Research on error correction model of surface acoustic wave yarn tension transducer based on DOA–SVR model

Abstract

We present an error correction model for improving the accuracy of Surface Acoustic Wave Yarn Tension (SAWYT) transducer measurements. The model utilizes Support Vector Regression (SVR) and Dingo Optimization Algorithm (DOA). Significantly, the input comprises the oscillation frequencies from the co-located measurement and calibration transducers. The model is trained using the yarn tension that bore on the measurement transducer as the output. As a Subsequently, for the training set, the DOA–SVR model achieves a mean square error (MSE) of 1.4999×10−5 and a determination coefficient (R2) of 0.99997, surpassing the other two models. On the test set, the DOA–SVR model continues to excel with an MSE of 3.8371×10−5 and an R2 of 0.99990, outperforming the other models. These results highlight the superior performance of the DOA–SVR model in error correction for SAWYT transducers, making it as the preferred choice for both the training and test sets.