基于神经网络的双向光散射粉尘浓度测量方法研究检测技术与自动化装置专业论文.docx

山东科技大学硕士学位论文摘要 山东科技大学硕士学位论文 摘要 摘 要 煤矿井下的粉尘不仅严重影响着煤矿的安全生产，而且对矿工身体健康构成危害， 研究粉尘浓度的测量方法对粉尘防治具有重要的意义。针对煤矿井下粉尘测量中存在的 问题，研究了一种基于神经网络的双向光散射粉尘浓度测量方法。 本文通过对光散射理论的分析，得出了大粒径颗粒群的光散射可以用夫琅禾费衍射 理论近似，通过对信号的积分变换，得到了前向散射光能的积分信号与大粒径颗粒群浓 度的关系；小粒径颗粒群的光散射情形可以用瑞利散射理论近似，通过确定后向散射角 度，可以得到后向散射光能信号与小粒径颗粒群浓度的关系。根据这两种关系，建立了 前后两向光散射信号与粉尘浓度关系的数学模型。基于这种模型和测量方法，进行了测 量系统的硬件和软件设计，包括测量系统的整体设计，光源模块的设计，光路系统的整 体设计，光电转换模块的设计，数据采集模块的设计等，并对以单片机为核心的数据采 集模块的软件系统进行了整体设计。对于这种测量方法得到的信号，论述了基于神经网 络的数据融合方法，并进行了仿真和误差分析。仿真结果表明，利用神经网络强大的学 习能力，可以较好地得到前后两向光束信号与粉尘浓度的具体关系，通过与理想数据对 比和误差分析，表明这种数据处理方法具有准确性，可靠性及应用的广泛性。 基于这种设计的粉尘浓度测量系统不仅得到了大粒径粉尘颗粒的浓度信号，还得到 了小粒径粉尘颗粒的浓度信号，并通过数据融合算法得到了更加准确的总粉尘浓度信息， 这对于防止呼吸性粉尘对矿工身体的危害和提高矿井粉尘浓度测量准确性有重要的意 义。 关键词：粉尘浓度;前后两向光散射;数据融合;神经网络 ABSTRACT Coal mine dust bring serious effect not only to safety production but also to the miners health hazards, so the research of dust concentration measuring method of has important significance to dust prevention. In view of the situation that the measuring at present cannot detect the large an small size of particles at the same time , the article has given a dust measuring method based on two-way light scattering. Based on the analysis of the light scattering, this article has demonstrated the light scattering of large size particles can be approximated by the theory of Fraunhofer diffraction, and we can obtain the relationship between the integral signal of forward light scatting and the large particle swarm by the integral transform of Fraunhofer diffraction；the light scattering of small particles can be approximated by Rayleigh scattering theory and we can get the relationship between the backward light scattering and the small particle swarm by determining the scattering angle. According to these relationship, the article has established the mathematical model of the relationship between dust concentration and the two-way scattering signal. Based on this model and measuring method ,the author has designed the hardware and software of the measuring system, i