CN106197252A - The method for building up of ground deformation position distribution measuring model based on parallel spiral transmission line - Google Patents
The method for building up of ground deformation position distribution measuring model based on parallel spiral transmission line Download PDFInfo
- Publication number
- CN106197252A CN106197252A CN201610560335.8A CN201610560335A CN106197252A CN 106197252 A CN106197252 A CN 106197252A CN 201610560335 A CN201610560335 A CN 201610560335A CN 106197252 A CN106197252 A CN 106197252A
- Authority
- CN
- China
- Prior art keywords
- transmission line
- voltage signal
- reflected voltage
- spiral transmission
- establishment method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000005259 measurement Methods 0.000 claims abstract description 16
- 239000013598 vector Substances 0.000 claims abstract description 12
- 238000000605 extraction Methods 0.000 claims abstract description 8
- 238000012706 support-vector machine Methods 0.000 claims description 14
- 239000012212 insulator Substances 0.000 claims description 8
- 230000008054 signal transmission Effects 0.000 claims description 8
- 238000002310 reflectometry Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 4
- 238000011425 standardization method Methods 0.000 claims description 2
- 238000013480 data collection Methods 0.000 claims 1
- 230000001681 protective effect Effects 0.000 claims 1
- 238000004804 winding Methods 0.000 claims 1
- 239000002689 soil Substances 0.000 description 16
- 239000011435 rock Substances 0.000 description 10
- 239000004568 cement Substances 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 4
- 230000003750 conditioning effect Effects 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 239000004576 sand Substances 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002243 precursor Substances 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 239000000741 silica gel Substances 0.000 description 1
- 229910002027 silica gel Inorganic materials 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 229920002379 silicone rubber Polymers 0.000 description 1
- 239000004945 silicone rubber Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B7/00—Measuring arrangements characterised by the use of electric or magnetic techniques
- G01B7/16—Measuring arrangements characterised by the use of electric or magnetic techniques for measuring the deformation in a solid, e.g. by resistance strain gauge
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
Description
技术领域technical field
本发明涉及岩土变形测试技术领域，尤其涉及基于平行螺旋传输线的岩土变形位置分布测量模型的建立方法。The invention relates to the technical field of rock-soil deformation testing, in particular to a method for establishing a rock-soil deformation position distribution measurement model based on parallel spiral transmission lines.
背景技术Background technique
我国是地质灾害多发国家，频繁的地质灾害给人民群众生命财产造成了严重伤害和巨大损失。山体滑坡、地面沉降等地质运动引起的局部岩土变形是一种重要的灾害前兆现象，准确定位岩土变形位置对于我国地质灾害的防治工作具有重要意义。my country is a country where geological disasters frequently occur, and frequent geological disasters have caused serious injuries and huge losses to the lives and properties of the people. Local rock-soil deformation caused by geological movements such as landslides and land subsidence is an important precursor to disasters. Accurately locating the location of rock-soil deformation is of great significance for the prevention and control of geological disasters in my country.
国内外学者对于岩土表面形变位置的监测作了大量工作，监测手段主要有以下几类：(1)GPS技术与全站仪观测技术。这类技术通过在岩土表面大量布点，测量三维坐标变化从而得到岩土变形位置，但是由于布点成本昂贵，难以覆盖整个监测区域，会造成大量观测盲区。(2)光纤传感技术。但是光纤拉伸量小，而且不能大幅度的扭折，在形变量较大的岩土变形中，容易被拉断。(3)拉线式传感技术。该技术将多个传感器布成网络测量岩土变形量，对形变量的测量精度能达到0.1mm，但是不能确定发生形变的具体位置。Scholars at home and abroad have done a lot of work on monitoring the deformation position of rock and soil surfaces. The monitoring methods mainly include the following categories: (1) GPS technology and total station observation technology. This type of technology obtains the deformation position of rock and soil by arranging a large number of points on the surface of the rock and soil and measuring the change of three-dimensional coordinates. However, due to the high cost of laying out points, it is difficult to cover the entire monitoring area, which will cause a large number of observation blind spots. (2) Optical fiber sensing technology. However, the amount of stretching of the optical fiber is small, and it cannot be greatly twisted. It is easy to be broken in the deformation of rock and soil with a large amount of deformation. (3) Pull wire sensing technology. This technology arranges multiple sensors into a network to measure the deformation of rock and soil, and the measurement accuracy of the deformation can reach 0.1mm, but the specific location of the deformation cannot be determined.
