Forecasting return and profit is a primary concern for financial practitioners. It is even more critical when it comes to forecasting energy market returns. This research attempts to propose an effective method for predicting the Brent Crude Oil return, which results in remarkable performance compared to the well-known models in the return prediction. The proposed model is a hybrid model based on LSTM and CNN networks where ARIMA and GARCH outputs are used as features along with return lags, price, and macroeconomic variables for training the models, resulting in significant improvement in the model's performance. According to the obtained results, our proposed model performs better than other models, including ANN, PCA-ANN, LSTM, CNN. In the second part of this study, by considering the spread of the COVID-19 and its impact on the financial markets, we present a precise LSTM model that can reflect this disease's impact on the Brent Crude Oil return. This paper uses the significance test and correlation measures to show similarity between the series of Brent Crude Oil during the SARS period and COVID-19; after that, we use SARS period data along the COVID-19 to train the LSTM. The results demonstrate that the proposed LSTM model, tuned by the SARS data, can better predict the Brent Crude Oil return during the COVID19 pandemic.