液压支架疲劳寿命近似估算

发布时间：2018-05-27 16:09

本文选题：液压支架 + 疲劳寿命　；参考：《工矿自动化》2017年03期

【摘要】：针对常规的液压支架寿命近似分析方法需对危险点进行循环加载和获取,导致计算机负载增加的问题,提出了一种基于遗传算法与BP神经网络的寿命估算模型。利用遗传算法的全局搜索性优化BP神经网络,使其不易陷入局部最小点;利用优化后的BP神经网络建立危险点结构参量到疲劳寿命的网络映射模型。针对样本容量和隐含层节点数进行了测试,测试结果表明,样本容量为40、隐含层节点数为7时,模型估算精度较高;液压支架平均寿命估算值为36 456次,与理论值的最大相对误差为5.27%。
[Abstract]:In order to solve the problem that the conventional approximate analysis method of the life of hydraulic support needs to load and obtain the dangerous point in circulation, which leads to the increase of computer load, a life estimation model based on genetic algorithm and BP neural network is proposed. The global search of genetic algorithm is used to optimize BP neural network to make it difficult to fall into the local minimum point, and the optimized BP neural network is used to establish the network mapping model from the structural parameters of dangerous points to fatigue life. The test results show that the sample size is 40, the hidden layer node number is 7, the model estimation accuracy is high, the average life of hydraulic support is 36,456 times. The maximum relative error with the theoretical value is 5.27.
【作者单位】：南京工业大学机械与动力工程学院;
【基金】：“十二五”国家科技支撑计划项目(2013BAF02B11)
【分类号】：TD355.4

【参考文献】