篇名 | Safety Risk Assessment of Shooting Test by Deep Learning Neural Networks |
---|---|
卷期 | 31:5 |
作者 | Chao Song 、 Yi-Zhuo Jia 、 Dong-Jun Wang 、 Hong-Tian Liu 、 Yang Cao |
頁次 | 277-289 |
關鍵字 | modeling 、 neural nets 、 risk analysis 、 safety management 、 shooting test 、 EI 、 MEDLINE 、 Scopus |
出刊日期 | 202010 |
DOI | 10.3966/199115992020103105021 |
To improve the safety of shooting tests in military tests, the safety risks of shooting tests are evaluated by the deep learning quantum gate circuit model, and the analysis of the vulnerability elements of the safety risk assessment is strengthened. The current safety management mechanism for shooting tests is analyzed. The practical experience of safety management is combined to build a quantum gate circuit neural network (QGCNN) model. The quantum revolving gate is utilized to control the qubit inversion and the phase deflection. The comprehensive risk value of shooting test safety is calculated. Simulation experiments have confirmed the reliability and effectiveness of the model, and the proposed model is compared with the traditional back propagation neural network (BPNN). The experimental results show that the proposed QGCNN has 28, 16, 14 and 12 iteration steps at different learning rates; for the output risk, the minimum error is 0.0025, and the maximum error is 0.0172, respectively; the performance of the proposed model is better than that of traditional BPNN. The constructed QGCNN model has a higher convergence rate for the safety risk assessment of shooting tests, which reduces the complexity of data processing, improves the accuracy of risk prediction.