CN102522148A公布了一种平行螺旋传输线传感器，在圆截面硅橡胶条外密绕一层有两根相互绝缘的电缆线，两根电缆线形成螺旋线，两根电缆线形成螺旋线外裹包有硅橡胶套管，两根电缆线的一端接匹配阻抗ZL，两根电缆线的另一端接时域反射测量仪，但没有公开具体的形变位置测试方法。CN102522148A discloses a parallel helical transmission line sensor. There are two mutually insulated cables wrapped around a layer of circular section silicone rubber strips. The two cables form a helix, and the two cables form a helix wrapped with silicon. One end of the two cables is connected to the matching impedance ZL, and the other end of the two cables is connected to a time domain reflectometry instrument, but the specific deformation position test method is not disclosed.
发明内容Contents of the invention
本发明提供了基于平行螺旋传输线的岩土变形位置分布测量模型的建立方法，利用建立的模型可以准确测试岩土形变位置。The invention provides a method for establishing a rock soil deformation position distribution measurement model based on parallel spiral transmission lines, and the rock soil deformation position can be accurately tested by using the established model.
基于平行螺旋传输线的岩土变形位置分布测量模型的建立方法，所述平行螺旋传输线包括中心的弹性绝缘体、两根平行螺旋缠绕在所述弹性绝缘体上相互绝缘的信号传输线，以及外层的绝缘保护套，两根信号传输线一端悬空，另一端连接时域反射测量装置；A method for establishing a geotechnical deformation position distribution measurement model based on a parallel helical transmission line, the parallel helical transmission line includes a central elastic insulator, two parallel helical signal transmission lines wound on the elastic insulator and insulated from each other, and an outer layer of insulation protection One end of the two signal transmission lines is suspended in the air, and the other end is connected to the time domain reflectometry device;
所述建立方法包括：The establishment method includes:
(1)将平行螺旋传输线平均分为若干段，对每段进行拉伸，利用时域反射测量装置，采集得到反射电压信号；(1) Divide the parallel spiral transmission line into several sections on average, stretch each section, and use the time-domain reflection measurement device to collect the reflected voltage signal;
(2)对反射电压信号进行预处理；(2) Preprocessing the reflected voltage signal;
(3)对经步骤(2)预处理的反射电压信号进行特征提取，获得特征向量；(3) performing feature extraction on the reflected voltage signal preprocessed through step (2) to obtain a feature vector;
(4)以所述特征向量作为输入层数据，以拉伸位置和起点之间的距离作为输出层数据，建立最小二乘-支持向量机模型。(4) Using the feature vector as the input layer data, and using the distance between the stretching position and the starting point as the output layer data, a least squares-support vector machine model is established.
平行螺旋传输线没有形变时，它的特性阻抗为固定的状态，发生形变后，变形位置的特性阻抗发生变化。在平行螺旋传输线的一端口输入脉冲信号，当阻抗发生变化，脉冲信号的反射电压信号也随之发生变化，反射电压信号蕴含变形位置的特征信息。本发明正是基于该原理，建立了最小二乘-支持向量机模型。When the parallel helical transmission line is not deformed, its characteristic impedance is in a fixed state. After deformation, the characteristic impedance of the deformed position changes. A pulse signal is input to one port of the parallel spiral transmission line. When the impedance changes, the reflected voltage signal of the pulse signal also changes accordingly, and the reflected voltage signal contains the characteristic information of the deformation position. Based on this principle, the present invention establishes a least squares-support vector machine model.
