle Scholar
  7. 7.
    Evensen G (1994) Sequential data assimilation with a nonlinear quasi-geostrophic model using Monte Carlo methods to forecast error statistics. J Geophys Res 99:10143–10162.  https://doi.org/10.1029/94JC00572 CrossRefGoogle Scholar
  8. 8.
    Eynard J, Grieu S, Polit M (2011) Wavelet-based multi-resolution analysis and artificial neural networks for forecasting temperature and thermal power consumption. Eng Appl Artif Intell 24:501–516.  https://doi.org/10.3390/w8050197 CrossRefGoogle Scholar
  9. 9.
    Fan J, Gijbels I (1996) Local polynomial modelling and its applications. Chapman and Hall/CRC, LondonzbMATHGoogle Scholar
  10. 10.
    Julier SJ, Uhlmann JK, Durrant-Whyte H (1995) A new approach for filtering nonlinear systems. In: Proceedings of the “American Control Conference”, Seattle, WA, pp 1628–1632.  https://doi.org/10.1109/acc.1995.529783
  11. 11.
    Kalman RE (1960) A new approach to linear filtering and prediction problems. J Basic Eng 82:35–45CrossRefGoogle Scholar
  12. 12.
    Kim HS, Eykholt R, Salas JD (1999) Nonlinear dynamics, delay times, and embedding windows. Physica D 127:48–60. https://doi.org/10.1016/S0167-2789(98)00240-1 CrossRefzbMATHGoogle Scholar
  13. 13.
    Lee S, Chang J (2016) On using multivariate polynomial regression model with spectral difference for statistical model-based speech enhancement. J Syst Architect 64:76–85.  https://doi.org/10.1016/j.sysarc.2015.10.007 CrossRefGoogle Scholar
  14. 14.
    Mallat GS (1989) A theory for multi resolution signal decomposition: the wavelet representation. Trans Pattern Anal Mach Intell 11:674–693CrossRefzbMATHGoogle Scholar
  15. 15.
    Masry E (1997) Multivariate regression estimation: local polynomial fitting for time series. Nonlinear Anal Theory Methods Appl 30:3575–3581.  https://doi.org/10.1016/S0304-4149(96)00095-6 MathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
    Messaci F, Nemouchi N, Ouassou I, Rachdi M (2015) Local polynomial modelling of the conditional quantile for functional data. Stat Methods Appl 24:597–622.  https://doi.org/10.1007/s10260-015-0296-9 MathSciNetCrossRefzbMATHGoogle Scholar
  17. 17.
    Mouatadid S, Adamowski J (2017) Using extreme learning machines for short-term urban water demand forecasting. Urban Water J 14:630–638.  https://doi.org/10.1080/1573062X.2016.1236133 CrossRefGoogle Scholar
  18. 18.
    Rosenstein MT, Collins JJ, De Luca CJ (1994) Reconstruction expansion as a geometry-based framework for choosing proper delay times. Physica D 73:82–89.  https://doi.org/10.1016/0167-2789(94)90226-7 MathSciNetCrossRefGoogle Scholar
  19. 19.
    Ruppert D, Wand PM (1994) Multivariate locally weighted least squares regression. Ann Stat 22:1346–1370.  https://doi.org/10.1214/aos/1176325632 MathSciNetCrossRefzbMATHGoogle Scholar
  20. 20.
    Shafaei M, Kisi O (2016) Lake level forecasting using wavelet-SVR, wavelet-ANFIS and wavelet-ARMA conjunction models. Water Resour Manag 30:79–97.  https://doi.org/10.1007/s11269-015-1147-z CrossRefGoogle Scholar
  21. 21.
    Stone CJ (1977) Consistent nonparametric regression. Ann Stat 5:595–620.  https://doi.org/10.1214/aos/1176343886 MathSciNetCrossRefzbMATHGoogle Scholar
  22. 22.
    Su L (2010) Prediction of multivariate chaotic time series with local polynomial fitting. Comput Math Appl 59:737–744.  https://doi.org/10.1016/j.camwa.2009.10.019 MathSciNetCrossRefzbMATHGoogle Scholar
  23. 23.
    Sun L, Nistor I, Seidou O (2015) Streamflow data assimilation in SWAT model using extended Kalman filter. J Hydrol 531:671–684.  https://doi.org/10.1016/j.jhydrol.2015.10.060 CrossRefGoogle Scholar
  24. 24.
    Valenzuela O, Pasadas M (2011) Fuzzy data approximation using smoothing cubic splines: similarity and error analysis. Appl Math Model 35:2122–2144.  https://doi.org/10.1016/j.apm.2010.11.046 MathSciNetCrossRefzbMATHGoogle Scholar
  25. 25.
    Wang X, Jiang S, Yin J (2012) Data sharpening methods in multivariate local quadratic regression. J Multivar Anal 105:258–275.  https://doi.org/10.1016/j.jmva.2011.09.004 MathSciNetCrossRefzbMATHGoogle Scholar
  26. 26.
    Wishner R, Tabaczynski J, Athans M (1969) A comparison of three non-linear filters. Automatica 5:487–496.  https://doi.org/10.1016/0005-1098(69)90110-1 CrossRefzbMATHGoogle Scholar
  27. 27.
    Zamani A, Azimian A, Heemink A, Solomatine D (2010) Non-linear wave data assimilation with an ANN-type wind-wave model and Ensemble Kalman Filter (EnKF). Appl Math Model 34:1984–1999.  https://doi.org/10.1016/j.apm.2009.10.013 MathSciNetCrossRefzbMATHGoogle Scholar
  28. 28.
    Zhang H, Hu J, Zou L et al (2018) Event-based state estimation for time-varying stochastic coupling networks with missing measurements under uncertain occurrence probabilities. Int J Gen Syst 47(5):506–521.  https://doi.org/10.1080/03081079.2018.1445740 MathSciNetCrossRefGoogle Scholar
  29. 29.
    Zheng X, Fang H (2015) An integrated unscented Kalman filter and relevance vector regression approach for lithium-ion battery remaining useful life and short-term capacity prediction. Reliab Eng Syst Saf 144:74–82.  https://doi.org/10.1016/j.ress.2015.07.013 CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Urban and Rural Construction Committee of Beibei DistrictChongqingChina
  2. 2.Faculty of Urban Construction and Environmental EngineeringChongqing UniversityChongqingChina
  3. 3.National Research Base of Intelligent Manufacturing ServiceChongqing Technology and Business UniversityChongqingChina
  4. 4.Institute of Groundwater and Earth SciencesJinan UniversityGuangzhouChina

Personalised recommendations