The recurrence plot (RP) method has been introduced into bearing fault diagnosis due to its capability of effectively analyzing nonlinear and non-stationary waveform signals in dynamic systems. However, the interference of noise increases the difficulty of RP-based fault diagnosis. To solve this problem, this paper proposed a novel anti-noise bearing fault diagnosis method based on improved RP and a convolutional neural network. First, different scales of approximation coefficients and detail coefficients were obtained and constructured for RP based on wavelet packet decomposition (WPD) on the vibrational signal. Meanwhile, redundant parts of each RP were removed according to its symmetry characteristics, and the remaining parts of these RPs were spliced into multi-scale asymmetric RP (MARP) containing all coefficients. Then, a fault diagnosis model for rolling bearing was established with MARP as the input of the pre-trained ResNet-34. Finally, the validity of the proposed fault diagnosis method was validated on the Paderborn bearing dataset. Experimental results showed that the proposed fault diagnosis method achieved an accuracy of 90% under Gaussian white noise with a signal-to-noise ratio of above -6 dB.

LuGre dynamic friction model has been widely used in servo system for friction compensation, but it increases the difficulty of controller design because its parameters are difficult to be identified and its internal state is immeasurable. This paper presents a parameter identification technique based on novel evolutionary algorithm (NEA) for LuGre friction model. In order to settle the practical digital implementation problem of LuGre model, this paper also proposes a modified dual-observer with discontinuous mapping and smooth transfer function. On the basis of the parameter identification results and the modified dual-observer, this paper designs an adaptive control algorithm with dynamic friction compensation for hydraulic servo system. The comparative experiments indicate that the proposed parameter identification technique and the adaptive control algorithm with modified dual-observer are effective with high tracking performance.

Superficial flow velocity perception is inspired by aquatic organisms and has become increasingly important for underwater robots and vehicles. Existing marine electromagnetic velometers require very large AC coils with high power consumption for excitation, limiting their mobility and underwater applications. Permanent magnet excitation is preferable for miniature flow sensors because of low power consumption, small size, and high sensitivity. However, permanent magnet excitation has not been applicable due to the high output drift caused by electrode polarization voltage. This paper proposes a permanent magnet-based flow velocity meter that measures the velocity based on the electrode output current instead of the output voltage. The relationship between the output current and the fluid velocity, fluid conductivity, permanent magnet parameters, and electrode position during permanent magnet excitation is derived, and a finite element simulation is conducted. A signal processing circuit is designed to obtain the current output. The experimental results show that the output drift ratio of the proposed sensor is 92% lower than that of the sensor measuring the voltage output, and the measurement R ² is 0.998 in the test range of 0-0.875m/s.