Modeling Daily Load Profiles of Distribution Network for Scenario Generation Using Flow-Based Generative Network

Ge, Leijiao; Liao, Wenlong; Wang, Shouxiang; Bak‐Jensen, Birgitte; Pillai, Jayakrishnan Radhakrishna

doi:10.1109/access.2020.2989350

Cited by 52 publications

(35 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper proposed a sociodemographic analysis of the Italian households' load profiles from a smart metering experimental study, while also considering households in energy poverty conditions. The outcomes can be exploited for further research activities, including the validation or the calibration of existing models [14,15,[20][21][22], or they can be used to improve the time step of broader energy consumption estimation models, such as MOIRAE [23], which is still under development. Indeed, simulation models based on the consumption behaviors, as developed by Gao et al [54], require data acquisition either at the calculation procedure phase or at validation phase.…”

Section: Discussionmentioning

confidence: 99%

“…Indeed, state-of-the-art models for residential demand forecasting are based on either survey methods [14,22,37] or high time-resolution data [38] (e.g., provided by SMs). For instance, Ge et al [20] modelled daily load profiles through a generative network, setting a training dataset of real load profiles and mapping such profiles' probability distribution. Similarly, Sousa et al [39] highlighted how high-resolution load profiling data are fundamental for neural-network-based forecasting models, and they proposed a day-ahead load profile prediction exploiting such data; moreover, the segmentation of consumers was carried out through clustering algorithms, although sociodemographic variables were not taken into account.…”

Section: Literature Reviewmentioning

confidence: 99%

“…However, a wider spread and installation of smart meters should take into account technical assessments as well as social issues, as discussed by Sovacool et al [46]. Nonetheless, smart metering has a significant potential in a variety of applications, including energy modelling implementation [20][21][22]36,38,39], households' behavior characterization [12,15,44], grid infrastructure technical assessments [24][25][26][27][28], and an end user's awareness raising [8,32,33].…”

Section: Literature Reviewmentioning

confidence: 99%

“…The defined clusters are thereby representative of the Italian households consumptions and behaviors, with accurate approximations; thus, by conducting measurements within the representative clusters, the results can be generalized and exploited to characterize the possible energy savings of the different types of households at a national level (e.g., waste reduction, appliances replacement) in the perspectives of awareness raising, training, and information campaigns. Moreover, the results regarding the load profiles of both total and single-appliance energy consumptions can be applied to validate load profile simulation models [14,15,[20][21][22] or to decrease the time steps of more general energy consumption estimation models, in particular in exploiting a bottom-up approach such as MOIRAE [23].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Italian Household Load Profiles: A Monitoring Campaign

2020

View full text Add to dashboard Cite

The increasing share of renewable energy sources on the supply side, as well as the so-called electrification pathways on the demand side, has led to peculiar challenges for electrical systems: Indeed, the increasing load demand has to be balanced from the supply-side viewpoint. In particular, the residential sector contributes to nearly 26% of the final energy consumption in Europe, suggesting that a further understanding of households’ consumptions and load profiles is needed to support an energy transition. In this context, this paper contributes to the existing discussion by proposing a sociodemographic analysis of Italian households’ load profiles using a smart metering experimental study, while also considering the households in energy poverty conditions. For the sake of generality, results are presented based on a previously proposed household segmentation of the Italian residential sector. The outcomes point out three prominent peaks on load profiles for all the identified clusters, with a notable distinction in intensity. Where children are present, a higher load profile is noted, reaching a maximum value of 600 W of absorbed power between 19:30 and 22:30. Conversely, households in an energy poverty condition show a relatively regular load profile, ranging from a minimum of 110 W of absorbed power in baseload conditions at night time to a maximum of 280 W in the evening hours. The findings in this paper are in agreement with existing research in the field, and accordingly the study proposes a better focus on domestic appliances and sociodemographic parameters.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Italian Household Load Profiles: A Monitoring Campaign

2020

View full text Add to dashboard Cite

show abstract

“…Zhang et al proposed a flow‐based conditional generative model to provide reliable and sharp scenarios for residential load 32 . To accurately capture the potential behavior of real samples, a generative network based on nonlinear independent component estimation was proposed by Ge et al to model the daily load profiles 33 …”

Section: Introductionmentioning

confidence: 99%

Day‐ahead renewable scenario forecasts based on generative adversarial networks

et al. 2020

View full text Add to dashboard Cite

Summary With the increasing penetration of renewable resources such as wind and solar, especially in terms of large‐scale integration, the operation and planning of power systems are faced with great risks due to the inherent stochasticity of natural resources. Although this uncertainty can be anticipated, the timing, magnitude, and duration of fluctuations cannot be predicted accurately. In addition, the outputs of renewable power sources are correlated in space and time, and this brings further challenges for predicting the characteristics of their future behavior. To address these issues, this paper describes an unsupervised distribution learning method for renewable scenario forecasts that considers spatiotemporal correlation based on generative adversarial network (GAN), which has been shown to generate realistic time series for stochastic processes. We first utilize an improved GAN to learn unknown data distributions and model the dynamic processes of renewable resources. We then generate a large number of forecasted scenarios using stochastic constrained optimization. For validation, we use power generation data from the National Renewable Energy Laboratory wind and solar integration datasets. The simulation results show that the generated trajectories not only reflect the future power generation dynamics, but also correctly capture the temporal, spatial, and fluctuant characteristics of the real power generation processes. The experimental comparisons verify the superiority of the proposed method and indicate that it can reduce at least 50% of the training iterations of the generative model for scenario forecasts.

show abstract