Summary of Day-ahead Regional Solar Power Forecasting with Hierarchical Temporal Convolutional Neural Networks Using Historical Power Generation and Weather Data, by Maneesha Perera et al.
Day-ahead regional solar power forecasting with hierarchical temporal convolutional neural networks using historical power generation and weather data
by Maneesha Perera, Julian De Hoog, Kasun Bandara, Damith Senanayake, Saman Halgamuge
First submitted to arxiv on: 4 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Regional solar power forecasting is crucial for energy sector stakeholders, but it’s challenging due to the vast amount of data from geographically dispersed locations. Previous work has focused on either aggregated or individual time series, disregarding location-specific weather effects. This paper proposes two deep-learning-based regional forecasting methods that leverage both types of time series and weather data. The first approach uses a single hierarchical temporal convolutional neural network (HTCNN) to generate a regional forecast, while the second divides the region into sub-regions based on weather information, training separate HTCNNs for each sub-region. The proposed work is evaluated using a large dataset from 101 locations across Western Australia and compares favorably with alternative methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine trying to predict how much solar power will be generated in a whole region, considering the weather conditions at different locations. This is like solving a big puzzle! Usually, people try to solve this problem by focusing on one part of it or another, but that’s not very accurate. In this paper, they propose two new ways to do regional solar power forecasting using special computer models called deep learning networks. These models can use information from many locations and weather conditions to make a more accurate forecast. The results show that their methods are better than previous ones. |
Keywords
* Artificial intelligence * Deep learning * Neural network * Time series
