No abstract

Although standard iterative learning control (ILC) approaches can achieve perfect tracking for active magnetic bearing (AMB) systems under external disturbances, the disturbances are required to be iteration-invariant. In contrast to existing approaches, we address the tracking control problem of AMB systems under iteration-variant disturbances that are in different channels from the control inputs. A disturbance observer based ILC scheme is proposed that consists of a universal extended state observer (ESO) and a classical ILC law. Using only output feedback, the proposed control approach estimates and attenuates the disturbances in every iteration. The convergence of the closed-loop system is guaranteed by analyzing the contraction behavior of the tracking error. Simulation and comparison studies demonstrate the superior tracking performance of the proposed control approach.

Electromagnetic sensors have been used for inspecting small surface defects of metals. Based on the eddycurrent thin-skin regime, a revised algorithm is proposed for a triple-coil drive-pickup eddy-current sensor scanning over long surface crack slots (10 mm) with different rotary angles. The method is validated by the voltage measurement of the designed EC sensor scanning over a benchmark (ferromagnetic) steel with surface defects of different depths and rotary angles. With an additional sensing coil for the designed EC sensor, the defect angle (or orientation) can be measured without spatially and coaxially rotating the excitation coil. By referring to the voltage change (due to the defect) diagram (voltage sum versus voltage different) of two sensing pairs, the rotary angle of the surface crack is retrieved with a maximum residual deviation of 3.5 %.

Remaining useful life (RUL) of cutting tools is concerned with cutting tool operational status prediction and damage prognosis. Most RUL prediction methods utilized different features collected from different sensors to predict the life of the tool. To increase the prediction accuracy, it is often necessary to mount a great deal of sensors on the machine in order to collect more types of signals, which can heavily increase the cost in industrial applications. To deal with this issue, this study, for the first time, proposed a new feature network dictionary, which can enlarge the number of candidate features under limited sensor conditions, and the developed dictionary can potentially contain as much useful information as possible. This process can replace the installation of more sensors and incorporate more information. Then, the sparse augmented Lagrangian (SAL) feature selection method is proposed to reduce the number of candidate features and select the most significant features. Finally, the selected features are input to the Gaussian Process Regression (GPR) model for the RUL estimation. Extensive experiments demonstrate that our proposed RUL estimation framework output performs traditional methods, especially for the cost savings for on-line RUL estimation.

Electromagnetic sensors have been used for inspecting small surface defects of metals. Based on the eddy-current thin-skin regime, a revised algorithm is proposed for a triple-coil drive-pickup eddy-current sensor scanning over long surface crack slots (10 mm) with different rotary angles. The method is validated by the voltage measurement of the designed EC sensor scanning over a benchmark (ferromagnetic) steel with surface defects of different depths and rotary angles. With an additional sensing coil for the designed EC sensor, the defect angle (or orientation) can be measured without spatially and coaxially rotating the excitation coil. By referring to the voltage change (due to the defect) diagram (voltage sum versus voltage different) of two sensing pairs, the rotary angle of the surface crack is retrieved with a maximum residual deviation of 3.5 %.