Optimization of kiwifruit irrigation strategies using multi-objective optimization algorithms coupled with water production functions

•A modelling and optimization framework was developed to optimize kiwifruit irrigation strategies.•Crop responses were accurately simulated using Jensen model for yield and Q-Rao model for fruit quality.•MOPSO produced optimal water allocation plans with low computational cost.•Optimized evapotransp...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computers and electronics in agriculture Jg. 237; S. 110579
Hauptverfasser: Zheng, Shunsheng, Cui, Ningbo, Gong, Daozhi, Wang, Zhihui, Chen, Fei, Liu, Quanshan, Jiang, Shouzheng
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.10.2025
Schlagworte:
Deficit irrigation
Jensen model
Multi-objective optimization
Precision irrigation
Water allocation strategy
Water deficit sensitivity index
Water deficit sensitivity index
Multi-objective optimization
Deficit irrigation
Precision irrigation
Jensen model
Water allocation strategy
ISSN:0168-1699
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•A modelling and optimization framework was developed to optimize kiwifruit irrigation strategies.•Crop responses were accurately simulated using Jensen model for yield and Q-Rao model for fruit quality.•MOPSO produced optimal water allocation plans with low computational cost.•Optimized evapotranspiration ratios were 0.53, 1.00, 1.00, 0.90 under sufficient water supply.•The optimized strategy reduced yield by 7.2 % but improved WP and quality by 2.2 % and 9.1 %. Efficient irrigation strategies are crucial for improving crop yield, fruit quality, and water productivity (WP), particularly under water scarcity conditions. This study developed dated crop water production functions (DCWPF) to simulate kiwifruit yield and quality responses under deficit irrigation, and integrated them with multi-objective optimization (MOO) algorithms to identify optimal irrigation strategies under varying total available water (TAW) conditions. The Jensen model exhibited robust performance in simulating yield, while the Q-Rao model effectively captured the nonlinear response of fruit quality to water stress. Water deficit sensitivity indexes revealed that fruit expansion stage (III) was the most critical for yield, while moderate deficit during fruit maturation stage (IV) could improve quality traits. The multi-objective particle swarm optimization (MOPSO) exhibited superior performance in both computational efficiency and solution quality, highlighting its suitability for optimizing kiwifruit irrigation strategies. Under adequate TAW condition, the optimal relative evapotranspiration allocations across the four growth stages of kiwifruit were 0.53, 1.00, 1.00, and 0.90. At this strategy, a 7.2 % reduction in yield was traded off for a 9.1 % improvement in fruit quality and a 2.2 % enhancement in WP. Under limited TAW, the recommended strategies prioritized irrigation during stages III and II (flowering to fruit set stage). The findings not only provide theoretical support for irrigation water management in kiwifruit cultivation, but also demonstrate the effectiveness of coupling DCWPF with MOO algorithms for optimizing irrigation strategies, offering valuable insights into the application of multi-objective optimization in agriculture practices.
ISSN:0168-1699
DOI:10.1016/j.compag.2025.110579