Journal of Researches in Mechanics of Agricultural Machinery

Abstract Order Rodentia is a massive group of animals. In case the population of which is not controlled, their sprawl and propagation will impose severe problems for humankind. One way to prevent and fight the growth of the Rodentia population is by poisoning their burrows, which requires finding the precise location of the holes made. In this study, a fast and intelligent algorithm for determining the position of rodent holes was presented. A quadcopter with a 1m.s-1 propeller speed at a constant 1m height from the ground was used for aerial filming. The filming rate was 25 frames per second. Movies prepared by the camera were transmitted to a computer via wireless communications and converted into images by Aoao software. From images, red, green, blue, and gray surfaces and co-occurrence matrices were extracted, from which, in turn, 88 color and 44 texture features were extracted. Once top features, most of which related to the texture of images, were selected, they were used to train and implement a support vector machine (SVM) classifier with the radial base kernel. The maximum classification accuracy was obtained with a Kernel breadth of 0.1. This method 
detected the rodent nest hole with very high accuracy (100%). The minimum and maximum differences between the exact and predicted positions were 1 and 24 cm. 

Abstract Solar energy as a renewable energy source can be used to produce heat and electricity. Among solar energy power production systems, photovoltaic thermal (PV/T) technology is almost a new and widely-used hybrid technology that can simultaneously produce heat and electricity. In recent years, nanomaterials in solar energy systems have significantly increased. In the present study, the effect of using nano-fluid water-aluminum oxide at three volumetric concentrations (0.10, 0.07, and 0.05%), and three inlet flow rates (0.6, 0.9, and 1.2 l/min ) on the electric efficiency, thermal efficiency and exergy efficiency of PV/T system at different times of the day were numerically calculated using CFD. Then, their results were compared with the observed experimental results. The inlet flow rate of 1.2 liters min-1 and the aluminum oxide nano-fluid concentration of 0.10% were optimized as input points due to the proper cooling of the system. The thermal efficiency is a function of the difference between the inlet and outlet temperatures of the fluid, and its highest value was observed at the maximum irradiation condition (960 W/m2 ) with 53%. Additionally, the highest and lowest exergy efficiency values were 21.6% and 15.9%, respectively, in the afternoon. 

Abstract In this study, a semi-automatic strawberry harvester was developed, and the performance of the system was evaluated. In the system, the operator located on the machine picks strawberries from the plant and puts them on the tray, and the harvesting process is performed. The system can change the forward speed and harvesting height. In three replications, the effect of harvester forward speed at three levels of 0.5, 1, and 2 km/h and harvesting height at 50, 65, and 80 cm on the percentage of harvest, losses, and plant damage were evaluated. The experimental data were analyzed using SAS software based on the factorial test in a completely randomized design. The results showed that the effect of the two discussed factors on the percentage of harvest, losses, and plant damage at the probability level of 1% are significant. However, the effect of harvest height on losses at the probability level of 5% was significant. Also, their interaction was significant for the percentage of harvesting and loss at 1% and plant damage at 5%. The best combination of parameters for the system, including harvester forward speed of 0.8 km/h and placement height of 60 cm, were obtained by Response surface method and Design expert software. 

Abstract One of the most important raw materials in the pharmaceutical and food industry is the essential oil of medicinal plants. The originality and purity of the essential oil is an important subject from an economical and practical point of view. Conventional methods for identifying and evaluating the authenticity of essential oils have weaknesses, and the use of electronic nose systems covers these disadvantages. In this study, an electronic nose system consisting of 8 metal oxide semiconductor sensors was designed to determine the purity of peppermint essential oil. The principal component analysis (PCA) method was used to reduce the dimensional of the data and identify the effective sensors in detecting the degree of purity and eliminating useless sensors. Then, classification accuracy was calculated by using the classification techniques of SVM. Based on the results, the PCA method with two main components, PC1, and PC2, explained 82% of the data variance. Also, based on the loading diagram, the sensors that had the most impact on detecting the degree of purity of essential oil, including TGS822, MQ8, MQ3, MQ5, MQ9, and TGS813, were identified. The diagnostic algorithm based on the support vector machine was also able to separate the samples of essential oils with a purity of zero to 100% from each other with an accuracy of 75%. Therefore, the proposed electronic nose system based on the mentioned  algorithm could detect the purity of peppermint essential oil with acceptable accuracy. 

Abstract Increasing the working width of agricultural machinery has benefits for the farmer that can reduce fuel consumption, reduce soil compaction, reduce crop crushing, reduce the depreciation of tractors and parts, and speed up the process of farm operations. In this research, a method has been proposed to increase the spray width of fertilizers. In this method, the removal of mechanical force to throw granular fertilizer and its replacement with air flow is proposed. In the proposed design, the airflow was provided by two screw blowers whose orifice is inclined to the sides of the tractor and produces a large wind flow. The tractor power axis provided the required rotational force of these two blowers and  agitators inside the fertilizer tank and a trolley wheel system with a ratio of 1 to 4. The manure is guided into the transfer pipes by an adjustable valve installed under the tank, and the screw blowers draw the manure inwards by suction and direct it outwards with high blowing force. As a result, the fertilizer is thrown to a much greater distance than the  conventional method. After completing the construction and installation of components, the device was evaluated in the field. In the evaluation of this device, the aim was to measure the spray width and uniformity. For this purpose, the device was evaluated in field conditions. Test treatments include advance speed (in terms of kilometers per hour in two levels), blower speed (in terms of percentage of blower valve opening in two levels), and fertilizer shedding rate (in terms of  percentage of sliding valve opening in two levels) and Traits also included spray uniformity and spray width. To  investigate the interaction effect of different factors, a factorial test with a randomized complete block design and SPSS software was used for the statistical analysis of data. The tests were performed in three replications, and to compare the results, the analysis of variance and Duncan's test 5% probability level were used to determine the statistically different independent parameters. The optimal progress speed of this device was 10 kmh-1 , its spray width was 46 m and its spray uniformity was measured and determined on average 5 grm-2. 

