This chapter addresses the critical challenge of accurately modeling shipboard loads across diverse navigation scenarios in all-electric ships (AES). The complex and varying operational conditions—including open waters, restricted waters, dynamic positioning (DP) mode, and radar scanning operations—exhibit significantly different load characteristics that are difficult to capture using traditional modeling approaches. The chapter presents comprehensive multi-scenario shipboard load models that combine physics-based mechanisms with data-driven methodologies. For open water navigation, a hybrid model incorporating coupled trim-heel dynamics and deep neural networks is developed using real vessel data from China Classification Society, achieving 98.7% accuracy through multi-scenario disturbance classification. In restricted waters, shallow and narrow channel effects are modeled using computational fluid dynamics and ship hydrodynamics, considering water depth ratios and bulkhead wall influences. For pulse loads during radar operations, an ultracapacitor-based cascade model with a “rolling charging/discharging" strategy is proposed to smooth power fluctuations and reduce DC bus voltage sags by 84.5%. The dynamic positioning model establishes three-degree-of-freedom motion control with flexible thrust allocation, reducing pulse loads by 28.32% under severe sea conditions while maintaining positioning accuracy. The integrated multi-scenario framework provides essential foundation for power management systems in electrified marine vessels, enabling accurate load forecasting across varying operational conditions and supporting optimal energy storage system deployment for improved grid stability and efficiency.