基于VMD-ConvKELM算法的燃气计量检测系统

Gas measurement and detection system basedon VMD-ConvKELM algorithm

燃气流量计量中存在信号噪声干扰和气流扰动,不能满足冶金、燃气、发电等工业领域精确计量需求。文章设计了一种基于气体流量传感器采集与变分模态分解(VMD)滤波优化预测补偿的燃气流量检测系统。在气体流量传感器采集信号进行变分模态分解的滤波处理,并使用卷积计算核极限学习机(ConvKELM)预测模型对数据误差进行预测补偿。实验结果表明,VMD-ConvKELM方法在信号分解和误差预测补偿任务中具有优越的性能,通过对比不同算法的预测精度,结果显示VMD-ConvKELM优化的燃气计量检测能够有效地测量实际流量值,具有较高精度且结果更加稳定可靠。

Due to signal interference and airflow disturbance in gas flow measurement,it cannot meet the precise measurement requirements of industrial fields such as metallurgy,gas,and power generation,a gas flow detection system based on gas flow sensor collection and variational mode decomposition(VMD)filtering optimization prediction compensation was designed.The signal collected by the gas flow sensor was filtered and processed using variational mode decomposition,And use convolutional kernel extreme learning machine(convKELM)prediction model to predict and compensate for data errors.The experimental results show that the VMD-ConvKELM method has superior performance in signal decomposition and error prediction compensation tasks.By comparing the prediction accuracy of different algorithms,the results show that the VMD-ConvKELM optimized gas metering detection can effectively measure the actual flow value,with high accuracy and more stable and reliable results.