所述平行螺旋传输线越长，所能检测的岩土形变范围越大，一般情况下，所述平行螺旋传输线的长度大于2米。The longer the parallel helical transmission line, the larger the range of geotechnical deformation that can be detected. Generally, the length of the parallel helical transmission line is greater than 2 meters.
为减小反射电压信号的基线漂移，需要对采集到的信号进行预处理，所述预处理采用变量标准化和归一化法。In order to reduce the baseline drift of the reflected voltage signal, the collected signal needs to be preprocessed, and variable standardization and normalization methods are used for the preprocessing.
采集的反射电压信号会包含较多的数据，所有数据构成一向量。由于是局部拉伸，只有拉伸变形位置的特性阻抗才会发生变化，因此仅部分数据包含了平行螺旋传输线阻抗变化的特征信息，需要采用特征变量提取算法将特征信息提取出来。The collected reflected voltage signal will contain more data, and all the data constitute a vector. Due to the local stretching, only the characteristic impedance of the stretching deformation position will change, so only part of the data contains the characteristic information of the impedance change of the parallel helical transmission line, and the characteristic variable extraction algorithm needs to be used to extract the characteristic information.
优选的，所述特征提取采用连续投影法。Preferably, the feature extraction adopts a continuous projection method.
所述最小二乘-支持向量机模型的核函数为高斯径向基函数，最小二乘-支持向量机模型具体为：The kernel function of described least squares-support vector machine model is Gaussian radial basis function, and least squares-support vector machine model is specifically:
b为偏置值，N为样本数量，K(x，xk)为高斯径向基函数，ak拉格朗日函数乘子，y为输出层数据，x为输入层数据，xk为样本输入层数据。b is the bias value, N is the number of samples, K(x, x k ) is the Gaussian radial basis function, a k Lagrange function multiplier, y is the output layer data, x is the input layer data, x k is Sample input layer data.
时域反射测量装置的采集频率越高，每个反射电压信号包含的数据越多，计算就越大，一般10-1000MHz，本发明实施例为100MHz，The higher the acquisition frequency of the time-domain reflectometry device, the more data each reflected voltage signal contains, and the larger the calculation, generally 10-1000MHz, the embodiment of the present invention is 100MHz,
所述脉冲信号幅值与检测信号的强弱有关，一般情况下其幅值为±1-100V，本发明实施例为±10V，脉冲宽度为1-100ns，本发明实施例为10ns。The amplitude of the pulse signal is related to the strength of the detection signal. Generally, the amplitude is ±1-100V. In the embodiment of the present invention, the amplitude is ±10V. The pulse width is 1-100ns, which is 10ns in the embodiment of the present invention.
本发明利用平行螺旋传输线发生形变时，其阻抗会发生相应变化，从平行螺旋传输线一端输入脉冲信号，检测其反射电压信号，因反射电压信号包含位置信息，基于该原理建立最小二乘-支持向量机的预测模型，用于准确测量形变位置。In the present invention, when the parallel helical transmission line is deformed, its impedance will change accordingly, and a pulse signal is input from one end of the parallel helical transmission line to detect its reflected voltage signal. Because the reflected voltage signal contains position information, the least squares-support vector is established based on this principle A predictive model of the machine for accurate measurements of deformation locations.
附图说明Description of drawings
图1为本发明平行螺旋传输线的结构示意图。FIG. 1 is a schematic diagram of the structure of the parallel helical transmission line of the present invention.
图2为本发明平行螺旋传输线局部变形结构示意图。Fig. 2 is a schematic diagram of the local deformation structure of the parallel helical transmission line of the present invention.
图3为平行螺旋传输线不同位置拉伸时TDR测量仪采集得到的反射电压信号。Figure 3 is the reflected voltage signal collected by the TDR measuring instrument when the parallel helical transmission line is stretched at different positions.
