My Contribution

1.

Developed an automation package to model a power system based on a set of parameters.

2.

Applied consensus-based alternating direction method of multipliers (ADMM) for distributed state estimation.

3.

Designed distributed moving horizon estimation (D-MHE) for power system state estimation (PSSE).

•

Guarantee robustness in state estimation by dealing with constraints within the objective function

•

Validated the robustness of state estimation to virtual data attack scenarios.

#### MHE Problem formulation for PSSE

1.

MHE concept diagram

2.

Problem Formulation

In Details

#### D-MHE for real-time state estimation

Assume that each local estimator can exchange information with its neighbors.

1.

Consensus form of optimization

•

${\mathcal I}_{\mathbb E}, {\mathcal I}_{\mathbb F}$ are the indicator functions corresponding to the equality and inequality constraints.

2.

Operator splitting algorithm for consensus-based optimization problem (3 steps)

a.

Solving an equality-constrained NLP

b.

Computing a set of separable projections

c.

Updating the scaled dual variables

#### Numerical Case Study - IEEE 118-bus test case

1.

Convert each transmission line into a partial 3-line physical mode to build a distributed power system

2.

Seperate measurements to enable distributed state estimation for each local estimator $\epsilon_i$

3.

Simulation Results (D-MHE vs D-EKF)

•

Validated the robustness of state estimation to virtual data attack scenarios.

◦

Nonlinear measurement function

◦

Linear measurement function