唐思豪,滕召胜,孙 彪,胡 清,潘喜福.ADAM 改进 BP 神经网络与动态称重应用[J].电子测量与仪器学报,2021,35(4):127-135 |
ADAM 改进 BP 神经网络与动态称重应用 |
Improved BP neural network with ADAM optimizer andthe application of dynamic weighing |
|
DOI: |
中文关键词: 动态称量 检重秤 非线性补偿 多层 BP 神经网络 ADAM 优化器 |
英文关键词:dynamic weighing checkweigher non-linear compensation multi-layer BP neural network ADAM optimizer |
基金项目: |
|
|
摘要点击次数: 561 |
全文下载次数: 3 |
中文摘要: |
为提高动态检重秤的运行效率和测量准确度,深入分析了机械振动对测量的干扰及传感器非线性特性的产生机理。 提
出一种基于自适应矩估计法(ADAM)优化器的多层 BP 神经网络,实现了检重秤传感器的非线性校正,并准确估计了动态称量
结果。 试验对比经典梯度下降法、附加动量法、均方根传播法以及 ADAM 算法,结果表明 ADAM 算法综合考虑了参数梯度的一
阶和二阶矩估计,具有更快的收敛速度,更准确的预测结果。 最终实现满量程 400 g,最高运行速度 2 m/ s 的高速动态检重秤,
型式测试结果表明其各指标均满足国家标准《GB/ T 27739—2011 自动分检衡器》对 XIII 级检重秤的要求。 |
英文摘要: |
To improve the operational efficiency and measurement accuracy of the dynamic check weigher, the interference of mechanical
vibration to the measurement and the generating mechanism of the sensor’s nonlinear characteristics are deeply analyzed. A multi-layer
BP neural network based on ADAM optimizer is proposed to realize the nonlinear correction of weighing sensor and estimate the dynamic
weighing results accurately. The classical gradient descent algorithm, gradient descent algorithm with momentum and root-mean-square
propagation algorithm are compared with the ADAM algorithm through experiment. According to the results, the ADAM algorithm had
faster convergence speed and more accurate prediction results as it comprehensively considered the first and second sample moment of
parameter’s gradient. The high speed dynamic check weigher with full range of 400 g and maximum running speed of 2 m/ s is
manufactured, The type test results showed that all of its indicators meet the requirements of national standard GB/ T 27739 - 2011
automatic divider for XIII check weigher. |
查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|