Search

The real-time monitoring and control of distribution systems are an essential step toward the future smart grid. This paper presents an interval state estimation model for low-voltage (LV) systems based on smart meter data provided by electrical customers. The formulation is based on a two-level approach, constituted of a nodal processor and a system state estimator. The nodal processor filters the smart meter data and provides a smaller amount of clean processed data to the upper level system state estimator. The system state estimator, based on the extreme value theory, provides the maximum and minimum values of the system states. To obtain narrow interval images, a scaling parameter is used to consider the fact that the errors of downstream smart meters are rarely at the maximum values simultaneously. The presented approach allows removing the assumption on the particular probability distribution of the smart meter measurement errors. Simulations are carried out on a sample LV network to show the characteristics and performance of the proposed approach.

State estimation has been widely used in power system energy management systems. However, the application of state estimation for integrated electrical and heating networks (IEHNs) remains in a preliminary stage. This paper addresses this issue by proposing a robust state estimation method for IEHNs based on the weighted least absolute value in conjunction with equality constraints. The robust performance of the proposed estimator resolves the disadvantages of existing combined state estimators. A heating load pseudo-measurement model based on an artificial neural network and real-time measurements is developed to suppress the negative effects of measurements that contain bad data, and thereby ensure an adequate basis for accurate state estimation and guarantee the observability of the heating network. The effectiveness of the proposed state estimation method and its robustness to bad data are verified by comparison with the performance of the conventional largest normalized residual test based on the equality-constrained weighted least squares state estimation of IEHNs in numerical simulations employing a simple IEHN and/or the Barry Island IEHN as case studies.

The number of real-time supervisory control and data acquisition (SCADA) measurements in power distribution systems is scarce. This limits the reliability of state estimation (SE) results for distribution systems. Therefore, some studies seek to enhance the observability and SE accuracy of distribution systems by incorporating advanced metering infrastructure (AMI) data with the SCADA measurements. However, the hourly updated AMI data may be too coarse to capture system changes, especially in the presence of intermittent renewable energy sources. This issue is addressed by proposing a hybrid SE framework integrating a data-driven estimator and a model-based estimator. To be specific, the data-driven estimator combined with a topology identification method is presented to solve the DSSE problem between AMI scans, and the model-based estimator is employed to ensure robust estimation results against gross errors at a lower time scale. The proposed hybrid SE switches from the data-driven estimator to the model-based estimator once the AMI data is updated. Such a solution allows for capturing system changes at different time scales and improving the real-time and reliability of distribution system state estimation. Simulations are conducted on a sample distribution system to illustrate the characteristics of the proposed hybrid SE.

Hybrid AC/DC distribution systems are becoming a popular means to accommodate the increasing penetration of distributed energy resources and flexible loads. This paper proposes a decentralized robust state estimation (DRSE) method for hybrid AC/DC distribution systems using multiple sources of data. In the proposed decentralized implementation framework, a unified robust linear state estimation model is derived for each AC and DC regions, where the regions are connected via AC/DC converters and only limited information exchange is needed. In this context, estimation accuracy may be suffering due to linearization. To enhance the estimation accuracy, a deep neural network (DNN) based on the smart meter data is used to extract hidden system statistical information and allow deriving nodal power injections that keep up with the real-time measurement update rate. This provides the way of integrating smart meter data, SCADA measurements and zero injections together for state estimation. Simulations on two hybrid AC/DC distribution systems show that the proposed DRSE has only slight accuracy loss by the linearization formulation but offers robustness of suppressing bad data automatically, as well as benefits of improving computational efficiency.

To accommodate a higher penetration of distributed energy resources, distribution systems are moving toward hybrid AC/DC configurations for secure and economic operation. In this regard, this paper proposes a forecasting-aided state estimator (FASE) for hybrid AC/DC distribution systems to obtain accurate estimates for online security monitoring and control. The proposed FASE is designed in a distributed framework, with decomposition into several subproblems and solution by a constrained ensemble Kalman filter algorithm. In the proposed methodology, a deep neural network-based state forecasting model is developed to imitate the complex temporal and spatial relationship between system states, avoiding the state transition model built by unfounded explicit formulations. Furthermore, smart meter data is integrated by deep regression learning to obtain power injections of consumers and address the system observability issue. Extensive comparisons with two alternatives are carried out on a sample 33-node hybrid AC/DC distribution system to show the effectiveness and benefits of the proposed FASE, and on a larger 106-node hybrid AC/DC distribution system to demonstrate scalability.

Conventional state estimators, such as the weighted least squares estimator, have been widely employed for state estimation in power systems. However, these methods have difficulty addressing the leverage effect and rejecting erroneous measurements simultaneously. In light of these factors, this paper proposes a multi-objective robust state estimator for systems measured by phasor measurement units (PMUs) to eliminate the leverage effect and reject erroneous measurements automatically. To reduce the computational complexity, the multi-objective state estimation is solved by the normal boundary intersection method, and is then transformed into a series of single-objective linear optimization problems. Numerical simulations based on the IEEE 14-bus system and an actual provincial 760-bus system monitored by PMUs are conducted to verify the effectiveness and robustness of the proposed estimator compared with two other estimators.

Grid security is the lifeline of civilization. As one of the most important function in State Estimation (SE), Bad Data Detection (BDD) is vulnerable to false data injection attacks. Therefore, it is very important to develop patent and effective method to detect false data injection attacks. In this paper, we propose an improved cusum algorithm and synergy with the Unscented Kalman Filter (UKF) to detect false data injection attack. The proposed algorithm tested on IEEE-14 systems successfully, confirm the advantages of the approach.

The modern power system is the integration and development of the physical grid and the information system. However, state estimation in the modern power system faces the rising threat of false data injection attacks, which lead to malfunction of the relay protection, dispatcher's misjudgment or faulty operation. In this paper, we propose an improved detection method to detect false data injection attacks based on the Jensen-Shannon distance. This method tracks the dynamics of measurement variations and calculates the distance between two probability distributions derived from real-time measurement variations and historical measurement variations. When false data is injected into the modern power system, the real-time probability distributions will deviate from the historical ones, thus leading to a larger Jansen-Shannon distance. The proposed method is simulated on the IEEE14-bus system using real load data. Test results demonstrate that our method can accurately detect most of the attacks, and has good stability as well as practicability.

With the proliferation of intermittent generations in distribution system, it is vital to monitor the system states based on real-time data. This paper presents a historical data-driven unscented kalman filter for distribution system state estimation. The aim of the proposed method is to exploit the historical data to improve the accuracy of state prediction and filtering. The system model in prediction stage is built by time series process and its parameters are estimated by Huber estimation. In filtering stage, the nonlinear measurement equations are addressed by unscented kalman filter. This paper mainly investigates the predicting and filtering performance of the proposed method in different distribution systems. The results of the proposed method compared with weighted least squares (WLS) and extended kalman filter (EKF) are presented along with briefly conclusions.