Renewable Power Forecasting Using Neighbor Data Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting power generation from renewable energy sources, such as solar and wind, are inadequate for short-term forecasting due to their reliance on expensive hardware and infrequent updates, leading to uncertainty and potential power imbalances.
Innovation Solution
A computer-implemented method that uses current and historical power generation data from neighboring renewable energy installations to create data maps, normalizing and interpolating data to predict future power generation values without the need for additional hardware, leveraging motion vector fields and computer vision techniques for accurate short-term weather forecasting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If numerical weather simulations from meteorological organisations are used for prediction, then predictions over longer timescales of several hours are improved, but predictions for shorter-term weather variations deteriorate
Solution Approach 1:
The patent combines data from multiple neighbouring renewable energy installations to create a comprehensive data map that captures short-term weather variations. By merging observations from multiple sources in the same geographical area, the system achieves improved short-term prediction accuracy without relying on expensive dedicated sensors at each location.
Solution Approach 2:
The patent uses power generation data from neighbouring installations as an intermediary indicator of local weather conditions. Instead of directly measuring weather parameters with expensive sensors, the system uses power generation variations as a proxy to infer weather changes and predict their impact on the target installation.
2Loss of information
If satellite imagery is used to track weather features, then weather tracking capability is improved, but prediction accuracy for shorter-term variations deteriorates due to infrequent updates
Solution Approach 1:
The patent achieves continuous monitoring of weather conditions by utilizing power generation data streams from multiple neighbouring installations in real-time. This continuous data flow replaces the intermittent satellite imagery updates, enabling the system to capture and respond to short-term weather variations as they occur.
3Measurement precision
If sensors are installed on-site at a renewable energy system to monitor weather conditions, then short-term weather prediction accuracy is improved, but cost and device complexity worsen due to installation and maintenance of dedicated hardware
Solution Approach 1:
The patent makes existing renewable energy installations serve a dual function: generating power and providing weather monitoring data. By utilizing the power generation measurements already being taken for energy production purposes, the system avoids the need for separate dedicated weather sensors, thereby reducing hardware complexity and costs while maintaining prediction accuracy.
Solution Approach 2:
The system uses the existing infrastructure and data collection capabilities of neighbouring renewable energy installations to serve the weather monitoring function. Each installation's power generation data automatically contributes to the collective weather prediction system without requiring additional equipment or dedicated maintenance resources at any single location.
4Measurement precision
If sensors are installed on-site to monitor weather conditions, then short-term weather prediction accuracy is improved, but cost worsens due to significant computational requirements
Solution Approach 1:
The patent divides the monitoring function across multiple neighbouring installations, with each installation contributing its own power generation data. This segmentation distributes the measurement burden and allows the system to build a comprehensive weather picture from distributed, low-cost data sources, reducing the computational burden compared to a single centralized sensor system.
Data Source
AI summary
The invention provides a computer-implemented method for determining predicted power generation of a renewable energy installation. The method includes receiving current power generation data indicative of a current power generation value for each of one or more neighbouring renewable energy installations, and determining, based on the received current power generation data, a current data map indicative of current power generation values across an area including the renewable energy installation and the one or more neighbouring renewable energy installations. The method includes retrieving a previous data map indicative of power generation values across the area at a previous time, and determining, based on the previous and current data maps, a future data map indicative of power generation values across the area at a future time, determining a predicted power generation value of the renewable energy installation at the future time based on the determined future data map.