Residual Life Prediction of Rotating Machinery Guided by Quantum Deep Neural Network

Main Article Content

Ganghong Ye
Ningxuan Shi


In order to avoid the low prediction rate and high evaluation rate in estimating the service balance life of rotating machinery, this paper presents a quantum gene chain encoded bidirectional neural network (QGCCBNN) for estimating the service balance life of rotating machinery. In QGCCBNN, quantum bidirectional transmission mechanism has been developed. In order to improve the global optimization ability and convergence speed, we have developed a quantum gene chain encoding method to transform the gradient descent into the data transmission and updating. Because of the advantages of QGCCBNN in nonlinear estimation ability and convergence speed, the proposed QGCCBNN for predicting the remaining service life of rotating machinery can achieve higher prediction precision and optimization.The predicted value of the proposed method for the remaining service life of double row roller bearings is 6.33h (actual value is 7.17h), with a prediction error of only 0.84h and a relative prediction error of only 11.72%. The experimental results demonstrate the effectiveness of the proposed method.

Article Details

Special Issue - Deep Learning-Based Advanced Research Trends in Scalable Computing