Present study links physical habitat simulation with the irrigation loss function to assess optimal environmental flow regime. The main motivation is Lack of optimization framework in the structure of reliable environmental flow methods such as physical habitat simulation. Data-driven model was used to simulate physical habitats. Moreover, a multi objective particle swarm optimization was utilized to optimize environmental flow regime by considering two objective functions including weighted useable area function and normalized benefit function for irrigation demand. Based on results, physical habitat data-driven model was robust to assess habitat suitability. Moreover, proposed multi objective optimization model is able to assess environmental flow properly. In the case study, physical habitat impact was minimized to 30% that means 70% of useable habitats would be protected. In contrast, 60% of maximum requested water demand would be supplied which implies a fair balance between demand and environmental flow.

本研究将物理栖息地模拟与灌溉损失函数联系起来，以评估最佳环境流量状况。主要动机是缺乏可靠的环境流方法（如物理栖息地模拟）的结构中的优化框架。数据驱动模型用于模拟物理栖息地。此外，通过考虑两个目标函数，包括加权可用面积函数和灌溉需求的归一化收益函数，利用多目标粒子群算法优化环境流态。基于结果，物理栖息地数据驱动模型对于评估栖息地的适宜性具有鲁棒性。此外，提出的多目标优化模型能够正确评估环境流量。在案例研究中 物理栖息地的影响最小化到30％，这意味着70％的可用栖息地将得到保护。相反，将提供最大需求水需求的60％，这意味着需求与环境流量之间的平衡。