Research on nonlinear compensation scheme of yarn tension sensor using SAW devices based on SSA–SVR model

Abstract

The detection accuracy of a yarn tension sensor using surface acoustic wave (SAW) devices has become increasingly important. We investigate a nonlinear compensation scheme based on SSA and SVR models to improve its detection accuracy, and the principle of SSA--SVR model and training method are also explored. We take the output frequency of the two sensors as input, the yarn tension applied to the working sensor as output, train an SSA--SVR model and use it for nonlinear compensation. We analyze and calculate the linearity, compensation accuracy and robustness of the SSA-SVR model, and compared it with the multiple regression model and BP neural network. The comparison results show that the SSA--SVR model has the best linearity, the highest compensation accuracy and the most robust. Finally, a novel nonlinear compensation scheme is proposed.

Accurate blood glucose level monitoring using microwave imaging

“…Accuracy (%) Support vector regression (SVR) [29] 89.2924 Random forest (RF) [30] 84.5236 Extreme learning machine (ELM) [31] 84.1132 General regression neural network (GRNN) [4] 74.4805 Proposed technique 91.9802…”

“…Microwave images reconstructed from VDSR-BL and conventional VDSR are extracted features, respectively. The obtained feature matrixes from the original microwave images, VDSR images, and VDSR-BL images are used in the SVR algorithm [29] for monitoring BGL, respectively. The 25 glucose concentrations with the range of 0.2-5 mg ml −1 are used as training set including 250 samples, and the 5 glucose concentrations with the range of 0.5-4.5 mg ml −1 are used as test set, which has 50 samples.…”

Nonlinear error compensation for 2D PSD based on SSA-BP neural network

“…SSA uses local search operations, which can accelerate the convergence speed of the algorithm and improve the quality of the optimal solution by selecting individuals within the population of local search. [10] These advantages of SSA can precisely solve the problems of slow iteration speed and easy getting stuck in local optimal solutions in BP neural networks.…”

“…n1 is the number of hidden layer nodes, n is the number of input layer nodes, m is the number of output layer nodes, and is a constant between [1,10]. According to the empirical formula, the range of hidden layer nodes in this model is between [3,12].…”