Evaluating the Impact of Abiotic Factors on Wheat Crop Production using Back Propagation Fuzzy Neural Network

Abstract

Abiotic components or abiotic factors are non-living chemical and physical parts of the environment that affect living organisms and the functioning of ecosystems. Abiotic factors and the phenomena associated with them underpin biology as a whole. One of the abiotic factors is the soil. Macronutrients are elements which plants require in relatively large amounts whereas micronutrients are those which plants require in much smaller amounts. A combination of macronutrients and micronutrients give the soil its optimum health. A lack of any one of these nutrients can significantly impact the health and longevity of a plant. Natural deficiencies in soil can be supplemented by using the proper applications and techniques to ensure maximum growth and yield of the plants. When plants get good nutrition, they grow well and give farmers the best possible yield in return. In this paper Back Propagation Neural Network using trapezoidal fuzzy number is applied to optimise the exact requirement of nutrient in the soil to maximise the yield of the plant. The wheat seeds sown in four different wheat fields are taken as the input. The weights denote the quantity of macronutrients viz., nitrogen, phosphorus, potassium and sulfur applied to each field. The inputs and weights are feed forwarded to find the output of the hidden layer which gives the yield on application of only the macronutrients. The micronutrients viz., boron, chlorine, iron and manganese are supplied to each field and the total yield of all the four fields is also recorded. If the yield obtained is not the expected output then by back propagation, the error is identified and the optimum quantity of macronutrients and micronutrients to be applied are calculated till the maximum yield is obtained. The application of nutrients to the wheat fields in each trial is plotted graphically.