Steady State Estimation of Power Distribution Networks with Distributed Generation using Integrated DC Load Flow and Weighted Least-Square Algorithm

Monitoring of a power system is essential in order to ensure reliability, safety, and security. However, metering devices, which are commonly used in a power system are prone to errors. Most of the time, these errors occur at random and due to man-made means, environmental, or system failures that can greatly affect the whole power system. Several methods and estimation techniques are emerging to anticipate such occasions. In this study, the researchers performed power system state estimation on an IEEE 14-Bus Test Case with Distributed Generation injection using Weighted Least Square method. At least three metering devices that were subjected to errors were considered; there were also cases that contained zero reading and contingency. To be able to perform state estimation, the researchers performed DC load flow analysis to acquire the power flows of the line. Then, an error ranging from -5% to + 5% were injected to the selected meters. The data coupled with error was then simulated to the modelled WLS algorithm. To verify the result from the algorithm, it was compared to ETAP, and the researchers conducted T – Test. Based from the parametric test (or statistical test), the t – value is less than the t – critical. The estimated data from MATLAB is almost equal to the true value from ETAP. Therefore, the researchers were able to develop an accurate PSSE using WLS method.

I. Introduction

A power system needs to be monitored every time to ensure reliability, safety, and security. Monitoring system is essential to determine the real-time condition of the power system. Power system state estimation (PSSE) was introduced to determine the approximate value of different parameters in a system using sets of given measurements [1]. In using state estimation, state variables like bus voltages can be determined. Bad Data Injection is a potential cause of error that is purposedly used by attackers to alter data reading within the meters, causing it to malfunction [12]. State estimation is useful in predicting future errors that may account in the system, therefore resolving it as soon as it happens. It uses statistical approach in approximating the state variables in a power system where it uses different criterion such as “maximum likelihood criterion”, “weighted least – squares criterion”, “minimum variance criterion”, and many more [2]. Although there are other algorithms used in State Estimations such as Least Square, Fast Decoupled, Orthogonal Transformation, Particle Swarm, etc. [1], most studies involving state estimations do not consider the possibility of the meters in this case to be malfunctioning. In this paper, the researchers will use an algorithm to solve the steady state parameters of an IEEE 14Bus power system with distributed generation (DG) injection. The meter readings of the power system are to be included and are assumed to have imperfect measurements. The researchers will also estimate the readings of at least 3 metering devices that are assumed to be exposed to noise disturbance such as bad data injection.