Energy Storage Device Sizing via Probabilistic Load Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for sizing energy storage devices (ESDs) at residential levels are inadequate, as they primarily rely on worst-case or average load profiles, failing to account for behavioral changes and intermittent renewable energy sources, leading to inefficiencies and increased costs.
Innovation Solution
A method and system that generate load profiles, ensemble net load profiles, and use cumulative distribution functions to determine optimal ESD capacities, incorporating user input and behavioral data, and graphical representations to size ESDs effectively for residential households, communities, and feeders.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If worst-case or average load profiles are used for sizing ESDs, then the sizing process is simple, but the accuracy and reliability of ESD capacity determination deteriorates
Solution Approach 1:
The patent transitions from static worst-case or average load profiles to dynamic, probabilistic load profiles that capture the full range of operational conditions. By using cumulative distribution functions and ensemble methods, the system dynamically models load variations based on behavioral changes and renewable energy intermittency, improving accuracy without excessive complexity
Solution Approach 2:
The patent changes the parameter representation from single-point estimates (average/worst-case) to probability distribution functions. This allows the sizing process to incorporate uncertainty and variability in load and generation, providing more reliable ESD capacity determination while maintaining computational feasibility through standardized statistical methods
2Ease of manufacture
If traditional load profiling methods are used, then data processing is straightforward, but the ability to account for behavioral changes and intermittent renewable energy deteriorates
Solution Approach 1:
The patent segments the load profile into multiple scenarios or ensembles that represent different operational conditions, including behavioral changes and renewable energy variations. This segmentation allows the system to process each scenario separately while capturing the full range of uncertainties, balancing simplicity and adaptability
Solution Approach 2:
The patent incorporates feedback mechanisms that use historical load data, behavioral patterns, and renewable generation data to continuously refine the probabilistic load profiles. This feedback loop enables the system to adapt to changing conditions while maintaining a structured data processing approach
3Reliability
If larger ESD capacities are selected to ensure reliability under all conditions, then system reliability improves, but cost and system size increase
Solution Approach 1:
The patent applies partial action by determining ESD capacity based on probabilistic criteria rather than requiring coverage of all possible conditions. By using cumulative distribution functions, the system identifies the capacity needed to meet reliability targets for the majority of operational scenarios, avoiding the excessive capacity required by worst-case approaches while maintaining adequate reliability
Data Source
AI summary
A method and system for sizing energy storage device capacity. The method includes generating a load profile of at least an entity based on at least information acquired by a user interface, generating an ensemble of net load profiles based on at least the load profile and based on at least one energy source profile, determining values of cumulative distribution functions associated with a plurality of energy storage device capacities as a function of at least the ensemble of net load profiles, generating a graphical representation of a combination of the composite cumulative distribution functions and a cluster of equal probability line, rendering the graphical representation, and sizing a capacity of the energy storage device based on the rendered graphical representation.