论文部分内容阅读
为实现对飞机维修人员安全性的有效预测,构建了维修人员安全性评价指标。通过专家打分法获得了维修人员安全性评价指标的原始训练数据和测试数据,分别采用BP-GDM模型、BP-LM模型和GA-BP-LM模型对数据进行了训练和预测,仿真结果表明:BP-GDM模型预测精度最差,且训练耗时较长,而GA-BP-LM模型不仅可以获得较高的预测精度,且耗时最短,BP-LM模型居中。此外GA-BP-LM模型相对误差值波动幅度最小,平均相对误差仅为3.33%,达到了预测的精度要求,为飞机维修人员安全性预测提供了有效手段。
In order to realize the effective prediction of the safety of aircraft maintenance personnel, the safety evaluation index of maintenance personnel is constructed. Through the expert scoring method, the original training data and test data of maintenance personnel safety evaluation index were obtained. The data were trained and predicted by BP-GDM model, BP-LM model and GA-BP-LM model respectively. The simulation results show that: The BP-GDM model has the worst prediction accuracy and takes a long time to train. However, the GA-BP-LM model not only achieves high prediction accuracy but also has the shortest time-consuming and the BP-LM model is centered. In addition, the relative error of GA-BP-LM model has the smallest fluctuation with the average relative error of only 3.33%, which meets the prediction accuracy requirements and provides an effective means for aircraft maintenance personnel’s safety prediction.