جستجوی مقالات مرتبط با کلیدواژه « آب » در نشریات گروه « مکانیزاسیون کشاورزی »

The purpose of this study was to identify, prioritize and control the factors affecting water productivity in sugar cane production. In this study, to calculate water productivity, the indices such as BPD and CPD, NBPD is used. Given that the research goal was to prioritize the factors affecting the productivity of water input during the sugar cane production process, the identified factors were evaluated using Analytical Hierarchy Process Analysis (AHP). Then, CART and CHAID decision trees were used in modeling the water input efficiency and the factors influencing it. According to the results, the average CPD, BPD and NBPD indices for sugar cane were 2.37 kg m-3, 1082.71 toman m-3, and 528.03 toman m-3, respectively. The variables of economic productivity of water, the value of sales of the product (income), the amount of sugar produced, production costs, water consumption per irrigation interval, irrigation intervals, electrical conductivity of the soil after harvest, river electrical conductivity, mean time of each irrigation interval, Plant height, plant age, number of irrigation cycles, drainage water drainage PH and width of the cultivating line before discovery are the most important and influential variables in the decision tree models of CHAR and CHAID. The accuracy of the CART model in training and testing was 96% and 92%, respectively, and the accuracy of the CHAID model in education and testing is 97% and 90%, respectively. Also, based on the prioritization of the factors affecting productivity, using the hierarchical analysis method, the quantity and quality of irrigation water, climate conditions, plant conditions, managerial and human factors, and soil conditions were ranked with a coefficient of 0.459, 0.231, 0, 0.091 and 0.069, respectively.

In this research some water quality parameters in fish pond includes pH, Electrical Conductivity (EC), Total Dissolved Solids (TDS) and Turbidity (Turb) were determined by standard methods and predicted by image processing using smart phone and artificial neural network. All experiments carried out in Kappur ponds in Sonqor city, Kermanshah province. Samples collected from three different depths. The 12 parameters consisted of 6 color features (red, green, blue, black, gray and white), and 6 tissue features (mean, standard deviation, softness, third torque, uniformity and entropy) were extracted from image samples and were selected as inputs to the neural network model. Based on the results, network with structure of 12-15-4 (12 neurons in the input layer, 15 neurons in the hidden layer and 4 neurons in the output layer) was the best model for predicting the parameters with R2 of 0.913, 0.993, 0.994 and 0.958 for pH, TDS, EC and Turb, respectively. These values for RMSE were 0.054, 1.835, 3.766 and 0.262, respectively.