Efficient use of energy in paddy production prevents the destruction of agricultural ecosystems by reducing greenhouse gas emissions and causes the development and promotion of sustainable agriculture. Meanwhile, intelligence agriculture has come to the aid of farmers and policy-makers by using up-to-date knowledge, to lead to sustainable welfare and comfort of human society in the present and the future. Therefore, the purpose of this study was to investigate energy consumption and production, modeling and optimization of two paddy cultivars yield by Artificial Bee Colony (ABC) and genetic algorithm.
Materials and Methods
Research data were collected from studying documentary and library information and using face-to-face questionnaires with 120 (including 80 high-grading and 40 high-yielding) paddy farmers and farm owners in Rezvanshahr city in Guilan province during the production year 2019-2020. The independent variables were, machinery, diesel and Gasoline fuels, electricity, seed, compost and straw, biocides, fertilizers and labor, and the dependent variable was paddy yield per hectare of the farm area. In the first step, calculations of energy consumption and production were obtained by multiplying the amount of the variables by the relevant and equivalent coefficients.
In the second step, all of the variables in order to maximize paddy yield are entered into MATLAB software. Therefore, the artificial bee colony algorithm with a novel and simple elitism structure was used for the fitness function in the genetic algorithm. The Sphere function, the Repmat function and the Unfrnd function were used as the objective function, defining position of the bee arrays and quantifying the position of the bee arrays, respectively. The number of new responses per each of the generations and algorithm iterations was 900 members and 200 iterations, respectively. Also, in the genetic algorithm, the population type and size were considered double vector and 100.
Results and Discussion
The results showed that total average energy consumption and production in the Hashemi (high-grading) paddy cultivar were obtained 55.973 and 30.742 GJ.ha-1, respectively, and in the Jamshidi (high-yielding) paddy cultivar were 54.796 and 62.522 GJ.ha-1, respectively. In both cultivars, the highest and lowest distributions of energy consumption were related to agricultural machinery and straw, respectively. The average energy consumption of tractor in the Hashemi and the Jamshidi cultivars were 25.111 and 25.865 GJ.ha-1, respectively, which were obtained 44.862 % and 47.202 % of total average energy consumption, respectively. Undoubtedly, it shows that inescapable effect of this input and refers to the operators' skill and experimental knowledge. The results of evaluation indexes such as R2, RMSE, MAPE and EF, and statistical comparison such as mean, STD and distribution indicated that the artificial bee colony algorithm had the essential condition for the fitness function. The optimization results of the bee-genetic algorithm demonstrated that most of the consumed resources are not a little difference from the optimum conditions but can be adopted the proper management in the farm, the Hashemi and the Jamshidi cultivars optimization of energy consumptionwill achieve 53.96 % and 39.41 %, respectively.
According to the results of the desirable performance of the ABC-GA algorithm and identifying the potential of saving energy consumption, policy-makers of the energy resource management and rice industry managers can define new strategies to reduce energy consumption.