The electricity consumption for agricultural irrigation is dynamically influenced by multiple factors, such as climatic conditions and crop growth cycles. Traditional estimation methods, including empirical formulas and static statistical models, often fail to adapt to dynamically changing agricultural scenarios, leading to low estimation accuracy and poor adaptability. To address this issue, this paper proposes a method for estimating the electricity consumption of farmland irrigation for mixed agricultural row users based on fuzzy clustering and quantum-behaved particle swarm optimization (QPSO). By employing convolutional neural networks, effective recognition and classification of crop images are achieved. Irrigation period characteristics for different crops are analyzed, and a farmland irrigation load model is constructed. Irrigation zoning is performed using fuzzy clustering algorithms, while regional irrigation behaviors are identified through quantum-behaved particle swarm optimization algorithms. A model for estimating irrigation electricity consumption for mixed agricultural row users is established to predict irrigation electricity usage. Experimental results demonstrate that the overall training accuracy of this method exceeds 0.96, with a clustering accuracy of 89.1%. The estimation error for daily irrigation electricity consumption of a single mixed rural drainage user is only 12.1 kWh, which represents 14.2% of the user’s average daily irrigation electricity consumption (mean 85.3 kWh). The computation time is only 5.7 s. Compared with existing methods, this approach achieves significant improvements in estimation accuracy, clustering reliability, and computational efficiency, providing technical support for agricultural water-saving scheduling and energy optimization. The estimated irrigation electricity consumption exhibits the smallest deviation from actual values and the highest estimation accuracy.
Li et al. (Fri,) studied this question.