Energy in Russia

The data is to evaluate the impact of restrictive measures introduced in connection with COVID-19 on consumption and, accordingly, on electricity generation in Russian cities, which were most susceptible to outbreaks of the coronavirus infection (Moscow, St. Petersburg, Yekaterinburg and Nizhny Novgorod). Currently, based on available data, the mobility of the population has decreased resulting in lower demand for electricity during self-isolation. Therefore, the study will be based on the hypothesis that similar changes in human behavior can be expected in the future with further spread of COVID-19 and/or the introduction of additional quarantine measures in major cities around the world. The analysis also yielded additional results: the strongest reductions in energy generation occurred in cities with high building density (7% in Moscow, 14% in Yekaterinburg). Furthermore, the decrease in energy generation in cities with low building density was not so dramatic (1% in St. Petersburg, 0% - Nizhny Novgorod). The study uses two models created with Keras LSTM. The first model forecasts power generation and uses 76 parameters. The second LSTM model forecasts new COVID-19 cases across countries, in which 10 parameters are involved.本数据集旨在评估新冠疫情（COVID-19）相关限制性措施对俄罗斯新冠疫情高暴发风险城市（莫斯科、圣彼得堡、叶卡捷琳堡及下诺夫哥罗德）的电力消费及相应发电量的影响。当前，基于现有数据，居家自我隔离期间民众的人口流动性下降，导致电力需求降低。据此，本研究提出假设：随着新冠疫情进一步蔓延，或全球主要城市出台新增防疫隔离措施时，人类行为将出现类似变化。本次分析还得到了额外研究结果：建筑密度较高的城市发电量降幅最为显著（莫斯科降幅7%，叶卡捷琳堡降幅14%）；而建筑密度较低的城市发电量降幅则相对平缓（圣彼得堡降幅1%，下诺夫哥罗德降幅0%）。本研究采用两个基于Keras长短期记忆网络（LSTM）构建的模型：第一个模型用于预测发电量，共使用76个参数；第二个LSTM模型用于预测全球各国的新增新冠确诊病例，涉及10个参数。