Engineering in Agriculture, Environment and Food最新文献

In this study, the application of a versatile approach for modeling and prediction of the moisture content of dried savory leaves using hybrid artificial neural network-genetic algorithm has been presented. Genetic Algorithm was used in order to find the best Feed Forward Neural Network (FFNN) structure for modeling and estimation of moisture content in the drying process of savory leaves. The experiments were performed at three air temperatures of 40, 60 and 80 °C and at three levels of relative humidity 20%, 30% and 40% and air velocity of 1, 1.5 and 2.0 m/s for drying the savory leaves in the forced conductive dryer. Optimized neural network by GA had two hidden layers with 9 and 17 neurons in first and second hidden layers, respectively. Mean Square Error (MSE) value (0.000094606) and correlation coefficient (0.9992) of FFNN-GA experiments showed that moisture content can be accurately predicted from the input variables: air temperature, airflow velocity, relative humidity and drying time. Moreover, results showed that the optimized neural network topology could denote the superior ability of this intelligent model for on-line prediction of the moisture content of Savory leaves in different drying conditions.

Mango kernel seed oil is rich in unsaturated fatty acid, bioactive compounds and a potential source of butter production. This study explores the unit operation of size reduction which is a vital step in the production of mango seed kernel flour and hence, mango seed kernel oil. The mechanical property of mango seed kernel, its grinding characteristics using hammer mill and then particle size distribution (PSD) of its flour were evaluated. Comparison was made of the PSD using laboratory sieve shaker and laser diffraction technique. The Kick's model had the best energy efficiency followed by Bond's and Rittinger's models respectively. PSD by Laser diffraction method indicated higher mass relative span (RSm), Inclusive graphic skewness (IGS), kurtosis of PSD (Kg) and graphic means. Rosin-Rammler-Bennet (RRB) and Gaudin-Schuhmann (GS) models were effectively used to describe the statistical and model parameters of the particle size distribution data.

Instrumentation and Artificial Intelligence (AI) recognition techniques were developed for automatic pest detection. This system is based on pest detection and monitoring, and it improves the efficiency of vegetable and fruit farming and food production. This paper presents an automatic pest detection system that applies artificial neural networks. The system automatically detects two defoliating pests on potato and bean crops: Mexican Bean Beetle (MBB) and Colorado Potato Beetle (CPB) in the adult stage. The neural classifiers utilized for the beetle detection are RSC (Random Subspace Classifier) and LIRA (Limited Receptive Area). The MBB images that were employed as inputs to the classifiers were obtained on Mexican plantations. The CPB images were collected from various Internet sources. We compared the results obtained with both classifiers on image databases. The RSC classifier demonstrates the better result for recognition, which is 89%, while LIRA presents a recognition rate of 88%. These results are good for pest detection and can be used for the diagnosis of pest locations in crops. The purpose was to contribute to the development of automatic detection applications based on images of potato and bean plantations. In Mexico and other countries, it is of great importance to solve pest problems in agriculture. We chose insect recognition due to the importance of potato and bean crop production and consumption. Pest detection in the adult phase is of high priority because of the high rate of crop defoliation and destruction. Our automatic pest-detection system can be employed in pest recognition in monitoring activities.

The present manual sorting technique is not effective to detect fungal infection in dates; especially at the early stage. The potential of near infrared (NIR) area scan imaging (900–1700 nm together as one image) to detect fungal contamination in three popular varieties of dates (Fard, Khalas and Naghal) was investigated. Date samples were treated as three groups: untreated control (UC), sterile control (SC) (surface sterilized, rinsed and dried) and infested samples (IS) (surface sterilized, rinsed, dried and fungal inoculated). The IS was then incubated for 10 days and imaged every 48hr to obtain 5 infection stages namely IS Day2, IS Day4, IS Day6, IS Day8 and IS Day10. In total, 3150 NIR images (UC + SC + five fungal infection stages × 150 images × three date varieties) were acquired and analyzed. The overall highest classification accuracy was 97, 96 and 100% for two-class, six-class and pair-wise models, respectively while comparing IS with UC. Similarly, it was 94, 89 and 94% for two-class, six-class and pair-wise models, respectively while comparing IS with SC. However, when the developed algorithm was tested on pooled dates images (all three varieties combined), the two class model yielded a higher classification accuracy of 83 and 86% for UC and IS, respectively; and 71 and 85% for SC and IS, respectively. Thus, NIR area-scan imaging has the potential to be used as a fast and cheaper technique to detect fungal infection in food industries.

