Branch current state estimation of three phase distribution networks suitable for paralellization

The evolution of distribution networks from passive to active distribution systems puts new requirements on the monitoring and control capabilities of these systems. The development of state estimation algorithms to gain insight in the actual system state of a distribution network has resulted in a wide range of distributed and decentralized algorithms that make use of parallel computing to deal with scalability and improve computational efficiency. From these state estimation algorithms, the branch current based state estimation has been proven suitable for distribution networks in terms of computational performance and convergence, but suffer from the fact that nodal voltage calculations are required within each iteration of the state estimation algorithm. This is usually accomplished using a forward sweep from the slack node, but this method is not suitable for parallelization. Therefore, this work proposes to replace the forward sweep with a Newton-Raphson optimization for calculating the nodal voltages, which is suitable for parallelization. The applicability of using Newton-Raphson for calculating nodal voltages within the state estimation is proven using numerical results, which clears the way for future work to implement the Newton-Raphson method within state estimation in a parallelized approach.