Lachlan O'Neil - Academia.edu (original) (raw)
Related Authors
Uploads
Papers by Lachlan O'Neil
Analysis of the VPP dynamic network constraint management
Applied Energy, 2019
• A big dataset of electricity load time series of residential customers is modelled. • Small set... more • A big dataset of electricity load time series of residential customers is modelled. • Small sets of modelled parameters are used to summarise the lengthy load time series. • The customers are clustered by the similarity of their time series' model parameters. • The 12 clusters show 61% distinction and 6.75% improved coincidence factor on average. • The clustering method is reasonably robust against a limited form of mixed noise.
Machine Learning and Knowledge Discovery in Databases. Applied Data Science and Demo Track, 2021
In this paper, the process of forecasting household energy consumption is studied within the fram... more In this paper, the process of forecasting household energy consumption is studied within the framework of the nonparametric Gaussian Process (GP), using multiple short time series data. As we begin to use smart meter data to paint a clearer picture of residential electricity use, it becomes increasingly apparent that we must also construct a detailed picture and understanding of consumer's complex relationship with gas consumption. Both electricity and gas consumption patterns are highly dependent on various factors, and the intricate interplay of these factors is sophisticated. Moreover, since typical gas consumption data is low granularity with very few time points, naive application of conventional timeseries forecasting techniques can lead to severe over-fitting. Given these considerations, we construct a stacked GP method where the predictive posteriors of each GP applied to each task are used in the prior and likelihood of the next level GP. We apply our model to a real-world dataset to forecast energy consumption in Australian households across several states. We compare intuitively appealing results against other commonly used machine learning techniques. Overall, the results indicate that the proposed stacked GP model outperforms other forecasting techniques that we tested, especially when we have a multiple short time-series instances.
Analysis of the VPP dynamic network constraint management
Applied Energy, 2019
• A big dataset of electricity load time series of residential customers is modelled. • Small set... more • A big dataset of electricity load time series of residential customers is modelled. • Small sets of modelled parameters are used to summarise the lengthy load time series. • The customers are clustered by the similarity of their time series' model parameters. • The 12 clusters show 61% distinction and 6.75% improved coincidence factor on average. • The clustering method is reasonably robust against a limited form of mixed noise.
Machine Learning and Knowledge Discovery in Databases. Applied Data Science and Demo Track, 2021
In this paper, the process of forecasting household energy consumption is studied within the fram... more In this paper, the process of forecasting household energy consumption is studied within the framework of the nonparametric Gaussian Process (GP), using multiple short time series data. As we begin to use smart meter data to paint a clearer picture of residential electricity use, it becomes increasingly apparent that we must also construct a detailed picture and understanding of consumer's complex relationship with gas consumption. Both electricity and gas consumption patterns are highly dependent on various factors, and the intricate interplay of these factors is sophisticated. Moreover, since typical gas consumption data is low granularity with very few time points, naive application of conventional timeseries forecasting techniques can lead to severe over-fitting. Given these considerations, we construct a stacked GP method where the predictive posteriors of each GP applied to each task are used in the prior and likelihood of the next level GP. We apply our model to a real-world dataset to forecast energy consumption in Australian households across several states. We compare intuitively appealing results against other commonly used machine learning techniques. Overall, the results indicate that the proposed stacked GP model outperforms other forecasting techniques that we tested, especially when we have a multiple short time-series instances.