Sparse and Orthogonal Method for Fast Bad Data Processing in Distribution System State Estimation

Real-time operation of Distribution Systems (DSs) demands the processing of a large volume of data in order to obtain the estimated state of the network. Accuracy and efficiency are requirements of the state estimation process which not only obtains the state variables but also detects gross errors in the input data. DSs may present ill-conditioning which badly affects the traditional WLS estimator and in this work an orthogonal formulation is presented so that the estimated state, as well as the processing of gross errors, are done using an computationally efficient and numerically robust method. Simulations are performed using three-phase IEEE test feeders and the results show that the detection of gross errors is effective with this orthogonal formulation. Sparse techniques are used to increase computational efficiency.