具体实施方式detailed description
如图1和图2所示，平行螺旋传输线6，包括呈圆柱状的弹性绝缘体3，弹性绝缘体3表面平行螺旋缠绕信号传输线1、2，外部设有绝缘套4。信号传输线1、2相互绝缘，可以选用漆包线，信号传输线1、2的直径大概在0.25mm。弹性绝缘体3和绝缘套4的材质为硅胶，弹性绝缘体3和绝缘套4的直径分别是3.5mm和5.5mm。As shown in Figures 1 and 2, the parallel spiral transmission line 6 includes a cylindrical elastic insulator 3, the surface of the elastic insulator 3 is parallel to the spirally wound signal transmission lines 1 and 2, and an insulating sleeve 4 is provided outside. The signal transmission lines 1 and 2 are insulated from each other, enameled wires can be used, and the diameter of the signal transmission lines 1 and 2 is about 0.25 mm. The elastic insulator 3 and the insulating sleeve 4 are made of silica gel, and the diameters of the elastic insulator 3 and the insulating sleeve 4 are 3.5mm and 5.5mm respectively.
如图2所示，对平行螺旋传输线进行局部拉伸，拉伸位置7处出现形变。信号传输线1、2一端悬空，另一端连接时域反射测量装置5。时域反射测量装置由脉冲信号发生电路，信号调理电路和数据采集电路构成，脉冲信号发生电路发出的脉冲信号幅值为±10V，脉冲宽度为10ns。信号调理电路有隔离的作用，能避免脉冲信号对反射电压信号接受造成干扰。反射电压信号经信号调理电路放大和滤波，由数据采集电路采集发送至上位机。数据采集电路的采用频率为100MHz，每次采集的数据为250个。As shown in Fig. 2, the parallel helical transmission line is partially stretched, and deformation occurs at stretching position 7. One end of the signal transmission lines 1 and 2 is suspended, and the other end is connected to the time domain reflectometry device 5 . The time domain reflection measurement device is composed of a pulse signal generating circuit, a signal conditioning circuit and a data acquisition circuit. The amplitude of the pulse signal sent by the pulse signal generating circuit is ±10V, and the pulse width is 10ns. The signal conditioning circuit has the function of isolation, which can prevent the pulse signal from interfering with the reception of the reflected voltage signal. The reflected voltage signal is amplified and filtered by the signal conditioning circuit, collected by the data acquisition circuit and sent to the host computer. The frequency of the data acquisition circuit is 100MHz, and the data collected each time is 250.
本发明测量模型建立方法具体如下：The measurement model establishment method of the present invention is specifically as follows:
(a)以平行螺旋传输线和时域反射测量装置连接的端部为起点，每隔10cm做一标记，每两相邻的标记之间进行一次拉伸，拉伸时使用两对木头夹具将两个标记位置加紧，每次拉伸2cm。每次拉伸时，采用时域反射测量装置5(TDR测量仪)采集平行螺旋传输线6的反射电压信号，并将信号发送至上位机，由上位机对数据进行处理并显示拉伸位置。(a) Starting from the end of the connection between the parallel helical transmission line and the time-domain reflectometry device, make a mark every 10cm, and stretch between every two adjacent marks. When stretching, use two pairs of wooden clamps to hold the two Tighten the marked position, stretching 2cm each time. During each stretching, the reflected voltage signal of the parallel helical transmission line 6 is collected by a time-domain reflectometry device 5 (TDR measuring instrument), and the signal is sent to the host computer, which processes the data and displays the stretching position.
(b)本发明采用的平行螺旋传输线总长度为5米，每隔10cm拉伸一次，共获得49组反射电压信号，每组信号包含250个数据，一共有250×49个数据，图3为平行螺旋传输线不同位置拉伸时TDR测量仪采集得到的反射电压信号。(b) The total length of the parallel helical transmission line used in the present invention is 5 meters, stretched once every 10cm, and a total of 49 groups of reflected voltage signals are obtained, each group of signals contains 250 data, and there are 250×49 data in total, as shown in Figure 3 The reflected voltage signal collected by the TDR measuring instrument when the parallel helical transmission line is stretched at different positions.
