仓储粮食三维温度分布声学成像法重建研究

The occurrence of molds or insects in stored grain will result in a localized high temperature zone in a grain bulk, thus the temperature of a grain bulk is the most important symbol to judge the condition of stored grain. Grain is a bad heat conductor. At present, temperature measurement cables containing contact temperature sensors are used to monitor the temperature of stored grain. Since the number of the temperature sensors in a bin is rather small, a developing hot spot usually can not be detected in time, which results in serious loss of grain. The reconstruction of the two-dimensional temperature distribution of a typical plane in stored grain by acoustic tomography has been successfully in our experimental silo. Based on that, this project will research the reconstruction of three-dimensional temperature distribution of stored grain by acoustic tomography. Arrange several sound transmitters/receivers around the region to be measured, measure the travel-times between these transmitters/receivers, and reconstruct the three-dimensional temperature distribution in stored grain from these travel-times by using a proper reconstruction algorithm. The main research contents are as follows: the measurement method of sound travel-time across stored grain in the size of a real bin; the optimizing design of multi-plane sound transmitter/receiver array; the high spatial resolution three-dimensional temperature field reconstruction algorithm; the influence of sound bending on reconstruction accuracy and the way to overcome it; the explain of reconstruction image; the acoustic tomography system for three-dimensional temperature field reconstruction. These studies will lay the foundation for the entire monitoring of stored grain temperature with high spatial resolution. In addition, these studies are expected to enrich the theory and application of temperature field measurement technology by acoustic tomography.

储粮中霉变或虫害的发生都会导致局部温升，储粮温度是判断储粮状态的最重要标志。粮食是热的不良导体。目前储粮温度监测皆采用接触式测温电缆，测温点少、热点发现不及时导致储粮损失严重。用声学成像法重建储粮典型层面的二维温度分布，已在我们的实验粮仓中取得成功。在此基础上，本项目研究储粮三维温度分布声学成像法重建。在被测区域多个层面的周围安装若干声波发射/接收器，测量出声波穿越储粮由任一发射/接收器至其它发射/接收器的传播时间，用适当的重建算法重建出储粮的三维温度分布。主要研究内容：适合实际粮仓尺度的储粮中声波传播时间测量法、多层声波发射/接收器阵列优化、高空间分辨率的三维温度场重建算法、声线弯曲对重建精度的影响及克服方法、重建图像的解释、声学成像法三维温度场重建系统。项目研究可望丰富声学成像法温度场检测理论与应用，为实现储粮温度分布高空间分辨率全面监测奠定基础。

储粮中霉变或虫害的发生都会导致局部温升，储粮温度是判断储粮状态的最重要标志。粮食是热的不良导体。目前储粮温度监测皆采用接触式测温电缆，测温点少、热点发现不及时易导致储粮损失严重。本项目研究储粮三维温度分布声学法重建。主要研究内容和取得的成果如下。①研究了储粮中声波的衰减特性，提出了稳定性好、精度高的储粮中声波传播时间测量法，为储粮温度分布声学法重建提供高质量的投影数据。②开发了声学法三维温度场重建仿真研究软件包。用该软件包，对布置在圆筒形仓、房仓的多层发射/接收器阵列进行了优化。给出了阵列设计的一般原则。③常用的重建算法，如最小二乘法，大都要求像素数小于投影数据数，空间分辨率低，难以适应储粮温度分布重建的要求。为此，提出了“最小二乘与克里金插值法”，有效地克服了像素少给温度场细致描述所带来的障碍，能以更高精度重建复杂的三维温度场；提出了步长优化的同步迭代法（OSIRT），其收敛速度比SIRT快得多，用相近的重建时间，OSIRT重建的温度场更接近实际温度场；提出了“径向基函数与正则化法”，该算法允许像素数远多于投影数据数，更适合储粮中位置随机的热点重建要求；研究了该算法中正则化参数的自适应选取，与“L曲线法”相比，所提的“最小变化法”对应着更小的重建误差。④声波在非均匀温度场中传播时其轨迹会因折射而弯曲。为提高温度场重建精度，提出两种考虑声线弯曲的重建算法，“考虑声线弯曲的最小二乘法”和“考虑声线弯曲的径向基函数与正则化法”，并验证了所提方法的有效性；⑤研制了一套基于虚拟仪器技术的声学法温度场重建系统。测试结果表明系统能按照设计构想正常运行。温度场重建实验表明重建温度场与实际温度场吻合，能正确地反映被测区域温度随热源数目、位置及加热时间的变化。上述研究为声学法温度场重建技术用于实际粮仓奠定了基础。