Processing math: 100%
System Identification With Fourier Transformation for Long-Term Time Series Forecasting | IEEE Journals & Magazine | IEEE Xplore

System Identification With Fourier Transformation for Long-Term Time Series Forecasting


Abstract:

Time-series prediction has drawn considerable attention during the past decades fueled by the emerging advances of deep learning methods. However, most neural network bas...Show More

Abstract:

Time-series prediction has drawn considerable attention during the past decades fueled by the emerging advances of deep learning methods. However, most neural network based methods fail in extracting the hidden mechanism of the targeted physical system. To overcome these shortcomings, an interpretable sparse system identification method without any prior knowledge is proposed in this study. This method adopts the Fourier transform to reduces the irrelevant items in the dictionary matrix, instead of indiscriminate usage of polynomial functions in most system identification methods. It shows an visible system representation and greatly reduces computing cost. With the adoption of l_{1} norm in regularizing the parameter matrix, a sparse description of the system model can be achieved. Moreover, three data sets including the water conservancy data, global temperature data and financial data are used to test the performance of the proposed method. Although no prior knowledge was known about the physical background, experimental results show that our method can achieve long-term prediction regardless of the noise and incompleteness in the original data more accurately than the widely-used baseline data-driven methods. This study may provide some insight into time-series prediction investigations, and suggests that a white-box system identification method may extract the easily overlooked yet inherent periodical features and may beat neural-network based black-box methods on long-term prediction tasks.
Published in: IEEE Transactions on Big Data ( Volume: 11, Issue: 2, April 2025)
Page(s): 474 - 484
Date of Publication: 30 May 2024

ISSN Information:

Funding Agency:


I. Introduction

Accurate time-series prediction is of crucial importance in the era of Big Data, which is widely used in many fields of modern industry, such as prediction on energy consumption, weather change, disease spreading, traffic flow and economic evolution [1], [2], [3], [4], [5]. It has attracted widespread attention of scientists ranging from mathematics, computer science, and other related specific engineering fields. Therein, a large number of achievements have been obtained and show appreciable performance. However, long (e.g., more than one thousand steps) and stable prediction has still been a tricky but significant problem to untangle due to the lack of deep understanding of the hidden mechanism in the targeted physical system [6]. Moreover, most existing methods are proposed in a deep learning framework, which cannot meets the requirement of reliable mechanism to be used in large-scale industrial systems. This motivates us to further investigate the mechanism of the system evolution with a white-box format and get rid of the usage of a black-box neural network module.

Contact IEEE to Subscribe

References

References is not available for this document.