Search

• A framework identifies energy contexts where racial injustice is more likely. • The 2019 American Housing Survey is used to assess five energy outcomes. • Housing-wealth explains 37% of racial inequality in solar uptake by US homeowners. • Racial differences exist for having washing machines, dryers, or broken equipment. • Black-alone households in public housing may have lower access to solar panels. This paper presents a framework to identify which household-energy outcomes are more likely to display injustice. Inequity is more likely when variation is pronounced at the lower end of wealth distributions. This paper compares solar-panel uptake with four other energy outcomes using the 2019 American Housing Survey. Analysis shows that solar-panel uptake in the United States has mainly been by high-wealth homeowners, across racial groups. Housing wealth explains around 37% of the solar-uptake difference between "Black alone" and other homeowners. In contrast, there are large racial differences for having washing machines or dryers. In addition, uncomfortably cold home temperatures and broken equipment are more likely for "Black alone" homeowners. Similar patterns are mostly observed for renters. However, "Black alone" households in public housing may have been less likely to benefit from solar panels than other public-housing tenants. There are three main implications. Solar policies targeting low wealth, rather than income, have greater potential for reducing racial inequality. Also, policymakers could focus more on equality in solar-panel access for public housing. Further, there are larger racial differences to address in non-solar contexts. [ABSTRACT FROM AUTHOR]

[Display omitted] • A new convolutional recurrent neural network is used for spatiotemporal prediction. • A new probabilistic deep learning model is proposed for solar radiation prediction. • A data preprocessing method for solar radiation uncertainty assessment is proposed. • The developed model performs very well in accuracy and uncertainty. High-quality and reliable solar radiation spatiotemporal probabilistic prediction is very significant for the planning and management of solar energy. In this paper, convolutional shared weight long short-term memory network is proposed to deal with solar radiation spatiotemporal prediction problems. The model can minimize the training time of model without significantly reducing prediction accuracy. Furthermore, a spatiotemporal probabilistic prediction model that combines convolutional shared weight long short-term memory network and deep ensemble method is proposed to deal with the solar radiation probabilistic prediction problem. The model obtains the uncertainty estimation of predictands by adjusting the network structure and optimizes the uncertainty by employing a proper scoring rule. In addition, a combination of ensembles are used in the model to improve the robustness of probabilistic prediction. The new spatiotemporal probabilistic prediction model is applied to predict solar radiation in a real area of the United States. Five state-of-the-art models and seven evaluation indicators are used for comparation. The comparation results show the proposed model is able to provide accurate point prediction, reasonable prediction interval and reliable probabilistic prediction results for a whole area. [ABSTRACT FROM AUTHOR]