Accurate and efficient condition monitoring is the key to enhance the reliability and security of wind turbines. In recent years, an intelligent anomaly detection method based on deep learning networks has been receiving increasing attention. Since accurately labeled data are usually difficult to obtain in real industries, this paper proposes a novel Deep Small-World Neural Network (DSWNN) on the basis of unsupervised learning to detect the early failures of wind turbines. During network construction, a regular auto-encoder network with multiple restricted Boltzmann machines is first constructed and pre-trained by using unlabeled data of wind turbines. After that, the trained network is transformed into a DSWNN model by randomly add-edges method, where the network parameters are fine-tuned by using minimal amounts of labeled data. In order to guard against the changes and disturbances of wind speed and reduce false alarms, an adaptive threshold based on extreme value theory is presented as the criterion of anomaly judgment. The DSWNN model is excellent in depth mining data characteristics and accurate measurement error. Last, two failure cases of wind turbine anomaly detection are given to demonstrate its validity and accuracy of the proposed methodology contrasted with the deep belief network and deep neural network.

To meet the requirements on industrial robots concerning their fast-moving pick-and-place or obstacle avoidance, this article puts forward a novel high-speed and smooth transfer control algorithm for spatial elliptic trajectory - an modified S-shaped acceleration/deceleration (ACC/DEC) control algorithm based on a piecewise continuous jerk curve in order to achieve the smooth motion at a high speed. Through mathematical modeling and interpolation point algorithm of spatial elliptic trajectory, the interpolation point information can be obtained. Then, using the speed look-ahead algorithm, the optimal speed at the switching point can be calculated during the trajectory transfer. According to the analysis on the experiment and simulation results of a sorting case, the velocity curve turned out to be smoother and the run time decreased by 33.69%. In this way, this transfer algorithm can guarantee the smooth motion of industrial robots and also improve the motion efficiency significantly.

Ultrasonic transducer based on rare-earth giant magnetostrictive materials was designed in accordance with the technical requirements of ultrasonic surface strengthening. The whole structure of the transducer was designed. Modal analysis is made to get the natural frequency of the compound oscillator. The working frequency of the transducer should be guaranteed at about 15.2 kHz and the composite oscillator should have relatively better vibration mode. The magnetic field of the transducer is well sealed and the transducer will not show obvious magnetic flux leakage phenomenon. Which shows the rationality of structural design. Based on this transducer, the ultrasonic surface strengthening experiment on 40 steel was conducted. The surface roughness and hardness of the parts were analyzed after the experiment. The results show that the surface of the parts reach the mirror surface result after the ultrasonic strengthening. When compared to the previous process, the roughness decreases by about 75%, and the surface hardness increases by more than 20%.

In order to realize the automatic strengthening for turbine blades, a path planning method for robotic ultrasonic surface strengthening is proposed. A constitutive model of nonlinear isotropic strengthening–kinematic hardening is analyzed to establish the dynamic response model of ultrasonic surface strengthening on the turbine blade. According to the dynamic response model, the impact depth of the ultrasonic working head was obtained. Then, a path planning method of robotic ultrasonic surface strengthening for turbine blades is proposed on the basis of impact depth of working head, and it can improve both the uniformity of path distribution and contour accuracy. It not only ensures the processing accuracy but also meets the uniformity requirement of coverage. This path planning method provides a new surface strengthening technology for turbine blades.

In order to satisfy the requirement of the automatic ultrasonic strengthening of an aviation blade surface, this paper puts forward a robotic compliance control strategy of contact force for ultrasonic surface strengthening. By building the force/position control method for robotic ultrasonic surface strengthening., the compliant output of the contact force is achieved by using the robot’s end-effector (compliant force control device). Based on the control model of the end-effector obtained from experimental determination, a fuzzy neural network PID control is used to optimize the compliance control system, which improves the adjustment accuracy and tracking performance of the system. An experimental platform is built to verify the effectiveness and feasibility of the compliance control strategy for the robotic ultrasonic strengthening of an aviation blade surface. The results demonstrate that the proposed method maintains the compliant contact between the ultrasonic strengthening tool and the blade surface under multi-impact and vibration conditions.

In order to enhance the automation level and achieve high precision in the ultrasonic strengthening of aviation blade surfaces, this study focuses on investigating the intelligent control strategy and optimizing the machining parameters for robotic ultrasonic surface strengthening. By designing an intelligent compliance control method, the end-effector can achieve the compliant output of contact force. The fuzzy PID control method is used to optimize the regulation performance of the compliant force control system. This compliance control strategy enables the optimization of the compliance device, effectively improving the static and dynamic characteristics of the compliance controller. Based on this, an experimental method (RSM) is designed to analyze the interaction effects of contact force, feed rate, and repetition times on the surface quality of the blade. The optimal combination of robotic strengthening parameters is determined, providing a practical reference for the application of robotic compliance control in the ultrasonic strengthening of aviation blade surfaces.