Abstract Soil is one of the most important sources of production in agriculture. Therefore, with the determination of soil and its important characteristics, proper management and sustainable use of agricultural lands can be achieved. The current study aimed to predict the soil texture using a machine vision system and deep convolutional neural network (DCNN) algorithm. The proposed CNN model was composed of two blocks, including convolutional layers, max pooling layers, a dropout layer, batch normalization layers, fully connected layers, and a support vector machine classifier. This model was trained and tested on the images of different soil samples (11 types of soil texture and a total of 790 soil sample images). The data is prepared by a machine vision system and a smartphone camera (Galaxy A8). Using the confusion matrix, important statistical parameters such as accuracy, precision, specificity, sensitivity, and area under the curve were obtained at 99.65%, 98.75%, 99.8%, 98.75, and 99.27%, respectively. The suggested model successfully and correctly classified the soil sample images with 98.1% accuracy. The obtained results indicated that this study's implemented deep learning model can be a proper alternative to costly and time-consuming laboratory methods for determining soil texture. 

Abstract Material handling systems are widely used in different sections of industry and agriculture. Pneumatic conveyors are methods of conveying granular and powdered materials that use airflow to transfer materials from one point to another. Particle conveying pattern and energy consumption of such conveyors are the parameters affecting the operation of this type of conveyor. In this study, to investigate the material conveying pattern and optimization of energy consumption, a divergent cross-section mechanism was designed and developed in the conveying pipeline. Its effect on barley grain conveying in a pneumatic conveying system was investigated experimentally numerically. Fluent software was used to solve the two-phase flow equations such as continuity and momentum, and for experimental evaluation, a factorial experiment in the form of a randomized complete block design with three replications was used. Duncan's test was conducted to compare the means of the treatments. The results of experiments and simulations of pipelines with divergent cross-sections indicate that the average inlet air velocity, the average pressure drop, and the average power consumption of the device in the dense phase in the auxiliary air mode were 8.6%, 6.67%, and 1.23% less than those obtained in no auxiliary air mode, and in the dilute phase in the auxiliary air mode were 3.56%, 4.66% and 5.21% less than the corresponding values in no auxiliary air mode, respectively. Also, the inlet of auxiliary air at a velocity higher than the inlet air of the environment increased the mass flow rate of the conveying particles by 17.3%. The particle velocity variation decreased in the divergent section, and the particle velocity became more uniform along the conveying path. 

Abstract One way of improving the thermal performance of greenhouses is using thermal insulators (thermal shading). In this study, the behavior of a novel dynamic shading system made of two-layer polycarbonate plates containing fluid was investigated by numerical method. In this system, according to the cooling and heating requirements also to control the incoming or outgoing radiation from the greenhouse, the appropriate fluid was pumped into the ducts between the polycarbonate plates by a hydraulic pump. Studying the hydraulic parameters, including intra-duct pressure and fluid velocity between plates, can be very useful for the proper design of the shading to minimize the energy required and run the system. The computational fluid dynamics (CFD) method and standard k-ε turbulence model simulated turbulent flow inside the shading ducts. Validation between experimental and numerical data showed an average difference of less than 5 percent between the experimental and numerical pressure drop data. The maximum velocity value of 0.96 ms-1 was obtained for the ducts inside. In comparison, the system pressure drops in single (series) and triple modes (parallel) it was 3793 Pa for the latter and 3439 Pa for the former, which indicates 9.3% decrease in pressure drop for the innovative and proposed design. 

Abstract Due to the fact that persimmon is a pressure-sensitive fruit, and it is difficult to store this fruit in the storehouse, in this study, the effective parameters for reducing color changes and its physical properties have been investigated. These  parameters were three types of packaging (foam containers with polyolefin film, polyethylene terephthalate, and ordinary box of wood) and three coatings (polyamines with concentrations of 1 and 2 mm, distilled water). Then, characteristics such as persimmon fruit color parameters including L*, a *, and * b, chroma index, and color change index as a  percentage of changes in the pre-and post-storage stage were investigated. The response surface method was used to analyze the data. According to the obtained results, the highest percentage of changes for the percentage of changes L* is in packaging with an ordinary wooden box, and the least changes are obtained in a foam container with polyolefin film and putrescine coating of 1 mm. These results were similar for b* factors, chroma index, and total color changes. For a*, the highest percentage of changes was when polyethylene terephthalate coating was used. The use of 1 mM putrescine and a foam container with polyolefin film had the lowest percentage of changes in the studied parameters. 

Abstract The use of optical sensors in assessing crop health, plant fertilizer demand, and yield prediction is the new method in the agricultural sector. Grennseeker optical sensor is one of the latest tools that has recently gotten much attention from worldwide farmers and researchers. Detailed examination of the variables mentioned above and measuring canopy coverage can improve production management and minimize input consumption. Therefore, this study aimed to investigate the GS sensor's ability to assess the nutritional status and surface area of the canopy and predict the performance of the garlic crop. In this research, five N-fertilizer treatments of N0, N50, N100, N150, and N200 were applied to garlic in a randomized complete block design with three replications. The image processing results in Labview software showed that the best separation performance for the elimination of background from the canopy could be achieved with the function of G-R. The results revealed that the best correlation coefficient between NDVI and LAI was 0.69, but no significant relationship was found between N and NDVI. Finally, it can be concluded that the GS could not estimate the N level of garlic. However, it could perform well in estimating tuber yield and canopy performance  satisfactorily.