摘要
设计了一种综合运用BP网络与GA灰色模型的时栅传感器信号状况预测方法。采用权重-尺度-平均方法构建对应的权重平方和均值联合模型,实现对多个样本的自适应性处理,显著提高预测准确性。研究结果表明:本次测试相关性均超过98%,达到了理想的预报精度。采用联合模型获得测试参数非常接近预报结果,得到优于单个模型的精确预报结果,能够达到实际预报需求,能实现较高预报精度。该研究有助于提高工业自动化水平,为后续的节能提高起到很好的效果。
A signal condition prediction method of time-gate sensor based on BP neural network and grey model is designed.The weight-scale-average method is used to construct the corresponding weight square and mean joint model,which realizes the adaptive processing of multiple samples and significantly improves the prediction accuracy.The results show that the correlation of this test is more than 98%,and the ideal prediction accuracy is achieved.The combined model is used to obtain the test parameters very close to the forecast results,and the accurate forecast results are better than that of a single model,which can meet the actual forecast requirements and achieve high prediction accuracy.This research is helpful to improve the level of industrial automation,and has a good effect for the subsequent improvement of energy saving.
作者
彭献峰
Peng Xianfeng(Puyang Vocational Secondary School,Puyang Henan 457000,China)
出处
《现代工业经济和信息化》
2024年第11期145-146,149,共3页
Modern Industrial Economy and Informationization
关键词
时栅传感器
激励信号误差
组合预测模型
健康诊断
time gate sensor
excitation signal error
combination prediction model
health diagnosis