Existing agricultural detection algorithms mainly detect a single object category (class) under specific conditions which restricts the farmer's ability to utilize them in different conditions and for different classes. What is needed are generic algorithms that can learn to detect objects from examples, thereby reducing the technical burden required to adapt to local circumstances. Among generic algorithms, deep learning methods recently are beginning to outperform other generic algorithms. In this study, we investigate the possibility of using a deep learning algorithm for recognition of two classes (mature and immature strawberry) of agricultural product using a deep convolutional neural network (DCNN) and greenhouse images taken under natural lighting conditions. To the best of our knowledge, this is the first application of deep learning to the detection of mature and immature strawberries as two classes. We evaluated the results using the following parameters: average precision (AP), a parameter that combines detection success and confidence of detection; and bounding box overlap (BBOL) which measures localization accuracy. The developed deep learning model achieved an AP of 88.03% and 77.21%, and a BBOL of 0.7394 and 0.7045 respectively for mature and immature classes.

A half-tracked tractor equipped with an oscillating crawler system has better stability on uneven ground than wheel type tractors, which respond directly to a concavo-convex surface. However, the crawler tension fluctuates according to the variation in the perimeter of the crawler unit, since the hitch point of the oscillating frame is set separately from the drive axis of the drive sprocket. It is suggested that this fluctuation of the crawler tension would influence the driving performance. This study aims to evaluate the relationship between the crawler tension and the driving performance of the oscillating crawler unit using kinematics simulation. The ultimate objective is to achieve traveling control with fluctuation in the crawler tension. This paper introduces a rubber crawler model that is able to simulate the fluctuation in the crawler tension. In the driving simulation, the effect of the elastic parameters of the crawler and support spring on the driving performance of the oscillating crawler unit was examined. It was observed that the driving performance of the oscillating crawler unit is affected by the parameters of the rubber crawler and the support spring of the idle roller.

A computer program is developed in Visual Basic 6.0 to locate the virtual hitch point of tractor with respect to depth of operation and to optimize the three-point linkage parameters for matching the virtual hitch point with the line of pull. The concept of Cartesian coordinate geometry was used in the program development. The virtual hitch point was calculated by solving the straight line equations of lower and upper link of three-point hitching system at particular depth of operation. For determining the line of pull, draft was calculated from ASAE equation and coordinate of centre of resistance was assumed to the 2/3 of depth of operation in the line of centre of gravity of implement. The optimum depth of operation is the depth for which line of pull passes above and near to VHP which was calculated in the program through the iteration process. The locus of virtual hitch point with respect to depth of operation was observed as parabolic. The distance of virtual hitch point from the rear axle centre increases with increase in depth of operation in all cases. The optimum depth of operation for a test tractor was observed as 269 mm using the developed program. This depth depends substantially on lower link (length and link position) and implements parameters (weight and height of hitch point). Therefore, this program may be very useful in simulating the three-point linkage and implement parameters as per the requirement of depth of operation.

Nowadays, due to the shortage of oil-based storages and environmental problems, the tendency toward the use of alternative fuels such as biodiesel and ethanol has been increased. The objective of the current study was to evaluate the energy cycle of biodiesel production from rapeseed oil using transesterification method. The energy consumption in the purification stage of rapeseed oil for edible purposes was analyzed, and the results were then compared with those of biodiesel production. According to the Life Cycle Analysis model, it was found that the input and output energy in biodiesel production from rapeseed oil were 28293.28 MJ/ha and 54727.07 MJ/ha, respectively. Energy indexes involving specific energy, energy efficiency and pure energy were also calculated for biodiesel production (equal to 16.002 MJ/kg, 0.06 kg/MJ and 7986.89 MJ, respectively). The emission of greenhouse gasses was calculated in the cycle of biodiesel production. It was revealed that the maximum pollutant factor was related to diesel fuel in transferring stage of the product to the oil-extraction company followed by agricultural production stage having the shares of 100% and 40%, respectively. Results of the study indicated that the ANFIS model could predict the yield of the product with a high accuracy (R = 0.99, RMSE = 3.94).

This study was to identify proper blowing velocity conditions to remove rice bran during milling. The milling characteristics were measured at four levels of blowing velocity 20, 25, 30, 35 m/s.

The rise of grain temperature of milled rice was the lowest, 16.9 °C at 35 m/s. The differences of cracked ratio and broken rice occurrence ratio between before and after milling were found to be slightly lower at 35 m/s than other blowing velocities. Turbidity was decreased at high blowing velocity, and showed the lowest value at 35 m/s.

At present, farmers who own a sugarcane harvester are becoming more interested in purchasing a sugarcane billet planter. However, the planting consistency of this type of planter is still low. Thus, its metering device should be improved. The objective of this study is to develop the shape and arrangement of the metering device of the billet planter and study its effect to the discharge consistency. The original billet metering device consists of a rubber conveyor belt with flat steel cleats that have a length equal to the conveyor width, while the developed metering device uses 20-degree inclined-edge cleats. The cleats were shortened to ⅔ of the conveyor width and aligned in left-right alternation. The developed metering device was evaluated compared with the original device, considering the precision of billet discharging during the stationary tests. The linear speed of the conveyor was controlled at 0.189 m/s. The test results showed that developed metering devices gave a precision index of 50.67%, which is 9.66% higher compared with that of the original. The developed metering device had higher torque than the first metering device and higher fluctuation.