Reduced Network Model Screening for Power Distribution Analysis

The patent segments the analysis process into two distinct phases: (1) a reduced-model analysis phase that screens multiple scenarios to identify critical ones, and (2) a full-model analysis phase that provides detailed accurate analysis only for the selected critical scenarios. This segmentation resolves the contradiction by applying different levels of model fidelity to different subsets of scenarios, thereby reducing overall computational time while maintaining accuracy for the most important cases.

The patent employs a reduced network model that is computationally inexpensive to execute, using it as a disposable screening tool to eliminate irrelevant scenarios. This cheap model sacrifices some accuracy but enables rapid evaluation of many scenarios, after which only the most critical ones proceed to the expensive full-model analysis, thus resolving the time-accuracy tradeoff.

The patent segments the analysis system into two distinct computational components: a reduced-model analysis engine for rapid scenario screening and a full-model analysis engine for detailed accurate analysis. This segmentation resolves the complexity contradiction by organizing the computational workload into manageable stages, where the simple reduced model handles the bulk of scenario evaluation and the complex full model is invoked only when necessary.

The reduced network model serves as an intermediary between the scenario definition and the full-model analysis. It acts as a filtering mechanism that translates the large set of input scenarios into a smaller set of critical scenarios suitable for full-model analysis, thereby reducing the complexity burden on the decision-making system while preserving accuracy for important cases.

The patent segments the scenario analysis into two stages: first, a reduced-model stage that rapidly evaluates all scenarios to identify critical ones, and second, a full-model stage that provides detailed accurate analysis only for the critical subset. This segmentation enables comprehensive scenario coverage while controlling computational time by applying different analysis depths to different scenarios.

The patent applies partial action by performing complete accurate analysis (full-model) only on the critical scenarios identified by the reduced model, rather than performing full analysis on all scenarios. This partial application of the expensive analysis method achieves reliable assessment of the most important cases while avoiding excessive computational time that would result from analyzing all scenarios with equal depth.