Research on crop cultivation optimization strategy based on adaptive hybrid differential evolution algorithm

This study presents a comprehensive approach to optimizing crop cultivation strategies using an Adaptive Hybrid Differential Evolution (AHDGA) algorithm. The research focuses on preprocessing agricultural data to ensure accuracy, employing a dual strategy of Z-score and boxplot analysis to identify and exclude outliers. A crop cultivation optimization model is established, incorporating decision variables for planting areas, economic benefits, and sustainability. The model accounts for sales revenue, production costs, and yield constraints, considering two scenarios: overproduction waste and discounted sales. The AHDGA is applied to solve the resulting Mixed Integer Nonlinear Programming (MINLP) problem, demonstrating improved efficiency and solution quality over traditional methods. The algorithm's performance is evaluated through comparison with Genetic Algorithm (GA) and Differential Evolution (DE), showcasing its superiority in convergence speed and stability. The study concludes that the AHDGA-based optimization model effectively balances economic efficiency, resource utilization, and risk management in crop cultivation planning. This research highlights the interdisciplinary application of advanced optimization algorithms and data analysis techniques in the field of agriculture, contributing to the development of smart agriculture.