機(jī)械外文文獻(xiàn)翻譯--電容層析成像（ECT）傳感器檢測土壤水分【中文6600字】 【PDF+中文WORD】
機(jī)械外文文獻(xiàn)翻譯--電容層析成像（ECT）傳感器檢測土壤水分【中文6600字】 【PDF+中文WORD】,中文6600字,PDF+中文WORD,機(jī)械外文文獻(xiàn)翻譯,電容層析成像（ECT）傳感器檢測土壤水分【中文6600字】,【PDF+中文WORD】,機(jī)械,外文,文獻(xiàn),翻譯,電容,層析,成像,ECT,傳感器
Soil Moisture Detection Using Electrical Capacitance Tomography (ECT) Sensor
Nurzharina Binti Abd. Karim
Electrical and Electronics Engineering Department Universiti Teknologi PETRONAS
31750 Tronoh, Perak Darul Ridzuan, Malaysia zharinabdkarim@gmail.com
Idris Bin Ismail
Electrical and Electronics Engineering Department Universiti Teknologi PETRONAS
31750 Tronoh, Perak Darul Ridzuan, Malaysia idrisim@petronas.com.my
Abstract— This paper briefly discusses the measurement on soil moisture distribution using Electrical Capacitance Tomography (ECT) technique. ECT sensor with 12 electrodes was used for visualization measurement of permittivity distribution. ECT sensor was calibrated using low and high permittivity material
i.e. dry sand and saturated soils (sand and clay) respectively. The measurements obtained were recorded and further analyzed by using Linear Back Projection (LBP) image reconstruction. Preliminary result shows that there is a positive correlation with increasing water volume.
Keywords-soil moisture distribution; ECT sensor; dielectric constant; volumetric water content
I. INTRODUCTION
Estimation of soil moisture content is important in agriculture especially during irrigation scheduling. Changes in the percentage of soil moisture during the irrigation event can be controlled by indicating the amount of applied water to soil. Continuous monitoring of soil moisture content within and below the rooting zone can facilitate optimal irrigation scheduling aimed at minimizing both the effects of water stress on the plants, and also leaching of water below the root zone which can have adverse environmental effects [1].
In this experimental work, ECT is used to measure soil moisture content based on the image of moisture distribution in soil. The soil moisture measurement is based on the concept of the dielectric constant in a dry material consisting soil particles and air is relatively small (1.5 to 4) whereas the dielectric water constant is larger (80 at room temperature) [15]. The small amounts of water in soil cause the dielectric constant of the resultant soil-water-air mixture to exhibit a composite dielectric constant that can be related to the soil moisture content through a simple calibration procedure [15].
Time-domain reflectometry (TDR) method, frequency- domain reflectometry (FDR) method, amplitude-domain reflectometry (ADR) method and capacitive method are the current methods available for the electrical measurements of the dielectric constant [12]. E.g. TDR instruments use measured pulse travel times to determine the apparent soil dielectric permittivity [14]. Some of the researchers found that the capacitance probe method was independent of soil type within wide ranges of soil moisture levels [2]. However, these methods do not determine the image distribution of the
moisture in the soil. In ECT, the image distribution of permittivity material can be seen in the form of image pixel depending on the scale of the permittivity material from lowest to highest value. ECT has been developed since 1980s for visualization and measurement of a permittivity distribution in cross section using a multi-electrode capacitance sensor [21]. The ECT sensor measures the distribution of permittivity of the soil moisture inside the insulating frame and captures the image distribution and measurement data.
II. SOIL MOISTURE
Soil moisture content impacts crop growth directly and also influences the fate of agriculture chemicals applied to soil [20]. One of the key components in managing irrigation scheduling is by determining the moisture content in soil. Zazueta and Xin (1994) reported that soil moisture content may be determined via its effect on dielectric constant by measuring the capacitance between electrode pairs implanted in the soil [7]. Other than that, the soil physical condition plays an important role in soil moisture determination. In normal soil condition, air and water are filled in the pore spaces of soil, thus the dielectric constant of the soil-water-air mixture can be measured. The dielectric constant for air is 1 while dielectric constant of water is 80 (room temperature). When volume of water content in the pore spaces increases, the dielectric constant of the soil will increase. Fig. 1 shows that the unsaturated soil is composed of solid particles, organic materials and pores [4]. The pore spaces of the soil contain air, water and root.
Root
Soil particle
Water Air
Figure 1. Illustration of soil physical contains soil particles, water, air and plant?s root [4].
978-1-61284-896-9/11/$26.00 ?2011 IEEE
The terminology of moisture content stated in ASTM D4643-08 is the ratio, expressed as percentage, of the mass of “pore” or “free” water in a given mass of soil to the mass of the solid particles [19]. The moisture content in soil can be analyzed by knowing the soil physical conditions of the soil. The soil?s bulk density, Db is weight of dry soil to the unit volume (solids and pore). Particle density, Dp involves measuring the weight of dry soil and the volumetric soil particle i.e. only solid and no pore space. From both bulk density and particle density values, the soil porosity can be calculated by using equation (1) [6] [17]:
A. ECT Sensor
There are several components need to be considered in designing ECT sensor. ECT sensor consists of four main parts [21]:
· Measurement electrodes.
· Outer, axial end and radial screens.
· An insulating frame.
· Coaxial cables and connectors.
Soil porosity= 1 - Db
Dp
(1)
The basic method is to surround the vessel or insulating frame with a set of electrodes (metal plates) and the capacitance measurements is measured between each unique
Moisture content, θm determines the amount of water with a given mass of soil. Moisture content is calculated by dividing the weight of water in the soil by the weight of dry soil. The expression of water content in terms of volume of water per volume of soil can be determined as volumetric water content, θv using equation (2) [6] [17]:
pair of electrodes. Typically, measurement electrodes of ECT sensor are placed outside insulating frame with two axial ends and outer screen. Two earthed axial end screens at both ends of the measurement electrodes can reduce external noise to some extent and have a negative effect on capacitance measurement because the electrical field is dragged to the earthed axial end screens [21]. Coaxial cables and connectors
qv = Db ′qm
(2)
are used to connect ECT sensor to the data acquisition unit. Key elements in ECT are the number of electrodes, length of electrodes, external or internal electrodes, earthed screens, and
From the ratio of volumetric water content and soil porosity, the amount of pores filled with water i.e. soil saturation can be calculated.
III. ECT
An illustrated ECT system shown in Fig. 2. consists of ECT sensor with 12 electrodes for soil moisture as detection device, data acquisition unit to process data, and a personal computer for data storage, image processing and display. In ECT the changes in inter-electrode capacitance due to the change in concentration and/or distribution of dielectric materials in the region are measured, and a cross-sectional image representing the permittivity distribution inside a pipe or vessel is reconstructed [16].
Data cable RG1741/U
Measurement electrodes
driven guard electrodes. The details of the key elements in ECT sensor are discussed by Yang (2010) [21].
In this experimental work, ECT sensor of 12 electrodes was constructed with electrodes length of 10.0cm and 1.0cm width, and the space between a pair of electrode is 0.6cm. The electrodes are mounted outside the insulating frame of 6.0cm diameter. There are two axial ends in the sensor which act as grounding portion. The material for outer earth shielding is aluminum. The outer shielding is connected with two axial ends as grounding purposes. Two types of measurement in ECT system are calibration and online measurements. ECT sensor is calibrated using two different test mediums as low permittivity material (dry soil) and high permittivity material (saturated soil). In online measurement, the ECT sensor is filled with soil and known amount of water is added in the soil.
The calibration in lower and higher permittivity material gives lower and higher capacitance (Cl and Ch) values respectively while online measurement gives measured capacitance value, Cm. All subsequent measured capacitance values, Cm are then normalized to have values, Cn between “0” (when the sensor is fully filled with the lower permittivity material) and “1” (when fully filled with higher permittivity material) according to the equation (3) [22]:
n
C = Cm - Cl
(3)
ECT Sensor Data Acquisition Unit Computer Figure 2. ECT system.
Ch - Cl
B. Data Acquisition System
Data acquisition unit measures the capacitance between electrode pairs and converts the measurement data into an image permittivity distribution. The high speed USB cable connects the data acquisition unit to the control computer. Data from the unit is then sent out to a control computer for data storage, processing and display. According to Yang (2010), to obtain a complete set of data for one image, electrode 1 is used for excitation and electrodes 2-12 for detection, obtaining 11 capacitance measurements [21]. The number of independent capacitance measurements can be
V. DIELECTRIC CONSTANT MEASUREMENT
For soil measurement, the dielectric constant is the ratio of the capacitance of the soil divided by the dielectric permittivity of free space. The relationship between the changes in water content and the dielectric constant of the medium depends upon soil type and the frequency range of the measuring apparatus [1]. From the relationship of water content and permittivity, several empirical models exist. The empirical equation that relates to volumetric water content and dielectric constant of material has been derived as in equation (6) [9]:
calculated by N (N-1)/2, where N is the number of electrodes. 2 3
Based on the number of independent capacitance measurements equation, there are 66 independent capacitance
qv = A + Be a + Ce a + De a
(6)
measurements for 12 electrodes in ECT sensor.
IV. IMAGE RECONSTRUCTION
Image of permittivity distribution in 12 electrodes ECT sensor with 66 independent measurements is projected onto a (32x32) square pixel grid. On a (32x32) square pixel grid there are 1024 pixels, but only 812 pixels are needed to construct the cross-sectional image of the vessel [18]. With 812 pixels, there are remaining pixels that are not being used because they are located outside the measurement area. The color scale used is from blue to red i.e. “0” to “1” to indicate the range of the lower permittivity material to higher permittivity material. In ECT, the measuring capacitance between two electrodes can be expressed by the following integral equation (4) [11]:
where θv is the volumetric water content, and εa is dielectric constant of material. E.g. the parameters value for A = -8.63, B
= 3.216, C = -9.54x10-2, and D = 1.579x10-3 for most sands
based on the calibration coefficients according to soil texture [9]. The parameters A, B, C, and D are differing for different types of soil texture and it is difficult to analyze. Topp et al. (1980) equation is used to provide a basis for comparison among soils [13]. The empirical equation by Topp et al. (1980) is generally applicable to coarse grained mineral soils [20]. Volumetric content from the range of 0 to 0.55 is used in the empirical relationship. Moreover, it is widely used in determining the volumetric water content and dielectric constant in soil using TDR and oven-dry. Equation (7) shows the empirical equation of third-order polynomial relationship based on volumetric water content and dielectric constant which is expressed by using equation (7) [8] [10] [15][20]:
C = òò G( x, y ) × S( x, y,G( x, y ))dxdy
D
(4)
qv = -5.3′10-2
+ 2.92′10-2
e a - 5.5′10-4
a
e
a
e 2 + 4.3′10-6 3
(7)
e a
One of the most commonly used reconstruction algorithm for image reconstruction in ECT system is the LBP because it is the simplest and fastest system for image reconstruction process. The LBP has been implemented to reconstruct images
Meanwhile, the calibration relationship for volumetric water content for most soil is expressed in refractive index model as in equation (8) [14] [20]:
for ECT sensor using 12 electrodes and transducer-based multiprocessor system [3]. With LBP, the permittivity distribution in ECT can be determined. The relationship can be
qv = a + b
(8)
seen in linear normalized form as in equation (5) [11]:
C = SG
(5)
The coefficients value for a and b parameter fall within the
range in TDR calibrations and the mean values of a = 0.110 and b = -0.180 [14]. The refractive index model is obtained by minimizing the sum of squared differences between the predicted permittivity and the permittivity according to equation (8) [20]. The refractive index model is a baseline
where C is the normalized capacitance matrix, S is the
transducer sensitivity matrix which contains the set of sensitivity matrices for each electrode pairs, and G represents the normalized permittivity. The product of G is multiplication of the transpose of the transducer sensitivity matrix with the normalized capacitance matrix.
equation developed as a reference for comparing different individual soil calibrations [14]. For Topp et al. (1980) empirical equation and refractive index model equation, only the real component of permittivity is considered. Soil bulk density and moisture content are used to calculate volumetric water content value.
VI. EXPERIMENTAL DESCRIPTIONS
The relationship between apparent soil permittivity and volumetric water content has been subject of much research in the last 30 years [5]. The experimental work has been conducted in the laboratory by using ECT system as the test apparatus. ECT sensor is connected to the data acquisition unit by using connectors that connect to the electrodes in ECT sensor via coaxial cables for 1 to 12 electrodes according to the plane arrangement on the device. The calibration procedure starts with the calibration of lower permittivity material i.e. dry soil and continued with higher permittivity material i.e. saturated soil. This calibration is important in order to measure the normalized capacitance value, Cn as in (3). Calibration is also used as lowest and highest set point values for online measurement data.
The online measurement is conducted after the calibration measurement has been performed. ECT sensor is filled with dry soil i.e. sand or clay (filled inside the insulating frame as illustrated in Fig. 3). An amount of water is added into the soil in the ECT sensor, i.e. 10ml, 20ml …, and 400ml accordingly. The data is captured for time interval of 10 seconds. The same procedure is repeated for the several times for data comparison and analysis. From soil moisture distribution image in ECT, the percentage of the soil moisture in the image pixel is calculated by referring to the color scale of from “0” to “1”. The moisture percentage and bulk density is used to calculate the volumetric water content, θv of the test soil.
Other than that, oven-drying method is a standard test
method to determine the soil moisture content. The oven- drying technique is probably the most widely used of all gravimetric methods for measuring soil moisture [7]. Microwave heating is a process by which heat is induced within a material due to the interaction between dipolar molecules of the material and an alternating, high frequency electric field [19]. In oven-drying method, a sample of soil is dried in the microwave oven for 24 hours with 105oC temperature in order is to ensure that there is no moisture content exists in the sample soil. The weight of dry soil is measured as well as the weight of wet soil. According to ASTM D4643-08, the water content of soil can be calculated as equation (9) and (10) [19]:
VII. RESULTS
A. ECT Calibration Measurement
The tomogram image has its unique color scale i.e. blue that indicates “0” (low permittivity material) and red indicates “1” (high permittivity material). In between blue to red there are other colors indicating different scale from 0.001 to 0.999 approximately. Fig. 3 shows lower calibration using dry sand and saturated sand for high calibration while Fig. 4 shows the image of the calibrations.
Low calibration (dry sand) High calibration (saturated sand) Figure 3. Low and high calibration for sand.
0
1
Figure 4. Low and high calibration images.
B. ECT Online Measurement
In online measurement, the soil moisture distribution for different volume of water content in soil can be verified in TABLE I. The types of soils used in this measurement are sand
w = massof water massof ovendriedsoil
x100%
(9)
and clay. In this image distribution, the color scale shows that, the sensitivity of the electrodes in ECT sensor depends upon the volume of water and air in the soil. As the volume of water added increases, the moisture percentage increases and the color scale moves towards “1” or red color that indicates
w = m1 - m2 x100%
m2 - mc
(10)
higher permittivity value of the test soils. It shows that the soil reaches its saturation value when the pore spaces of the soil are fully filled with water content. From the images shown in
where w can be denoted as θm i.e. the moisture content, mc is the mass of container (g), m1 is the mass of container and wet soil (g), and m2 is the mass of container and dry soil (g). From
(9) and (10), the moisture content value is substituted into (2) to calculate the volumetric water content, θv by knowing the soil bulk density. The volumetric water content from online measurement in ECT and oven-drying methods can be substituted into empirical equations to calculate the dielectric constant measurement.
TABLE I, it can be seen that the moisture distributions in sand are evenly distributed than clay. This is due to the soil properties itself where the sand has larger soil particles and pore spaces that allows the water to flow through it easily. Meanwhile, in clay, the water is hardly distributed when the volume of clay is larger than volume of water. A mixture of higher volume of clay and smaller amount volume of water results in a mixture that can be shaped.
TABLE I. IMAGE OF SOIL MOISTURE DISTRIBUTION IN ECT
Water volume (ml)
ECT image distribution
Percentage of moisture distribution in ECT image pixel (%)
Sand
Clay
Sand
Clay
0
0
0
80
18.09
17.03
160
36.22
33.43
240
56.88
51.50
320
78.13
72.65
400
97.30
91.02
From the ECT image distribution in TABLE I, when 80ml water is added into dry sand, the percentage of moisture distribution obtained is 18.09% by calculating the color scale ratio in the ECT image pixel. The changes in color scale indicates an existence of water content in that particular area. The color scale turns to red when the percentage of moisture in the sand 56.88% i.e. with 240ml water is added. This result shows that the percentage of moisture increases as the volume of water added in the sand increases. The saturation level is indicated when the pore spaces in the sand is fully filled with water. For the image of moisture distribution in clay, the percentage of moisture is lower than sand. The clay reaches its saturation l
收藏