State estimation in power engineering using the Huber robust regression technique

The least-squares method has been used extensively for state estimation in electric power engineering. However, this method is not always suitable when the process matrix is characterized by structural defects. This letter demonstrates the Huber function technique in a power engineering application. This technique is a slight departure from the least-squares method and downweighs large residuals. A statistical test to identify possible structural imperfections in the model is introduced.