Search

• A transfer learning framework and a CNN model are proposed for the black swan small-sample events. • (BEST-L). • Experiments show that the BEST-L outperforms traditional methods. • The transfer learning method can effectively use small samples to improve the model's generalization. • performance. • The BEST-L is tested for the load forecasting case with the COVID-19 scenario in Central China. The black swan event will usually cause a great impact on the normal operation of society. The scarcity of such events leads to a lack of relevant data and challenges in dealing with related problems. Different situations also make the traditional methods invalid. In this paper, a transfer learning framework and a convolutional neuron network are proposed to deal with the black swan small-sample events (BEST-L). Taking the COVID-19 as a typical black swan event, the BEST-L is utilized to achieve accurate mid-term load forecasting using the relationship between economy and electricity consumption. The experiment results show that the transfer learning model can effectively learn the basic knowledge about the relationship between the adopted input and output data and use a relatively small amount of data during the black swan event to improve the target areas' generalization. The approach and results can provide an effective approach to respond and react to sudden changes quickly and effectively in similar open problems. [ABSTRACT FROM AUTHOR]

[Display omitted] • Clean energy transitions based on the circular economy (CE) are analysed for a town in China. • This could reduce energy use, CO 2 emissions and PM 2.5 emissions compared to the BAU scenario. • COVID pandemic is a new opportunity to introduce circular economy measures. • Circular economy measures could accelerate clean energy transitions. This paper models the energy and emissions scenarios for a circular economy based clean energy transitions in a 140,000-population town in China, taking into account the new situation encountered by the COVID-19 pandemic. The modelled scenarios propose new clean energy transition roadmaps towards a sustainable urban system through the implementation of circular economy strategies. This is represented by the cascading use of industrial excess heat to form symbiosis between factories and to cover the growing building heat demand, as well as by the electrification of the transport sector and reusing the batteries for a second life as energy storage devices. The results show that for a circular economy scenario, during 2020–2040, an accumulated saving of 7.1 Mtoe final energy use (34%), a decline in 14.5 Mt CO 2 emissions (40%) and 592 t PM 2.5 emissions (43%) could be achieved compared with the business-as-usual scenario. The outcomes of the circular economy strategies are at least 7% better than the new policy scenario which simply has energy efficiency improvements. The outbreak of the COVID-19 tremendously impacts the socio-economic activities in the town. If taking the pandemic as an opportunity to enhance the circular economy, by 2040, compared with the scenario without introducing circular economy measures, the extra avoided final energy use, CO 2 emissions and PM 2.5 emissions could be 1.6 Mtoe (8%), 3.8 Mt (11%) and 229 t (17%) respectively. [ABSTRACT FROM AUTHOR]

With the coronavirus pandemic wreathing havoc around the world, power industry has been hit hard due to the proposal of lockdown policies. However, the impact of lockdowns and shutdowns on the power system in different regions as well as periods of the pandemic can hardly be reflected on the foundation of current studies. In this paper, a prediction-based analysis method is developed to point out the electricity consumption gap resulted from the pandemic situation. The core of this method is a novel optimized grey prediction model, namely Rolling IMSGM(1,1) (Rolling Mechanism combined with grey model with initial condition as Maclaurin series), which achieves better prediction results in the face of long-term emergencies. A novel initial condition is adopted to track data with various characteristics in the form of higher-order polynomials, which are then determined by intelligent algorithms to realize accurate fitting. Historical power consumption data in China are utilized to carry out the monthly forecasts during COVID-19. Compared with other competitive models' prediction results, the superiority of IMSGM(1,1) are demonstrated. Through analyzing the gap between predicted consumption values and the actual data, it can be found that the impact of the pandemic on electricity varies in different periods, which is related to its severity and the local lockdown policies. This study helps to understand the impact on power industry in the face of such an emergency intuitively so as to respond to possible future events. [ABSTRACT FROM AUTHOR]