A real-time fuzzy decision support system for alfalfa irrigation

Abstract

An irrigation decision support system is a critical component in precision water management. In this study, a real-time irrigation decision support system (IDSS) based on a fuzzy inference system and a relevant software were developed for irrigation management in alfalfa. The proposed IDSS was a comprehensive model, with inputs of soil water, alfalfa growth and weather conditions. On the basis of soil moisture and difference of alfalfa height, the IDSS determined the most appropriate irrigation strategy in order to maintain the soil moisture above a predefined value. In the fuzzy inference system, all variables were fuzzified using triangular membership functions. In the process of fuzzification, the max-min inference and the rule-based Mamdani-type fuzzy modeling were adopted to generate the amount of irrigation. The IDSS was incorporated with weather forecasting and the models of alfalfa growth and soil water. It provided the predictions of alfalfa growth stages, height, and soil water status. The test results indicated that the prediction models made a good estimation of alfalfa growth stages and soil water with NRMSE of 8.28% and 6.29%, respectively. Eventually, the IDSS provided the timing and amounts of irrigation for high efficiency of the alfalfa irrigated system. The test in Zhuozhou, Hebei Province, China indicated that this IDSS is very promising in facilitating alfalfa irrigation and harvest management.