(c)为了减小仪器状态、实验环境变化对反射电压信号测量带来的影响，需要对反射电压信号进行预处理。常见的预处理算法包括平滑滤波、导数校正、多元散射校正、变量标准化和归一化等。本实施例先采用变量标准化法用于消除反射电压信号的基线漂移，然后采用归一化法对反射电压数据进行处理，以消除反射电压信号的随机误差。(c) In order to reduce the impact of instrument status and experimental environment changes on the measurement of the reflected voltage signal, it is necessary to preprocess the reflected voltage signal. Common preprocessing algorithms include smoothing filtering, derivative correction, multivariate scatter correction, variable standardization and normalization, etc. In this embodiment, the variable normalization method is first used to eliminate the baseline drift of the reflected voltage signal, and then the normalized method is used to process the reflected voltage data, so as to eliminate the random error of the reflected voltage signal.
(d)每次拉伸测量得到的反射电压信号包含250个数据，由于是局部拉伸，而只有拉伸变形位置的特性阻抗才会发生变化，因此这250个数据中只有部分包含了平行螺旋传输线阻抗变化的特征信息，需要采用特征变量提取算法将特征信息提取出来，本实施例采用的特征变量提取算法为连续投影算法。连续投影算法的基本思路是：随机选择49组反射电压信号中的36组作为建模集，剩余13组作为验证集。通过对建模集数据进行投影映射构造出新的特征变量集合，根据这些特征变量集合依次建立多元线性回归(MLR)模型，然后利用验证集数据对MLR模型的预测结果进行评估，从而选择出含有最低限度冗余信息的特征变量集合，即特征向量。特征变量选择以后，每一组反射电压信号包含27个数据。(d) The reflected voltage signal obtained by each stretching measurement contains 250 data, because it is a local stretch, and only the characteristic impedance of the stretch deformation position will change, so only part of the 250 data contains parallel helix The characteristic information of the impedance change of the transmission line needs to be extracted by using a characteristic variable extraction algorithm, and the characteristic variable extraction algorithm adopted in this embodiment is a continuous projection algorithm. The basic idea of the continuous projection algorithm is: randomly select 36 groups of 49 groups of reflected voltage signals as the modeling set, and the remaining 13 groups as the verification set. A new feature variable set is constructed by projection mapping of the modeling set data, and a multiple linear regression (MLR) model is established in turn based on these feature variable sets. A collection of feature variables with minimal redundant information, namely feature vectors. After feature variable selection, each group of reflected voltage signals contains 27 data.
(e)预处理和特征变量选择以后的反射电压信号作为输入层数据，以拉伸位置与起点之间的距离作为输出层数据，根据建模集，建立最小二乘-支持向量机模型。设建模集D＝{(x1,y1),(x2,y2),…,(xk,yk)}，1≤k≤N，本实施例中N为36。集合中输入层数据xk∈RN，输出层数据yk∈R；然后利用一个非线性函数将xk映射到高维空间并建立回归模型：(e) The reflected voltage signal after preprocessing and feature variable selection is used as the input layer data, and the distance between the stretching position and the starting point is used as the output layer data. According to the modeling set, a least squares-support vector machine model is established. Suppose the model set D={(x 1 , y 1 ), (x 2 , y 2 ), . . . , (x k , y k )}, 1≤k≤N, and N is 36 in this embodiment. The input layer data x k ∈ R N in the set, the output layer data y k ∈ R; then use a nonlinear function Map x k to a high-dimensional space and build a regression model:
上式中，b为偏置值，w∈RN为权值向量。最小二乘-支持向量机模型的函数拟合问题可以转化为对以下方程进行求解：In the above formula, b is the bias value, and w∈RN is the weight vector. The function fitting problem of the least squares-support vector machine model can be transformed into solving the following equation:
上式约束条件为： The above constraints are:
