[20p-C205-1~12] CS.2 Code-sharing Session of 3.2 & 4.4

▲ [20p-C205-5] A machine learning approach for simultaneous measurement of temperature and strain using MSM

〇(D)KOUSTAV DEY¹, Vangety Nikhil¹, Anirban Majee¹, Sourabh Roy¹ (1.Nat. Inst. of Tech. WL)

Keywords:Machine Learning, Optical Fiber, Optical Sensor

In this paper, a novel and efficacious method for simultaneous measurement of temperature and strain by employing a multi-single-multi mode (MSM) fiber segment using an Artificial Neural network (ANN) has been reported and studied. An experiment is performed to accumulate the input dataset and employed in the ANN model for training and testing thoroughly. Captivating performance in terms of different metrics has been illustrated. The results ensure that our proposed model has attained enhancement in the root mean squared error (RMSE) of 1294 times and 301 times in comparison with the transfer matrix method for measuring the temperature and strain respectively. Furthermore, higher sensitivities are achieved using the proposed techniques. These primary findings suggest that the proposed scheme could be employed for various multiparameter sensing applications with enhanced performance capabilities.