Recently computer science has evolved to a level wherein it is being tried to develop an artificial human brain with the help of BCI (Brain computer interface) which can help the society in various ways like help the patients to remove hearing problems, develop artificial hands for disabled patients, help in curing depression using EEG signals and much more. In fact, using machine learning a lot of unconventional sources of energy like wind energy, solar energy etc have benefitted immensely throughout the years. This is leading to a reduction in cost, improvement in forecasts and powers the return rate of their portfolio. And the similar positive trend is expected to go up in coming times. This study uses deep learning to undertake a systematic analysis of EEG categorization, resulting in an EEG brainwave dataset with five people' mental states and just a minute session for each mental state category to train and examine multiple approaches. Relaxed, Concentrating, and Neutral are the three main categories of states. We classified the three possibilities based on a few mental states identified via cognitive behavioural studies utilising the Muse headband with four Electroencephalography sensors namely, TP9, AF7, AF8, TP10. Only 44 traits out of a total of approximately 2100 are required, according to the results. And applying ML algorithms in Wind Energy to get the weather reports, they were able to anticipate output 36 hours in advance and this helped them enhance their energy's value by 20%. It has been observed that in the past classifiers like Support Vector Machines (SVM) and Random Forests (RF) attained a total accuracy of around 87% only. From the exploratory analysis of deep learning methods like BCI, confusion matrix, classification report it has been perceived that the EEG brainwave dataset model capitulates a test accuracy rate of 95.027% while classifying responses rendered by the brainthroughEEG signals.