Piezo-electrical transducer (PZT) driven Fabry-Perot filter (F-P) filters are widely employed to implement a fast speed and high-resolution wavelength interrogation. Under a certain voltage, the transmission band of FFP can be tuned by changing the voltage applied on PZT. For the ideal design of the tunable filter actuation, the displacement–voltage relationship of PZT is assumed to be linear. However, temperature changes make the piezoelectric constant and dielectric constant of the material gradually change, and temperature drift leads to deterioration of the demodulation performance of tunable FabryPerot filters. Traditional calibration approaches are mainly based on F-P etalon and absorption lines of inert gas, but these approaches dramatically increase system cost and complexity. Considering the time series characteristics of temperature drift data, this paper proposes long and short-term memory (LSTM) neural network to compensate temperature drift of FP filters. During the cooling process of the F-P filter, the temperature and the drift of the reference fiber Bragg grating (FBG) are employed as the features of the LSTM model to characterize the wavelength drift of the sensing FBG. The experiment results show that after the wavelength is compensated with LSTM, the wavelength drift is reduced to 8.45 pm, while the compensated wavelength drift with the least square support vector machine (LSSVM) is 15.51 pm. Compared with LSSVM, LSTM is more suitable in long-term temperature-changing environments. Additionally, no additional hardware is required and the whole C band is covered in the proposed method.
Fiber Fabry-Perot filter (FFP-TF) is one of the key components of the fiber Bragg grating (FBG) demodulation system. Its main principle is to realize wavelength scanning with the inverse piezoelectric effect of piezoelectric ceramics (PZT), but the inherent hysteresis and creep characteristics of PZT make the relationship curve between the transmission wavelength of FFP-TF and the control voltage of the PZT unable. Furthermore, in the temperature-varying environment, the relationship between the transmission wavelength and the control voltage keeps drifting. Aiming at the temperature-induced wavelength drift problem of the tunable optical filter, this paper proposed an improved least square support vector machine (LSSVM) model to capture the internal law of the transmission wavelength drift with temperature, and the BAS-PSO algorithm is employed to search penalty factor and nuclear parameters. Experimental results show that after the optimized least squares support vector machine compensates for the tunable filter's sweep fluctuations, the temperature drift error of the tunable filter is ±0.77 pm, and the standard deviation is 0.35 pm, which improves the temperature stability of the tunable filter demodulation in a variable temperature environment.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.