Search

The diagnosis performance of Deep Convolutional Neural Network (DCNN) method is closely related to the generalization ability of the training model. An empirical training strategy is to randomly disperse the training samples and train the model with mini-batch training samples. But there are still two problems in the empirical method that need to be solved urgently. Firstly, what is the theoretical basis for random discretization of samples? Secondly, how to scientifically quantify batch division? Aiming at these two problems, the theoretical basis of sample random discretization has been deduced and proved, furthermore, a scientific quantitative batch division method is proposed based on the proved thesis. The fault diagnosis results of the planetary gearbox show that: (1) The model obtained by the training guide proposed in this paper has stronger generalization ability; (2) The DCNN with the training guide can accurately and effectively diagnose the faults of planetary gearbox and obtain ideal diagnosis results.

During the working process of the turbine, some types of faults can cause changes in the vibration characteristics of the blades. The real-time vibration monitoring of the blades is of great significance to the stable operation and state-based maintenance. Torsional vibration is a kind of blade vibration modes and results in failures such as cracks easily. Thus, it is important to measure it due to the harmfulness of torsional vibration. Firstly, the principle of blade tip timing (BTT) is introduced, and the models of the blade are built to analyze the characteristics of torsional vibration. Then, the compressed sensing theory is introduced, and its related parameters are determined according to the measurement requirements. Next, based on the condition that the blade rigidity axis is not bent and bent, respectively, the layout method of sensors is proposed and the numerical simulation of the measurement process is performed. The results of the above two types of numerical simulation verify the proposed measurement method. Finally, by analyzing the influencing factors of measurement uncertainty, the optimization method of sensors’ layout is further proposed. This study can provide important theoretical guidance for the measurement of blade torsional vibration.

According to the physiological structure of the hand index finger and the motion characteristics of each joint, the movement of the hand index finger during the maintenance operation was observed and analyzed. By measuring the movement trajectory of the index finger based on the binocular stereo vision camera, the motion angles of each joint were calculated by the index finger kinematics model, and then the rotation angles of each joint were fitted and simulated with the robot toolbox, comparing and analyzed the simulation trace and the experimental trace. The results show that the rotation angle function of each joint can simulate the movement of a single index finger of the manipulator, which plays an important role in the motion simulation and control of the intelligent maintenance manipulator.

Monitoring the vibrations of high-speed rotating blades is significant to the security of turbomachineries. Blade tip timing (BTT) is considered as a promising technique for detecting blade vibrations without contact online. However, extracting blade vibration characteristics accurately from undersampled BTT signals measured at varying rotational speed (VRS) has become a big challenge. The existing two methods for this issue are restricted within the order bandwidth limitation and require prior information and precise sensor installation angles, which is often unpractical. To overcome these difficulties, a compressed sensing-based order analysis (CSOA) method was proposed. Its feasibility comes from the sparsity of BTT vibration signals in the order domain. The mathematical model for the proposed method was built, and the optimizing principles for sensor number and sensor arrangement were given. Simulated and experimental results verified the feasibility and advantages of the proposed method that it could extract order spectrum accurately from BTT vibration signals measured at VRS without the drawbacks in the existing two methods.

Blade tip-timing measurements (BTT) are increasingly used in the monitoring of aero-engine blades. BTT signal is heavily under-sampled, which is difficult to recover. In order to solve this problem, the sparse reconstruction algorithm has been applied in many studies. However, the speed of sparse reconstruction for the BTT signal has been rarely mentioned till now. In this paper, for the purpose of improving computation speed, an improved sparse reconstruction algorithm is proposed Besides, a series of simulation results are performed to verify that with higher accuracy, the improved method can significantly shorten computing time compared with traditional sparse reconstruction algorithm. Therefore, the improved method can enhance accuracy and real-time performance of BTT, for which its advantage of online measurement can be better shown.

Blade tip timing (BTT) technique has become a promising way of online damage diagnosis for compressor and turbine blades by monitoring their vibrations. However, there are some disadvantages in previous algorithms for BTT signals, including not high precision, strict sensor layout, signal bandwidth limit, and weak anti-noise ability. To solve these problems, a novel interpolation method for wideband signal reconstruction (IWSR) is presented. Simulations are conducted using a lumped parameter model, which shows that IWSR reconstructs noiseless signals with overall errors mainly below 1%, and identifies the exact frequencies of 5 dB noise signals. In addition, the results validate its signal separation function and verify that its sensor layout is almost arbitrary, and its maximum allowable bandwidth is half the product of sensor number and rotational frequency. Experimental results demonstrate its effectiveness in frequency identification. In conclusion, IWSR provides precise information for real-time blade damage diagnosis, which benefits turbomachinery safety.