Method for predicting service life of epoxy resin adhesive through small sample data

• Main Authors: FAN YE, YANG LU, ZHOU JIE, CHEN JINHU, LIU YUFENG, DE XINLIN, ZHEN JIANGYUN, ZHU YIPENG, NING WEIWEI, XU YUZHEN, ZHAO WEI, YANG XUEYIN, LIU DEJUN
• Format: Patent
• Language: Chinese; English
• Published: 25.03.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; PHYSICS

The invention belongs to the field of reliability accelerated test methods, and particularly relates to a method for predicting the service life of an epoxy resin adhesive by small sample data, which comprises the following steps: 1) determining test conditions, test samples and time; 2) measuring tensile shear strength test data and recording; 3) performing normalization processing on the data in the step 2); 4) fitting an epoxy resin adhesive degradation track model; 5) expanding the sample data; and 6) substituting the expanded sample data into the BP neural network for iteration to obtain a neural network training result, and finally obtaining a predicted value of the storage life of the epoxy resin adhesive. According to the method, a new thought is developed for ensuring the precision of an evaluation result under the condition that test data is small, and compared with a traditional test method, a large amount of time and expenditure are saved. 本发明属于可靠性加速试验方法领域，具体涉及一种小样本数据预测环氧树脂胶粘剂寿命的方法，采用下述步骤：1）确定试验条件