其中，ek为误差变量，γ是正则化参数，控制对超出误差样本的惩罚程度，为内核映射函数。将上式转换至对偶空间，得到拉格朗日函数的形式为：Among them, e k is the error variable, γ is the regularization parameter, which controls the degree of punishment for samples exceeding the error, Map functions for the kernel. Converting the above formula to the dual space, the form of the Lagrangian function is obtained as:
式中拉格朗日函数乘子ak∈R被称作支持值，对上式各变量求偏导数可以得到以下等式：In the formula, the multiplier a k ∈ R of the Lagrangian function is called the support value, and the partial derivative of each variable in the above formula can be obtained as follows:
变量w和e被迭代消除后，可得线性方程组：After the variables w and e are eliminated iteratively, a system of linear equations can be obtained:
式中In the formula
其中K(xk,xl)是内核函数，本实施例采用的是RBF函数，计算公式如下：Wherein K(x k , x l ) is a kernel function, and what this embodiment adopts is the RBF function, and the calculation formula is as follows:
式(7)中σ是RBF核函数的半径，核函数σ影响着特征空间中的样本分布复杂度。得到最小二乘-支持向量机(LS-SVM)的拟合模型为：In formula (7), σ is the radius of the RBF kernel function, and the kernel function σ affects the complexity of sample distribution in the feature space. The fitting model obtained by the least squares-support vector machine (LS-SVM) is:
选择RBF核函数后，LS-SVM模型还需要确定正则化参数γ和RBF核函数参数σ。利用Matlab软件LS-SVM工具包中的tunelssvm函数，采用完全搜索方式，寻优范围设置为[10-6,106]，计算得到γ和σ分别为6448和166287，再利用LS-SVM工具包中的trainlssvm函数就可以建立最小二乘-支持向量机模型。余下13组反射电压信号作为验证集用于对预测模型进行评估，评估参数包括均方根误差(RMSE)和决定系数(R2)，本实施例最小二乘-支持向量机模型的RMSE和R2分别为2.390和0.993。After selecting the RBF kernel function, the LS-SVM model also needs to determine the regularization parameter γ and the RBF kernel function parameter σ. Using the tunelssvm function in the LS-SVM toolkit of Matlab software, using the complete search method, the optimization range is set to [10 -6 , 10 6 ], and the calculated values of γ and σ are 6448 and 166287 respectively, and then use the LS-SVM toolkit The trainlssvm function in can build the least squares-support vector machine model. The remaining 13 groups of reflected voltage signals are used as a verification set to evaluate the prediction model, and the evaluation parameters include root mean square error (RMSE) and coefficient of determination (R 2 ), the RMSE and R of the least squares-support vector machine model in this embodiment 2 are 2.390 and 0.993, respectively.
为了进行对比，将建模集输入层数据作为自变量，建模集输出层数据作为因变量，建立线性的偏最小二乘预测模型：For comparison, the input layer data of the modeling set is used as the independent variable, and the output layer data of the modeling set is used as the dependent variable to establish a linear partial least squares prediction model:
Y＝A*X+bY=A*X+b
其中Y为平行螺旋传输线拉伸位置，X为反射电压信号，A为系数，b为常数。Among them, Y is the stretching position of the parallel helical transmission line, X is the reflected voltage signal, A is a coefficient, and b is a constant.
(f)平行螺旋传输线形变位置在线测量：固定平行螺旋传输线的某一点，采集该点位置拉伸2厘米以上时的时域反射电压信号，将该信号进行预处理和特征提取以后，分别代入到最小二乘-支持向量机模型与偏最小二乘模型中，计算得到平行螺旋传输线的形变位置，结果见表1。(f) On-line measurement of the deformation position of the parallel helical transmission line: fix a certain point of the parallel helical transmission line, collect the time domain reflection voltage signal when the position of the point is stretched by more than 2 cm, and preprocess the signal and extract the feature, and then substitute it into the In the least squares-support vector machine model and the partial least squares model, the deformation position of the parallel spiral transmission line is calculated, and the results are shown in Table 1.
