基于超声波传感器阵列及多重信号分类算法的风速风向测量方法

发布时间：2018-05-18 18:26

本文选题：超声波测量 + 传感器阵列　；参考：《吉林大学》2017年硕士论文

【摘要】：风作为一种重要的自然现象,在现今生活中有着广泛的应用,在铁路、航空、航海、军事和气象等方面,风速风向的准确测量至关重要。传统的测风仪器由于存在可测风速范围不高、机械磨损、使用寿命短、维护成本高和精度较低等诸多缺点,逐渐被结构简单,测量范围宽,测量速度快,准确度高的超声波测风仪所取代。现有的超声波测风仪应用最为广泛的原理是时差法,即利用超声波在顺风和逆风情况下传播时间的不同来测量风速。国内外学者的相关研究也都是围绕时差法展开的,而时差法的测量精度完全取决于对超声波传播时间的测量精度。但是,在强电磁干扰和低信噪比的环境下,超声波传播时间的测量将变得困难甚至失效。阵列信号处理方法可以在复杂环境中对噪声进行有效抑制,目前将阵列信号处理的思想应用于气象仪器的研究,特别是超声测风方面的研究较少。本文利用阵列信号处理的优势,借鉴多重信号分类算法对风速风向信息进行估计,以突破时差法的原理限制。本文的主要工作内容如下:1.本文设计了一种弧形超声波传感器阵列结构,在此基础上建立了风速风向作用下的近场阵列接收数据的数学模型,并对其阵列流型矢量进行了推导,最后借鉴多重信号分类算法的基本思想,根据待测风速风向的范围及分辨率,通过谱峰搜索估计出风速风向信息;2.对弧形阵列结合多重信号分类算法估计风速风向信息方法的相关性能和问题展开分析和讨论:对所提方法的风速和风向估计方差公式分别进行推导,得到的公式可用于任意阵列结构,同时也分别给出了阵列测风方法的风速风向估计克拉美-罗界公式;利用估计方差的计算结果以及阵列模糊问题的分析,讨论了均匀线阵对于本文提出的方法的缺陷;接着对方法测向范围、阵列模糊和分辨力问题展开了详细的分析和讨论;通过设计仿真实验验证了方法的可行性以及相关问题和性能。3.根据待估计的风速风向范围以及分辨率,利用推导得到的方差公式,分别计算得到所提方法对风速和风向的估计方差值,从理论上对阵列测风方法估计的准确性做了一定的说明,着重分析讨论了估计方差相对较大的部分;推导得到的估计方差公式,对于分析这一类方法的准确性有着很高的应用价值。4.通过搭建硬件实验平台验证本文提出的测风方法:首先对硬件实验平台的整体结构及各部分组成进行了详细描述,利用Gill仪器对搭建的实验风洞进行了风速标定;接着说明了在硬件实验平台的实际应用中,对数据进行处理的几种方式;最后对实验平台在风洞中的实验结果以及产生误差的原因进行分析和讨论,证明了方法的可行性以及算法应用的稳定性。
[Abstract]:Wind, as an important natural phenomenon, is widely used in modern life. In railway, aviation, navigation, military and meteorology, the accurate measurement of wind speed and wind direction is very important. Traditional wind measuring instruments have many disadvantages, such as low range of measurable wind speed, mechanical wear, short service life, high maintenance cost and low precision, etc., so they are gradually simple in structure, wide in measurement range and fast in measuring speed. The ultrasonic wind meter with high accuracy was replaced. The most widely used principle of the existing ultrasonic wind meter is the moveout method, which is to measure the wind speed by using the different propagation time of the ultrasonic wave in the downwind and the headwind. The related research of domestic and foreign scholars is also focused on the time difference method, and the accuracy of the time difference method is completely dependent on the measurement accuracy of ultrasonic propagation time. However, in the environment of strong electromagnetic interference and low signal-to-noise ratio, the measurement of ultrasonic propagation time will become difficult or even ineffective. Array signal processing method can effectively suppress noise in complex environment. At present, the idea of array signal processing is applied to the research of meteorological instruments, especially the research on ultrasonic wind measurement. In this paper, the advantages of array signal processing are used to estimate wind speed and wind direction by using multi-signal classification algorithm to break through the principle limitation of moveout method. The main work of this paper is as follows: 1. In this paper, a structure of an array of curved ultrasonic sensors is designed. Based on this, a mathematical model of near-field array receiving data under wind speed and wind direction is established, and its array flow pattern vector is deduced. Finally, based on the basic idea of multi-signal classification algorithm and the range and resolution of wind direction to be measured, the information of wind speed and wind direction is estimated by spectral peak search. This paper analyzes and discusses the related performance and problems of the method of estimating wind speed and direction information by arc array combined with multi-signal classification algorithm. The formulas of variance of wind speed and wind direction estimation are deduced respectively. The formulas obtained can be used for arbitrary array structures, and the Clemei-Row bound formula for wind velocity direction estimation by array wind measurement method is given, and the calculation results of estimated variances and the analysis of array fuzzy problems are used. The defects of uniform linear array to the method proposed in this paper are discussed, and then the problems of direction finding range, array ambiguity and resolution are analyzed and discussed in detail. The feasibility, related problems and performance of the method are verified by simulation experiments. According to the range and resolution of wind speed and direction to be estimated, the estimated variance of the proposed method for wind speed and direction is calculated by using the derived variance formula. In this paper, the accuracy of the estimation of the array wind measurement method is explained in theory, the part of the estimation variance is analyzed and discussed, and the formula of the estimated variance is derived. It has high application value for analyzing the accuracy of this kind of method. Firstly, the whole structure and components of the hardware experimental platform are described in detail, and the wind speed of the experimental wind tunnel is calibrated by using Gill instrument. Then it explains several methods of data processing in the practical application of the hardware experimental platform, and finally analyzes and discusses the experimental results of the experimental platform in wind tunnel and the causes of the errors. The feasibility of the method and the stability of its application are proved.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH765.4;TP212.9

【相似文献】