The transient and dynamic loading accuracy of the valve controlled cylinder force loading system of the undercarriage actuator cylinder wear and life experiment platform is low, which cannot meet the accuracy requirements of the load spectrum, thus affecting the safety and reliability judgment of the actuator. An improved nonlinear active disturbance rejection control (INADRC) algorithm with higher accuracy and anti-interference ability is proposed based on control algorithm. First, the AMESim/Simulink co-simulation model of the electro-hydraulic servo force loading system is established. Secondly, in order to optimize its parameters, the INADRC controller is designed, and the genetic particle swarm algorithm is used. Finally, the performance of the controller is verified by simulating and experiment with three target signal tracking. The simulation and experimental results show that compared with PID control, nonlinear ADRC (NADRC) and other improved nonlinear ADRC (ONADRC), the average accuracy of the INADRC is improved by 4.15%, 1.15% and 0.65%, which reflects the characteristics of high servo force transient, dynamic loading accuracy and strong anti-interference ability.

Rolling bearing is a critical component in the rotating machinery, which directly affects the reliability of the equipment. The artificial intelligence-enabled bearing fault diagnosis model has achieved impressive successes over the years. However, rolling bearings’ imbalanced data sets (normal samples are much larger than failure samples) degrade the diagnostic performance. To address this issue, a bidirectional generative adversarial network(BiGAN) based fault diagnosis method was proposed. First, the signal was denoised via the ensemble empirical mode decomposition(EEMD) to automatically distribute it to a suitable reference scale and avoid modal aliasing. Then, the BiGAN model with gradient penalty term was constructed to expand the fault samples, where the min-max normalization was included. Finally, based on the enhanced training set, the convolutional neural network was established with batch normalization and maximum pooling layers. Experimental results proved that the proposed method improved fault diagnosis accuracy and robustness.