Representation Learning in Power Time Series Forecasting

Henze, Janosch; Schreiber, Jens; Sick, Bernhard

doi:10.1007/978-3-030-31760-7_3

Cited by 9 publications

(8 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to research areas like computer vision, there is limited research on ITL for renewable power forecasts [6,7]. There has been some work to learn a transferable representation of the input utilizing autoencoders [8,9,10,11]. While the principle approach of transferring an autoencoder for a target is combinable with our approach, we argue that considering the conditional distribution of the power forecast is more relevant for model selection and combination.…”

Section: Related Workmentioning

confidence: 96%

Model Selection, Adaptation, and Combination for Transfer Learning in Wind and Photovoltaic Power Forecasts

Schreiber¹,

Sick²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

There is recent interest in using model hubs, a collection of pre-trained models, in computer vision tasks. To utilize the model hub, we first select a source model and then adapt the model for the target to compensate for differences. While there is yet limited research on a model selection and adaption for computer vision tasks, this holds even more for the field of renewable power. At the same time, it is a crucial challenge to provide forecasts for the increasing demand for power forecasts based on weather features from a numerical weather prediction. We close these gaps by conducting the first thorough experiment for model selection and adaptation for transfer learning in renewable power forecast, adopting recent results from the field of computer vision on six datasets. We adopt models based on data from different seasons and limit the amount of training data. As an extension of the current state of the art, we utilize a Bayesian linear regression for forecasting the response based on features extracted from a neural network. This approach outperforms the baseline with only seven days of training data. We further show how combining multiple models through ensembles can significantly improve the model selection and adaptation approach. In fact, with more than 30 days of training data, both proposed model combination techniques achieve similar results to those models trained with a full year of training data.

show abstract

Section: Related Workmentioning

confidence: 96%

Model Selection, Adaptation, and Combination for Transfer Learning in Wind and Photovoltaic Power Forecasts

Schreiber¹,

Sick²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We ensure that the network is not learning an identity mapping during training by having a lower dimension in the bottleneck than in the input. Other alternatives to avoid this problem are, e.g., denoising autoencoders, where we induce random noise on the input features [1]. However, we excluded those variants as the current results suggest that they are not beneficial over vanilla autoencoders for day-ahead power forecasting [1].…”

Section: Inputs Outputsmentioning

confidence: 97%

“…The article closest to ours is [1]. This article compared traditional feature extraction methods with feature extraction techniques from deep learning.…”

Section: Related Workmentioning

confidence: 99%

“…Due to the increase in the amount of renewable energy in the electric grid, it is essential to find suitable compact representations that allow prediction of expected power generation. Such compact representations often improve the prediction quality and can be used to save computational resources and time [1]. The area of research that deals with finding a suitable representation, e.g., through autoencoders, is called representation learning [2].…”

Section: Introductionmentioning

confidence: 99%

“…The generative approach learns to represent the (whole) distribution, while a discriminative autoencoder learns the most efficient data encoding to represent the data. On the other hand, discriminative autoencoders have the advantage that they are easy to implement and often are sufficient for various tasks [1].…”

mentioning

confidence: 99%

See 2 more Smart Citations

Multi-Task Autoencoders and Transfer Learning for Day-Ahead Wind and Photovoltaic Power Forecasts

Schreiber

Sick

2022

Energies

Self Cite

View full text Add to dashboard Cite

Integrating new renewable energy resources requires robust and reliable forecasts to ensure a stable electrical grid and avoid blackouts. Sophisticated representation learning techniques, such as autoencoders, play an essential role, as they allow for the extraction of latent features to forecast the expected generated wind and photovoltaic power for the next seconds up to days. Thereby, autoencoders reduce the required training time and the time spent in manual feature engineering and often improve the forecast error. However, most current renewable energy forecasting research on autoencoders focuses on smaller forecast horizons for the following seconds and hours based on meteorological measurements. At the same time, larger forecast horizons, such as day-ahead power forecasts based on numerical weather predictions, are crucial for planning loads and demands within the electrical grid to prevent power failures. There is little evidence on the ability of autoencoders and their respective forecasting models to improve through multi-task learning and time series autoencoders for day-ahead power forecasts. We can close these gaps by proposing a multi-task learning autoencoder based on the recently introduced temporal convolution network. This approach reduces the number of trainable parameters by 38 for photovoltaic data and 202 for wind data while having the best reconstruction error compared to nine other representation learning techniques. At the same time, this model decreases the day-ahead forecast error up to 18.3% for photovoltaic parks and 1.5% for wind parks. We round off these results by analyzing the influences of the latent size and the number of layers to fine-tune the encoder for wind and photovoltaic power forecasts.

show abstract

Visualizing the Behavior of Convolutional Neural Networks for Time Series Forecasting

Henze

Sick

2021

Studies in Computational Intelligence

View full text Add to dashboard Cite

Representation Learning in Power Time Series Forecasting

Cited by 9 publications

References 13 publications

Model Selection, Adaptation, and Combination for Transfer Learning in Wind and Photovoltaic Power Forecasts

Model Selection, Adaptation, and Combination for Transfer Learning in Wind and Photovoltaic Power Forecasts

Multi-Task Autoencoders and Transfer Learning for Day-Ahead Wind and Photovoltaic Power Forecasts

Visualizing the Behavior of Convolutional Neural Networks for Time Series Forecasting

Contact Info

Product

Resources

About