Evaluation of Operation Status and Analysis of Transformation Effectiveness of Medium Voltage Distribution Network based on Ensemble model

In present scenario, evaluating transformation effectiveness in medium voltage distribution networks involves assessing the impact of upgrades, repairs, or reconfigurations on network performance, and efficiency by ensuring with optimal power supply and minimizing losses, faults, and downtime. It faces difficulties in measure the true impact of network upgrades or changes. To overcome those drawbacks this paper, proposed Convolutional Neural Network- Long Short-Term Memory (CNN-LSTM) technique efficiently analysis the verification of collecting input data, initially utilized a power grid dataset, which undergoes data preprocessing through three key steps such as cleaning, grouping, and interpolation. Then, feature extraction is performed using time-frequency analysis. The proposed classification model is a hybrid CNN-LSTM approach, leveraging the assets of both CNN and LSTM. This integrated model is trained and tested on the pre-processed data, demonstrating its effectiveness in classifying operation status and transformation effectiveness enables robust extraction of spatial features from images and temporal modeling of sequential and for forecasting of complex phenomena. The performance of proposed model evaluated the various metrics such as precision (0.995%), recall ($\mathbf{0. 9 8 9 \%)}$) and F1 score ($\mathbf{0. 9 9 6 \%}$) metrics on power grid dataset.