Hybrid Method for Forecasting Next Values of Time Series for Intelligent Building Control

  • Andrzej Stachno
  • Andrzej JablonskiEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9520)


The method for forecasting successive values of time series with the application of Artificial Neural Networks and Moving Window Fourier takes into account additional parameters. Such parameters are determined by an external observer based on an analysis of the accuracy of forecasts obtained. For the purpose of automating the above method, a parameter selection module based on decision trees has been proposed. In this way the Hybrid Method for Forecasting (HMF) successive values of time series has been created.


Artificial neuron networks FFT Forecasting Successive values of a time series Moving Window Fourier Environmental measurements in an intelligent building 


  1. 1.
    Jabłoński, A.: Intelligent buildings as distributed information systems. Int. J. Comput. Anticip. Syst. 21, 385–394 (2008). (Liege)Google Scholar
  2. 2.
    Stachno, A., Suproniuk, M.: Environmental performance measurement system designed for forecasting in intelligent buildings, Przegląd Elektrotechniczny, R89, 9, pp. 152–155 (2013). ISSN 0033-2097Google Scholar
  3. 3.
    Peitgen, H.O., Jurgens, H., Saupe, D.: Introduction to Fractals and Chaos. PWN (2002)Google Scholar
  4. 4.
    Stachno, A., Jablonski, A.: Application of artificial neuron networks and hurst exponent to forecasting of successive values of a time series representing environmental measurements in an intelligent building. In: Moreno-Díaz, R., Pichler, F., Quesada-Arencibia, A. (eds.) EUROCAST. LNCS, vol. 8111, pp. 483–490. Springer, Heidelberg (2013)CrossRefGoogle Scholar
  5. 5.
    Mitchell, T.M.: Machine Learning. McGraw-Hill, New York (1997)zbMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Faculty of ElectronicsWroclaw University of TechnologyWroclawPoland

Personalised recommendations