Large Scale Solar Power Plant Automated Monitoring and Maintenance

Description

Smart and sustainable development is the key aim of the new generation of technology and advancement. Therefore, green and clean sources of energy are needed to promote and use. In this context, government and private sector organizations are making large-scale investments in developing large-scale solar power plants. In this project, the main aim is to contribute an Internet of Things (IoT) enabled architecture to monitor the productivity of solar power plants. Additionally, by using machine learning algorithms the IoT data is processed to identify the deviation of solar power plant performance. By using this difference the maintenance of solar power panels has been carried out automatically.

The simulation of the proposed automated solar power plant monitoring and maintenance has been carried out using Python technology. Additionally, the dataset from Kaggle has been considered. The dataset can be downloaded from the download link. The dataset contains the invertor-based data. Therefore, first monitoring data is personalized and then invertor-level data analysis is performed. Further, the personalized data based on the inverter is being used to train a Long Short Term Memory (LSTM) architecture and sequential Convolutional Neural Network (CNN). Deep learning techniques are now in these days used in various applications among them prediction is also one of the most essential tasks.

Next, using the next 36-hour monitoring data the model has been validated and performance analysis has been performed. The performance analysis included the analysis of the performance of LSTM and CNN, in terms of accuracy, loss, precision, recall, and f-score. In this project, the main aim is to demonstrate how the machine-learning technique can be used for identifying performance degradation to make decisions for effective maintenance, which keeps the solar power generation performance updated. Finally, the conclusion of the work has been reported and an extension plan has been discussed.