表1Table 1
从表1数据可以看出，支持向量机模型测得的平行螺旋传输线形变位置与实际值更为接近，优于偏最小二乘模型。这是因为，平行螺旋线含有分布式电感与分布式电容，当高频信号作用时，分布式电容和分布式电感在阻抗变化中占据主导地位，这就意味着当平行螺旋线拉伸时，反射电压信号包含大量阻抗变化的非线性信息。相比于线性偏最小二乘模型，支持向量机模型能够更好的利用这些非线性信息。It can be seen from the data in Table 1 that the deformation position of the parallel spiral transmission line measured by the support vector machine model is closer to the actual value, which is better than the partial least squares model. This is because the parallel helix contains distributed inductance and distributed capacitance. When a high-frequency signal acts, the distributed capacitance and distributed inductance occupy a dominant position in the impedance change, which means that when the parallel helix is stretched, The reflected voltage signal contains a lot of non-linear information of impedance changes. Compared with the linear partial least squares model, the support vector machine model can make better use of these nonlinear information.
在实际应用中，为提高平行螺旋传输线6与岩土体的耦合性，先在埋设平行螺旋传输线6的岩土体挖一条宽30cm、深30cm的沟，然后按横截面直径为10cm的尺寸，倒入条状的手捏可成型的黄沙水泥混合浆。再将平行螺旋传输线6埋入，在黄沙水泥混合浆尚未固化时，将挖掉的岩土覆盖上去，这样当黄沙水泥混合浆固化后，平行螺旋传输线6就会被一层水泥体包裹。当岩土体变形移动时会带动水泥体，水泥体会断裂并带动平行螺旋传输线6的拉伸，测量系统便能检测到该段岩土体的变形。In practical application, in order to improve the coupling between the parallel spiral transmission line 6 and the rock-soil mass, a trench with a width of 30 cm and a depth of 30 cm is first dug in the rock-soil body where the parallel spiral transmission line 6 is buried, and then the diameter of the cross section is 10 cm. Pour in strips of hand-kneaded formable yellow sand cement mixture. Embed the parallel spiral transmission line 6 again, and cover the excavated rock and soil when the yellow sand cement mixture is not yet solidified, so that after the yellow sand cement mixture solidifies, the parallel spiral transmission line 6 will be wrapped by a layer of cement . When the rock and soil mass deforms and moves, it will drive the cement body, and the cement body will break and drive the stretching of the parallel helical transmission line 6, and the measurement system can detect the deformation of the rock and soil mass in this section.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610560335.8A CN106197252A (en) | 2016-07-13 | 2016-07-13 | The method for building up of ground deformation position distribution measuring model based on parallel spiral transmission line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610560335.8A CN106197252A (en) | 2016-07-13 | 2016-07-13 | The method for building up of ground deformation position distribution measuring model based on parallel spiral transmission line |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106197252A true CN106197252A (en) | 2016-12-07 |
Family
ID=57474717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610560335.8A Pending CN106197252A (en) | 2016-07-13 | 2016-07-13 | The method for building up of ground deformation position distribution measuring model based on parallel spiral transmission line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106197252A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109238128A (en) * | 2018-11-06 | 2019-01-18 | 江苏柔世电子科技有限公司 | Big strain sensor of flexible inductive and preparation method thereof, big strain transducer |
| CN110243278A (en) * | 2019-07-10 | 2019-09-17 | 浙江水利水电学院 | A kind of rock displacement amount distributed measurement method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102522148A (en) * | 2011-11-15 | 2012-06-27 | 中国计量学院 | Rock-soil body deformation distribution type sensing measuring cable of parallel spiral transmission line structure |
| CN104598996A (en) * | 2015-02-02 | 2015-05-06 | 北京交通大学 | Prediction method of surface deformation due to construction based on least square support vector machine |
| CN105735370A (en) * | 2016-03-02 | 2016-07-06 | 铁道第三勘察设计院集团有限公司 | Foundation settlement deformation prediction method based on Rayleigh waves |
-
2016
- 2016-07-13 CN CN201610560335.8A patent/CN106197252A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102522148A (en) * | 2011-11-15 | 2012-06-27 | 中国计量学院 | Rock-soil body deformation distribution type sensing measuring cable of parallel spiral transmission line structure |
