Abstract

VAZQUEZ SEISDEDOS, Luis; TRUJILLO CODORNIU, Rafael A.; LLOSAS ALBUERNE, Yolanda  and  DIAZ MARTINEZ, David. Method to steady state detection: application to electrical generation units. EAC [online]. 2014, vol.35, n.2, pp. 45-61. ISSN 1815-5928.

Detecting windows or intervals of when a continuous process is operating in a state of steadiness is useful for long term monitoring and especially when steady-state models are being used to optimize the process or plant on-line. In this paper a method, which is called as sigma gamma, on moving windows based that improves significantly existent algorithms is presented. On the measurements, it combines algorithms based on standard deviation analysis with moving average method and it can be applied not only on measurements which have been corrupted by additive white noise; but also on time series which have been corrupted by additive colored noise. Its performance is evaluated by comparing with two very recent methods. For different levels and noise types, tests point out that proposed method offers a significantly reduction on Type I and Type II errors. The base time series for algorithm's testing are related with the essential waveform's response below electrical generation units' work.

Keywords : steady state detector; time series; hypothesis testing; colored noise; electrical generation unit.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )