DESIGN AND DEVELOPMENT OF AN AFFORDABLE SOIL MOISTURE SENSOR AND SYSTEM FOR IN-SITU MEASUREMENTS
The rapid progress of the technology has opened avenues for the enhancement of the productivity of resource-challenged agricultural practices in developing countries. In a country such as India, where dry farming is pervasive, efficient optimum irrigation technology would be valuable. To enhance the crop productivity, optimum irrigation practices need to be followed in the agricultural fields. Moreover, optimum irrigation is not only required in dry farming, but also in all locations in which irrigation is expensive or waste of water is forbidden. Optimum irrigation helps to enhance the crop productivity and also to conserve water. To maintain the optimum irrigation in agricultural fields, soil-moisture sensors are widely used. Among various affordable soil-moisture sensors, the dual-probe heat-pulse (DPHP) sensor is one of the main candidates due to its optimum price and accuracy. The DPHP sensor has excellent potential in soil-moisture measurements. However, DPHP sensor dissipates high power, its measurement is dependent on the soil temperature, and has narrow cylindrical sampling zones. These shortcomings restrict the use of DPHP sensors for in situ soil-moisture measurements. These requirements and constraints are motivation for this thesis with focus on the design of low-power, temperature-independent, and wide sampling zone for the DPHP sensors.
A DPHP sensor has two probes: a heater probe and a temperature sensor probe that is placed at a distance of 3 mm from the heater probe. Reported DPHP sensors consume as high as few Watts of power while developed DPHP sensors in this thesis dissipate only 165mW, which is among the lowest in this category of sensors. Besides keeping the power consumption low, methods for the calibration is the primary challenge in the heat-pulse technique for in situ deployments. Standard calibration in the heat-pulse technique is performed on the agar stabilized water, and specific heat of the agar is estimated. However, for the in situ measurements, calibration on the standard soil would be preferable considering the soil texture variations in the Indian agriculture field for the heat-pulse sensors. Thus, we have developed customized calibration process, which includes estimation of specific heat of the dry soil. The fabricated DPHP sensor is used to determine the specific heat of the dry soil of the sand and white clay as standard soil samples and results are benchmarked with the literature. The customized calibration process is validated by measuring the moisture content using the DPHP sensor and benchmarking with the standard oven-dried gravimetric technique. With proper deployment in the soil, fabricated DPHP sensor, reported in this thesis, measures the soil moisture with a maximum acceptable difference of ? 3 % with reference to the standard oven-dried technique, for agriculture applications.
SoilgSenS 20.2, an automated and self-sustained system using the developed low power DPHP sensor with a standard procedure for the calibration is developed. The system stores the data in a micro SD card. The developed system automatically turns on the DPHP sensor at one-hour intervals and enters into a sleep mode for 54 minutes, thus requiring no manpower. It is worthwhile to mention that SoilgSenS 20.2 is the first reported affordable DPHP-based field prototype sensor system. The cost of the SoilgSenS 20.2 is Rs. 4000, which is almost three to ten times less than commercially available moisture sensors such as capacitive-based and TDR sensor systems.
The soil temperature in the agriculture fields varies from time to time during a 24-hour period, and it should not affect the results of the soil moisture measurement by the in situ sensor. In order to use field deployable DPHP sensors, it is important to minimize the effect of the temperature of the soil. For this purpose we have developed a temperature compensation method for in situ DPHP sensors. Accordingly, the existing soil-moisture model that is used for DPHP devices is modified and used for the temperature compensation. Field measurements indicate that volumetric moisture content when measured without the developed temperature compensation leads to an error of about 3% and with temperature compensation the error is reduced to 0.5%.
In sensing soil moisture, there are inherent field-related issues such as uncertainty and unpredictable distribution of minute amounts of trapped water, and morphological changes of soil such as lumps and cracks due to desiccation. These issues pose problems for the accurate measurements performed using DPHP sensors that use narrow cylindrical sampling zone between the heater and the temperature probe. To remedy these shortcomings, we have designed and developed a multi-point heat-pulse (MPHP) sensor with three temperature probes that have enhanced the sampling zone to a volume of 840 mm3, which is three times that of the DPHP sensor. Deployed DPHP and MPHP sensors with soil spatial variations have revealed that the maximum difference between the measured percentage of moisture from the DPHP sensor and the standard oven-dried instrument in such specific conditions is more than 10%, whereas for the MPHP sensor, the error is only around 3%. Furthermore, when cracks appear in the soil mass, it is observed that the error of the DPHP sensor is around 16% while that for the MPHP sensor is still only 3% when compared with the standard oven-dried method.
For agricultural scientists it is important to measure the penetration of soil moisture across different depths to quantify the moisture content that is available at the root zone. To use the conventional soil moisture sensor for examining the soil-moisture profile at different depths, either sensor must be stacked in a single probe or their design need to be modified. Staking these sensors into single probe will increase the probe thickness and disturbs the soil matrix. Due to this, the soil moisture measurement may not be accurate. It is also unaffordable for Indian farmers and agriculture scientists. Hence, there is a dearth of affordable thin probe soil-moisture sensor (typically with a thickness of less than 5 mm) that disturbs the soil matrix only negligibly. This has motivated us to fabricate microsensor by which multiple microsensors can be incorporated in a single probe, and the soil-moisture profile at different depths can be studied. For this purpose, for the first time, we have fabricated graphene-based microsensor for the soil-moisture measurements. Graphene oxide (GO) microsensor has a sensitivity of around 340% and 370% when the soil moisture changes from 1% to 55% in red and black soils, respectively. A GO sensor array shows a fast response time of 100?120 seconds in soil moisture measurements. For in situ soil moisture measurements, the diurnal temperature and salt concentration (soil conductivity) are the variable parameters that might affect the sensor?s response. We observed that the GO sensor output changes by 6% when the ambient temperature varies from 25 ?C to 65 ?C leading to only 3% difference in the soil moisture measurements. In the salt concentration (soil conductivity) measurements, we noted that the sensor output changes by 4% when the concentration of salt in the soil sample varies from 0 moles to 0.35 moles and differences across soil-moisture measurements are around 2%.
Graphene-based microsensor has opened avenues for new sensing platform for the agriculture applications. However, for the MEMS-based sensors challenge lies in the packaging, accuracy and calibration for every soil. Like heat-pulse sensors, graphene based sensors also needs copious calibration and field testing to establish it as the standard moisture sensors.
|
