Izudin Dzafic

Real Time Estimation of Loads in Radial and Unsymmetrical Three-Phase Distribution Networks

This paper presents a load estimator for large scale asymmetric distribution networks. It is based on iterating between 1) a weighted least squares (WLS) estimator that attempts to adjust the forecasted load group values to the telemetered measurements and 2) a three-phase load flow employing the Fortescue transformation. The WLS estimation phase explicitly handles Y- and Δ-connected unbalanced loads and area losses; it can also identify bad measurements based on the estimated load scaling factors. Numerical results show that the proposed load estimator is superior to conventional load allocation based on current balancing, and that its performance is commensurate with the requirements of real-time applications on unsymmetrical radial distribution networks having up to one million nodes.

Three-phase power flow in distribution networks using Fortescue transformation

Analysis of power systems is typically made using the method of either phase coordinates or symmetrical components depending on the type of network (un/symmetrical). The use of the symmetrical components method, a three-phase specific application of the Fortescue transformation, has been limited to pure three-phase symmetrical networks. Therefore phase coordinates has been the method used most for the analysis of distribution networks. This paper is extending the use of the symmetrical components method to distribution networks combining one-phase, two-phase, and three-phase symmetrical network parts. This paper derives a new integrated approach for the description of such networks based on Fortescue transformation. This approach is demonstrated using the forward-backward sweeping algorithm (FBS).

Robust Optimization of Storage Investment on Transmission Networks

This paper discusses the need for the integration of storage systems on transmission networks having renewable sources, and presents a tool for energy storage planning. The tool employs robust optimization to minimize the investment in storage units that guarantee a feasible system operation, without load or renewable power curtailment, for all scenarios in the convex hull of a discrete uncertainty set; it is termed ROSION-Robust Optimization of Storage Investment On Networks. The computational engine in ROSION is a specific tailored implementation of a column-and-constraint generation algorithm for two-stage robust optimization problems, where a lower and an upper bound on the optimal objective function value are successively calculated until convergence. The lower bound is computed using mixed-integer linear programming and the upper bound via linear programming applied to a sequence of similar problems. ROSION is demonstrated for storage planning on the IEEE 14-bus and 118-bus networks, and the robustness of the designs is validated via Monte Carlo simulation.

A Sensitivity Approach to Model Local Voltage Controllers in Distribution Networks

Local controllers are essential in distribution networks; they are employed in classical devices such as load tap-changing (LTC) transformers and switchable shunt capacitors, and more recently in distributed generation (DG). The effective use of distribution management system (DMS) applications requires an accurate model of the interaction between the local controllers through the distribution system. This paper presents a new sensitivity matrix approach for modeling such interactions, and demonstrates its application in the implicit ZBus Gauss method for power flow computation. The sensitivity method models both PV buses (for the connection of DG) and tap position adjustments through current source injections, and consequently avoids re-factorization of the network bus admittance matrix. Numerical results on distribution networks with up to 3145 buses show that the sensitivity-based power flow method for simulating the operation of local controllers is superior to a sequential control action adjustment approach previously proposed in the literature, and that its computing time is commensurate with the performance requirements in real-time DMS applications.

High performance State Estimation for smart grid distribution network operation

Distribution System State Estimation (DSSE) stays in focus in all smart grid related topics. In Medium Voltage (MV) networks, in particular, and in distribution networks, in general, DSSE is a prerequisite for smart grid functionality. In this paper a new approach for a three-phase DSSE is presented. It utilizes the most types of analog real time measurements (including current magnitude, active and reactive power measurements). Loads are modeled based on historical or AMI/AMR information. Applying traditional state estimation on distribution networks, by using load data as pseudo measurement and zero injection constraints, causes state estimation formulations that have very large size. Approach presented in this paper is based on network reduction which reduces size of the problem. The estimation problem is solved by weighted least squares error method. The approach is very fast and provides feasible results

State Estimation in Two Time Scales for Smart Distribution Systems

The monitoring of distribution systems relies on a critical set of pseudomeasurements and a varying but low number of redundant measurements. In the light of the different refreshing rates of both types of information, this paper considers a state estimation model structured in two time scales. Possibilities and limitations of the proposed model are discussed, and illustrated on a real distribution system comprising a diversity of load patterns.

Real-time Distribution System State Estimation

In this paper a new approach for a three-phase Distribution System State Estimator (DSSE) is presented. It utilizes the most types of analog real time measurements (including voltage and current magnitude, active and reactive power measurements). The correct utilization is supported by topology analysis and creation of measurement areas. Loads are modeled based on historical or AMR information. The estimation problem is solved by weighted least squares error method. Applying traditional state estimation on distribution networks, by using load data as pseudo measurement, causes state estimation formulations that have very large size. Approach presented in this paper is based on network reduction which reduces size of the problem. The approach is very fast and provides feasible results.

Generalized

This paper develops a generalized admittance matrix approach in Fortescue coordinate system to solve unbalanced/unsymmetrical distribution networks including different number of phases. This generalized Fortescue π equivalent is defined in this paper for solving the heterogeneous phase, and thus Fortescue, network model. The performance of the approach is demonstrated in different model networks with number of nodes ranging between 168 and 14200. It is found that the current iteration method exploiting the decoupling in admittance matrix in Fortescue coordinate is substantially faster than the typical unbalanced three-phase solution in phase domain. The method has a significant potential for application in real time active power network management.

\pi

This paper presents a symmetrical components approach for computing the phase-coordinates nodal admittance matrices of three-phase transformers, where each connection is comprised either of three single-phase units or one three-phase unit. The approach is based on assembling the symmetrical component circuits into the different three-phase connections; it can straightforwardly handle unsymmetrical transformer banks with different phase electrical characteristics or missing phases, and grounding impedances. The proposed approach easily accounts for the various clock numbers quoted in the British Standard (BS) and the International Electrotechnical Commission (IEC) Standard on power transformers; this presents an advantage as compared to classical phase-coordinates approaches that require defining a distinct connection matrix corresponding to each transformer vector group.

Fortescue Equivalent Admittance Matrix Approach to Power Flow Solution

This paper proposes an optimized approach for Real-Time Distribution System State Estimation (RT-DSSE) for solving three-phase radial, unbalanced and unsymmetrical type of networks by utilizing Active and Reactive Power, Current magnitude and Voltage magnitude measurements. As the preparation step, topological network analysis and network object model optimization is done to reduce the problem size. Loads are modeled using the Advanced Load Scheduling System to obtain best possible initial load values. Subsequently, special per phase measurement areas are created and pre-estimation of current magnitude measurements to power is done, further reducing the problem size. Only then is the main estimation algorithm executed by weighted least squares error method. Unlike the traditional state estimation on distribution networks, which by using load data as pseudo measurements produces formulations of very large size, this algorithm is designed to achieve real-time performance requirements in several optimizing steps. Results are presented to show estimation and performance achievements.