Energy System Sizing via Intra-Hour Variability Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methodologies for planning Distributed Energy Resources (DER) and microgrids rely on data reduction methods that average time series data, leading to loss of information and inaccurate sizing of energy systems, particularly for volatile energy sources like photovoltaic and wind, due to the elimination of intra-hour variability.
Innovation Solution
A method combining energy balance with original time series data at smaller time steps using a power balancing framework to capture intra-hour variability, allowing for robust and fast techno-economic sizing of energy systems through mixed integer optimization, which includes down-sampling and dispatch optimization to ensure accurate sizing and operational schedules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If data reduction methods average time series data into 60, 30, or 15 min time steps, then computational runtime is reduced, but intra-hour variability and power spikes are eliminated causing loss of information
Solution Approach 1:
The patent segments the time series data processing into two distinct parts: (1) averaging time series data at coarse time steps (60, 30, or 15 min) for computational efficiency, and (2) separately capturing and processing intra-hour variability and power spikes at fine time steps (5 min or 1 min). This segmentation allows each part to be optimized independently, resolving the contradiction between runtime reduction and information preservation.
Solution Approach 2:
The patent introduces a new dimension by adding intra-hour variability capture as a separate processing layer alongside the traditional averaging approach. Instead of choosing between coarse averaging or fine-resolution processing, the method operates in both temporal dimensions simultaneously, using averaged profiles for baseline sizing and intra-hour variability for volatility adjustments.
2Device complexity
If data reduction methods average time series data, then computational complexity is reduced, but sizing accuracy for volatile energy systems deteriorates
Solution Approach 1:
The patent segments the sizing calculation into two components: base sizing based on averaged energy profiles (low complexity) and volatility adjustments based on intra-hour variability (higher precision). This segmentation maintains overall computational simplicity while improving sizing accuracy for volatile energy systems like PV and wind.
Solution Approach 2:
The patent introduces intra-hour variability as an intermediary factor that mediates between simple averaging and complex high-resolution processing. The variability capture acts as a bridge, providing the additional precision needed for accurate sizing of volatile energy systems without requiring full high-resolution time series processing.
3Measurement precision
If original time series data at smaller time steps is used, then power spikes and intra-hour variability are captured, but computational runtime increases significantly
Solution Approach 1:
The patent extracts only the critical intra-hour variability and power spike information from the original high-resolution time series data, rather than processing the entire dataset. This extraction approach captures essential power spike characteristics while discarding redundant data points, significantly reducing computational runtime compared to full high-resolution processing.
Solution Approach 2:
The patent applies partial action by processing only the necessary portion of the time series data - specifically the intra-hour variability component - rather than processing the complete high-resolution dataset. This partial processing achieves sufficient power spike detection accuracy while avoiding the computational burden of full high-resolution analysis.
Data Source
AI summary
The disclosed embodiments combine an energy balance (e.g., averaged energy profiles) with original time series data having smaller time steps to establish an energy balance (e.g., averaged profiles) and a power balance (e.g., smaller time step with original time series data) for a Distributed Energy Resources (DER), microgrid, or other energy system. In an embodiment, a method comprises: solving, with at least one processor, a first optimization problem on time series data related to energy system planning, the first optimization including applying a power balancing framework to the time series data that captures intra-hour variability; and selecting, with the at least one processor, technology assets and sizing for the energy system based on an average hourly and sub-hourly datasets.