Diagnosis of malfunction in fluidized beds is always a very difficult problem for a long time in academic circles and industrial circles. In this paper,we use acoustic sensor to measure the passive acoustic emissions ...Diagnosis of malfunction in fluidized beds is always a very difficult problem for a long time in academic circles and industrial circles. In this paper,we use acoustic sensor to measure the passive acoustic emissions (AE) from a bubbling fluidized bed.By comparison between acoustic and pressure signals,we found that acoustic signals are very sensitive to large particles such as agglomeration and particle sintering occurred in the bed and so on.Analysis of AE mechanism shows that AE measurement mainly stems from particle-chamber collisions and friction.AE measurements are completely non-intrusive and are not subject to erosion, corrosion, or plugging.Therefore, study and measurement of AE in different fluidization conditions will provide many new methods and trains of thought for detecting early deterioration of the fluidization quality.展开更多
中国西部地区正在发展集约化和规模化的设施养羊业,通过监测羊舍内的声信号可以判别羊只的行为状态,从而为设施养羊的福利化水平评估提取基础依据。梅尔频率倒谱系数(mel frequency cepstrum coefficient,MFCC)模拟了人耳对语音的处理...中国西部地区正在发展集约化和规模化的设施养羊业,通过监测羊舍内的声信号可以判别羊只的行为状态,从而为设施养羊的福利化水平评估提取基础依据。梅尔频率倒谱系数(mel frequency cepstrum coefficient,MFCC)模拟了人耳对语音的处理特点且抗噪音性强,被广泛用于畜禽发声信号的特征提取,但其没有考虑各个特征分量表征声信号的能力。该研究构建羊舍无线声音数据采集系统,采集20只羊在设施羊舍内的打斗、饥饿、咳嗽、啃咬和寻伴共5种行为下的声信号,并通过Audacity音频处理软件选出720个清晰且不重叠的声音样本数据。根据MFCC各分量对羊舍声信号表征能力,特征参数提取采用一种熵值加权的MFCC参数,再求其一、二阶差分并进行主成分分析降维,得到优化的19维特征参数。通过对羊舍声信号的声谱图分析,设计了支持向量机二叉树识别模型,并对模型内的4个分类器参数进行网格化寻优测试,该识别模型对羊只5种行为下的声信号进行分类识别,用改进的特征参数与传统MFCC和线性预测倒谱系数(linear predictive cepstrum coefficient,LPCC)进行对比分析。结果表明,该特征参数对5种行为的识别率平均可达83.6%,分别高于MFCC和LPCC参数14.1%和26.8%,羊只打斗和咳嗽行为的声信号属于相似的短时爆发类声音,其识别率分别仅为80.6%和79.5%,啃咬声特征显著不易混淆,其查全率可达到为92.5%,改进特征参数更好的表征了羊舍声信号的特征,提高了羊只不同行为的识别率,为羊只健康和福利状况的监测提供理论依据。展开更多
文摘Diagnosis of malfunction in fluidized beds is always a very difficult problem for a long time in academic circles and industrial circles. In this paper,we use acoustic sensor to measure the passive acoustic emissions (AE) from a bubbling fluidized bed.By comparison between acoustic and pressure signals,we found that acoustic signals are very sensitive to large particles such as agglomeration and particle sintering occurred in the bed and so on.Analysis of AE mechanism shows that AE measurement mainly stems from particle-chamber collisions and friction.AE measurements are completely non-intrusive and are not subject to erosion, corrosion, or plugging.Therefore, study and measurement of AE in different fluidization conditions will provide many new methods and trains of thought for detecting early deterioration of the fluidization quality.
