Improved unscented Kalman filter based interval dynamic state estimation of active distribution network considering uncertainty of photovoltaic and load

Abstract

State estimation of active distribution network (ADN) plays an important role in distribution energy management system. The increase penetration of distributed generations, especially the distributed photovoltaic (PV), in ADN leads to high uncertainty of ADN’s operation and the state of the ADN varies with the variation of the PV output power. For the uncertainty of PV power output, an interval dynamic state estimation (IDSE) method, which estimates the interval of ADN state variables is proposed in this paper. Firstly, considering the slow computation speed of the Unscented Kalman Filter (UKF), the square root UKF is used to predict the real-time operating level of the state variables. Secondly, since the power output of PV has features of the variation randomly, the neural network-based prediction intervals is employed to predict the power output interval of PV. Finally, the normal fluctuation range of ADN state is modelled as a bilevel non-linear programming problem to perform IDSE, which in turn monitors the operating state of ADN. The proposed method is evaluated on the IEEE 33-node and IEEE 123-node systems, respectively. The test results demonstrate that the dynamic status of the ADN can be tracked accurately using the proposed method.

Research on deep-learning-based photovoltaic output estimation and multi-layer collaborative load prediction in distribution networks

“…The second type is data-driven methods, which do not involve any explicit modeling and have high prediction accuracy in practical applications. However, the main limitation is that they require a large amount of historical photovoltaic output data [6]. Although photovoltaic output information can be obtained by installing metering devices for each distributed photovoltaic system, due to the large number of distributed photovoltaics, installation costs are relatively high.…”

Research and Application of Power Monitoring System Based on Internet of Things

“…In order to solve the above problems, this paper studies a method of applying Internet of Things technology to the power monitoring system to achieve remote monitoring of power equipment, reduce capital investment and accident risks, and integrate the unscented Kalman filter algorithm into the power monitoring system The software platform enables accurate processing and analysis of power data and provides strong support for the monitoring and maintenance of the power system [6]. The monitoring platform studied in this paper can well solve the problems of high power system maintenance costs and untimely fault detection [7].…”