| CN104598996A (en) * | 2015-02-02 | 2015-05-06 | 北京交通大学 | Prediction method of surface deformation due to construction based on least square support vector machine |
| CN105735370A (en) * | 2016-03-02 | 2016-07-06 | 铁道第三勘察设计院集团有限公司 | Foundation settlement deformation prediction method based on Rayleigh waves |
Non-Patent Citations (2)
| Title |
|---|
| 刘飞等: "连续投影算法在油菜叶片氨基酸总量无损检测中的应用", 《光谱学与光谱分析》 * |
| 童仁园: "基于螺旋线的岩土变形分布式测量技术研究", 《中国博士学位论文全文数据库（电子期刊）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109238128A (en) * | 2018-11-06 | 2019-01-18 | 江苏柔世电子科技有限公司 | Big strain sensor of flexible inductive and preparation method thereof, big strain transducer |
| CN110243278A (en) * | 2019-07-10 | 2019-09-17 | 浙江水利水电学院 | A kind of rock displacement amount distributed measurement method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101344217A (en) | Measuring device and method for buried pipeline geomagnetically induced current and pipeline-ground potential | |
| CN109541695B (en) | Method for rapidly imaging visual resistivity of remote region of electric field gradient of artificial field source frequency domain | |
| CN105044793B (en) | A kind of inversion method and device of multiple tracks transient electromagnetic detecting data | |
| CN107228772A (en) | A damage assessment method for shield tunnel structure | |
| CN103207412A (en) | Method for detecting solution leaching and groundwater pollution scope of acid in-situ leaching of uranium | |
| CN109579726A (en) | A kind of long gauge length distribution type fiber-optic Brillouin sensing-demodulating system and strain measurement method | |
| CN103941152B (en) | A kind of cable mixed line fault distance-finding method of k-NN algorithm based on waveform similarity | |
| CN107272068A (en) | It is a kind of to utilize the method without anti-interference across the hole resistivity method Underground space of cable | |
| US20200110050A1 (en) | Reflectometry devices and methods for detecting pipe defects | |
| CN104897995A (en) | Transformer station grounding network corrosion detecting system based on earth surface potential, and method thereof | |
| CN104898024A (en) | Transformer station grounding network fault diagnosis method based on synthesis of surface potential and magnetic induction strength | |
| CN106197252A (en) | The method for building up of ground deformation position distribution measuring model based on parallel spiral transmission line | |
| CN117307980A (en) | A method for electromagnetic monitoring and detection of pipeline integrity status and abnormal point positioning | |
| CN106767583B (en) | Longitudinal profile Equivalent Pile diameter calculation method for pile detection sound wave transmission method | |
| CN203307792U (en) | Micro-scale pore pressure static sounding probe for effectively identifying ultra-thin soil layer | |
| CN115522914A (en) | Radial long-distance high-precision detection method and system for cased reservoir | |
| Aghda et al. | Comparison of performance of inclinometer casing and TDR technique | |
| US20180245456A1 (en) | Defect evaluation using holographic imaging | |
| CN110174593A (en) | The apparatus and method of grounded screen breakpoint location are judged using electromagnetic induction | |
| CN105092706A (en) | Belt-type seismic wave imaging system and method for tree trunk interior abnormity | |
| CN105445330A (en) | Method for measuring distribution and depth of cracks in soil | |
| CN103207322B (en) | Techniques in Electromagnetic Measurement is utilized to remove the method for ground unrest | |
| CN104833723B (en) | A kind of forest textural anomaly seismic wave CT round-looking scan device and methods | |
| CN112083507A (en) | Transient electromagnetic rescue well detection method based on empirical mode decomposition | |
| CN116973708A (en) | Sea cable partial discharge key parameter inversion method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161207 |