【摘要】 高低温试验是微波组件生产中的重要环节,对微波组件生产周期具有很大影响。围绕提升微波组件高低温试验一次通过率,采用神经网络算法及测试曲线包络线提取法对常温测试数据及高低温试验数据进行分析,建立常温测试数据与高低温试验数据之间的映射关系模型或对应区间范围,并嵌入测试软件进行应用,对常温测试调试进行指导,以提升微波组件高低温试验一次通过率。结果表明:基于数据分析的高低温性能预测式测试具有可行性和应用潜力,为缩短微波组件生产周期提供了一种可选的新方式。更多还原

【Abstract】 High and low temperature testing plays a key role during the manufacturing of microwave modules,which has a great impact on the production cycle of microwave module.Aiming at enhancing the first pass yield of high and low temperature testing,the neural network algorithm and the enveloping line extracting method are adopted in the data analysis for defining the relationship between the room temperature RF testing results and high and low temperature testing results.Based on the data analysis,the approximate relationship model and the instructing zone are obtained and integrated into the software.The application of the software in the manufacturing of microwave modules indicates the feasibility of predicating the high and low temperature testing results,which paves a new road for reducing the delivery time of microwave modules.更多还原