Deep ensemble learning-based approach to real-time power system state estimation

Power system state estimation (PSSE) is commonly formulated as weighted least-square (WLS) algorithm and solved using iterative methods such Gauss-Newton methods. However, have become more sensitive to operating conditions than ever before due the deployment of intermittent renewable energy sources, zero emission technologies (e.g., electric vehicles), demand response programs. Appropriate PSSE approaches are required avoid pitfalls WLS-based computations for accurate prediction conditions. This paper proposes a data-driven real-time deep ensemble learning algorithm. In proposed approach, setup with dense residual neural networks base-learners multivariate-linear regressor meta-learner. Historical measurements states utilised train test model. The trained model can be used in estimate power (voltage magnitudes phase angles) measurements. Most current assume availability complete set measurements, which may not case real data-acquisition. adopts multivariate linear regression forecast instants missing assist technique. Case studies performed on various IEEE standard benchmark systems validate approach. results show that approach outperforms existing techniques. developed source code solution publicly available at https://github.com/nbhusal/Power-System-State-Estimation.