|
| GA-BP神经网络在AgCuNi电接触材料的性能预测研究 |
| Optimal design of AgCuNi series electrical contact materials based on GA-BP neural network |
| Received:August 12, 2021 |
| DOI： |
| 中文关键词: AgCuNi系电接触材料 遗传算法 BP神经网络 硬度 导电率 |
| 英文关键词: AgCuNi series electrical contact material genetic algorithm BP neural network hardness conductivity |
| 基金项目:国家科技部“科技助力经济2020”重点专项；云南省稀贵金属材料基因工程(202002AB080001-1)；云南省2019材料基因工程项目(2019ZE001-2)；云南省科技人才与平台计划(202105AE160027) |
| Author Name | Affiliation | E-mail | | ZHANG Nai-qian* | AECC Shenyang Liming AERO-engine Co.Ltd., Shenyang 110043, China | powder@ipm.com.cn | | WEI Ming-xia | AECC Shenyang Liming AERO-engine Co.Ltd., Shenyang 110043, China | | | ZHAO Jun | State Key Laboratory of Advanced Technologies for Comprehensive Utilization of Platinum Metals, Sino Platinum Metals Co.Ltd., Kunming 650106, China | | | WANG Jian-ping | State Key Laboratory of Advanced Technologies for Comprehensive Utilization of Platinum Metals, Sino Platinum Metals Co.Ltd., Kunming 650106, China | | | YANG You-cai | State Key Laboratory of Advanced Technologies for Comprehensive Utilization of Platinum Metals, Sino Platinum Metals Co.Ltd., Kunming 650106, China | | | ZHAO Tong-ming | State Key Laboratory of Advanced Technologies for Comprehensive Utilization of Platinum Metals, Sino Platinum Metals Co.Ltd., Kunming 650106, China | | | FANG Ji-heng | State Key Laboratory of Advanced Technologies for Comprehensive Utilization of Platinum Metals, Sino Platinum Metals Co.Ltd., Kunming 650106, China | | | XIE Ming | State Key Laboratory of Advanced Technologies for Comprehensive Utilization of Platinum Metals, Sino Platinum Metals Co.Ltd., Kunming 650106, China | |
|
| Hits: 1201 |
| Download times: 665 |
| 中文摘要: |
| 本文针对现有AgCuNi系电接触材料对硬度和导电率预测方法不足等问题，采用遗传算法对BP神经网络的权值和阈值优化，加快了算法的收敛速度，建立了基于遗传算法优化的BP神经网络AgCuNi系电接触材料的硬度和导电率预测模型。训练精度和实际测试精度分别达到了0.98和0.89，误差9.18%。研究结果表明，本文建立的BP神经网络预测模型，有助于提高合金的成分设计效率，提高银合金电接触材料的开发效率。 |
| 英文摘要: |
| In this paper, aiming at the problems of insufficient prediction methods of hardness and conductivity of AgCuNi series electrical contact materials, the weight and threshold of BP neural network are optimized by genetic algorithm, which speeds up the convergence speed of the algorithm, and the prediction model of hardness and conductivity of AgCuNii series electrical contact materials based on BP neural network optimized by genetic algorithm is established. The training accuracy and actual test accuracy are 0.98 and 0.89 respectively, with an error of 9.18%. The results show that the BP neural network prediction model established in this paper is helpful to improve the efficiency of alloy composition design and the development efficiency of silver alloy electrical contact materials. |
|
View Full Text
View/Add Comment Download reader |
| Close |
|
|
|