Stabilizing Electric Grid Load Forecasts with Distributed PV
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed solar photovoltaic (PV) generation introduces volatility to real-time load measurements, leading to forecast instability and increased costs due to the need for additional spinning reserves in electrical utility systems.
Innovation Solution
A two-stage ensemble smoothing algorithm is implemented to stabilize real-time load forecasts by separating the impact of volatile solar PV generation from underlying power consumption patterns, using a combination of historical and instantaneous cloud cover data to smooth load data and reduce the effect of solar PV generation variability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If distributed solar PV generation is integrated into the electrical grid, then renewable energy utilization is improved, but load forecast stability deteriorates due to volatility from changing cloud conditions
Solution Approach 1:
The patent segments the load measurement into two distinct components: solar-affected load (volatile portion due to cloud changes) and non-solar load (stable baseline consumption). By separating these components through statistical analysis of historical data, the system can forecast the stable non-solar portion independently while accounting for the volatile solar portion separately, thereby improving overall load forecast stability while maintaining renewable energy utilization benefits
Solution Approach 2:
The patent introduces an intermediary statistical model that acts as a mediator between the volatile solar PV generation and the load forecast system. This model characterizes the relationship between solar generation and load measurements, allowing the forecast system to compensate for solar volatility without directly incorporating the unstable solar data, thus maintaining forecast stability while preserving renewable energy integration
2Reliability
If spinning reserve capacity is increased to cover PV generation volatility, then grid reliability is improved, but operational costs increase due to gas-powered turbines running at low gas levels
Solution Approach 1:
The patent performs preliminary separation of solar-affected and non-solar load components before the spinning reserve activation decision is made. By pre-characterizing the volatile solar portion through statistical analysis, the system can distinguish between load changes caused by solar volatility versus actual demand changes, allowing more accurate determination of when spinning reserve is truly needed, thereby reducing unnecessary reserve activation and lowering operational costs while maintaining grid reliability
Solution Approach 2:
The patent implements a feedback mechanism where historical load data and solar generation data are continuously analyzed to refine the statistical model of solar impact. This feedback loop allows the system to learn and adapt to patterns in solar volatility, improving the accuracy of load forecasts over time and enabling more precise spinning reserve management, which reduces unnecessary reserve capacity deployment and associated operational costs
Data Source
AI summary
In the context of an electrical utility system, changing cloud conditions may cause a customer having solar panels to greatly increase or decrease electrical demand in a difficult-to-predict manner. Accordingly, a spinning reserve maintained by an electric utility company must be larger, and is therefore more expensive. In an example, the spinning reserve may be managed by: calculating a stable sequence of forecasts of smoothed real-time consumption, wherein the calculating is based at least in part on smoothed estimates of consumption data. A stable sequence of forecasts of real-time measured load may be calculated by subtracting forecasts of real-time distributed solar photovoltaic (PV) generation data from the stable sequence of forecasts of smoothed real-time consumption. The spinning reserve of the electricity system may be controlled based at least in part on the stable sequence of forecasts of real-time measured load.