Hidden Markov Models revealing the household thermal profiling from smart meter data

Ulmeanu, Anatoli Paul; Barbu, Vlad Stefan; Tanasiev, Vladimir; Badea, Adrian

doi:10.1016/j.enbuild.2017.08.036

Cited by 13 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first step to setup a prediction model is to determine the number of lag periods, which can also be understood as input to the time-series model. AIC and BIC have been widely used in literature [42,43]; these two criteria can find the best balance between the complexity of the model and the accuracy of the model. Therefore, we select AIC and BIC to be the selection criteria for the number of lag periods, and they are implemented using MATLAB.…”

Section: Model Trainingmentioning

confidence: 99%

A Novel Interval Energy-Forecasting Method for Sustainable Building Management Based on Deep Learning

Duan

2022

Sustainability

View full text Add to dashboard Cite

Energy conservation in buildings has increasingly become a hot issue for the Chinese government. Compared to deterministic load prediction, probabilistic load forecasting is more suitable for long-term planning and management of building energy consumption. In this study, we propose a probabilistic load-forecasting method for daily and weekly indoor load. The methodology is based on the long short-term memory (LSTM) model and penalized quantile regression (PQR). A comprehensive analysis for a time period of a year is conducted using the proposed method, and back propagation neural networks (BPNN), support vector machine (SVM), and random forest are applied as reference models. Point prediction as well as interval prediction are adopted to roundly test the prediction performance of the proposed model. Results show that LSTM-PQR has superior performance over the other three models and has improvements ranging from 6.4% to 20.9% for PICP compared with other models. This work indicates that the proposed method fits well with probabilistic load forecasting, which could promise to guide the management of building sustainability in a future carbon neutral scenario.

show abstract

Section: Model Trainingmentioning

confidence: 99%

A Novel Interval Energy-Forecasting Method for Sustainable Building Management Based on Deep Learning

Duan

2022

Sustainability

View full text Add to dashboard Cite

show abstract

“…The performance was validated through a public data set and accurate disaggregation results could be obtained. Ulmeanu et al adopted HMMs to analyze hourly-recorded time series of building thermal loads [71]. The research results showed that various temporal patterns, each corresponding to different seasonal effects and occupancy behaviors, could be successful identified.…”

Section: Miller Et Al Developed An Automated Filter To Find Infrequen...mentioning

confidence: 99%

“…The temporal associations discovered were found to be useful for identifying time lags in status transitions and anomalies in temporal space. Not specified [70] Building power usage Motif discovery Hidden Markov Not Specified [71] disaggregation Models…”

Section: Miller Et Al Developed An Automated Filter To Find Infrequen...mentioning

confidence: 99%

Unsupervised data analytics in mining big building operational data for energy efficiency enhancement: A review

Fan

Xiao

et al. 2018

Energy and Buildings

172

View full text Add to dashboard Cite

“…A common and quite successful approach is the factorial hidden Markov model (FHMM) [23,24]. The main drawback of most unsupervised methods is their requirement of a relatively high sampling rate (typically greater than 1/60 Hz), although recently the method has also been applied also to low-frequency data [25,26].…”

Section: Introductionmentioning

confidence: 99%

A Bayesian Approach to Unsupervised, Non-Intrusive Load Disaggregation

Massidda

Marrocu

2022

Sensors

View full text Add to dashboard Cite

Estimating household energy use patterns and user consumption habits is a fundamental requirement for management and control techniques of demand response programs, leading to a growing interest in non-intrusive load disaggregation methods. In this work we propose a new methodology for disaggregating the electrical load of a household from low-frequency electrical consumption measurements obtained from a smart meter and contextual environmental information. The method proposed allows, with an unsupervised and non-intrusive approach, to separate loads into two components related to environmental conditions and occupants’ habits. We use a Bayesian approach, in which disaggregation is achieved by exploiting actual electrical load information to update the a priori estimate of user consumption habits, to obtain a probabilistic forecast with hourly resolution of the two components. We obtain a remarkably good accuracy for a benchmark dataset, higher than that obtained with other unsupervised methods and comparable to the results of supervised algorithms based on deep learning. The proposed procedure is of great application interest in that, from the knowledge of the time series of electricity consumption alone, it enables the identification of households from which it is possible to extract flexibility in energy demand and to realize the prediction of the respective load components.

show abstract

Hidden Markov Models revealing the household thermal profiling from smart meter data

Cited by 13 publications

References 24 publications

A Novel Interval Energy-Forecasting Method for Sustainable Building Management Based on Deep Learning

A Novel Interval Energy-Forecasting Method for Sustainable Building Management Based on Deep Learning

Unsupervised data analytics in mining big building operational data for energy efficiency enhancement: A review

A Bayesian Approach to Unsupervised, Non-Intrusive Load Disaggregation

Contact Info

Product

Resources

About