Abstract
The ongoing battle against global warming is rapidly increasing the amount of renewable power generation, and smart solutions are needed to integrate these new generation units into the existing distribution systems. Smart grids answer this call by introducing intelligent ways of controlling the network and active resources connected to it. However, before the network can be controlled, the automation system must know what the node voltages and line currents defining the network state are.
Distribution system state estimation (DSSE) is needed to find the most likely state of the network when the number and accuracy of measurements are limited. Typically, two types of measurements are used in DSSE: real-time measurements and pseudomeasurements. In recent years, finding cost-efficient ways to improve the DSSE accuracy has been a popular subject in the literature. While others have focused on optimizing the type, amount and location of real-time measurements, the main hypothesis of this thesis is that it is possible to enhance the DSSE accuracy by using interval measurements collected with automatic meter reading (AMR) to improve the load profiles used as pseudo-measurements.
The work done in this thesis can be divided into three stages. In the first stage, methods for creating new AMR-based load profiles are studied. AMR measurements from thousands of customers are used to test and compare the different options for improving the load profiling accuracy. Different clustering algorithms are tested and a novel twostage clustering method for load profiling is developed. In the second stage, a DSSE algorithm suited for smart grid environment is developed. Simulations and real-life demonstrations are conducted to verify the accuracy and applicability of the developed state estimator. In the third and final stage, the AMR-based load profiling and DSSE are combined. Matlab simulations with real AMR data and a real distribution network model are made and the developed load profiles are compared with other commonly used pseudo-measurements.
The results indicate that clustering is an efficient way to improve the load profiling accuracy. With the help of clustering, both the customer classification and customer class load profiles can be updated simultaneously. Several of the tested clustering algorithms were suited for clustering electricity customers, but the best results were achieved with a modified k-means algorithm. Results from the third stage simulations supported the main hypothesis that the new AMR-based load profiles improve the DSSE accuracy.
The results presented in this thesis should motivate distribution system operators and other actors in the field of electricity distribution to utilize AMR data and clustering algorithms in load profiling. It improves not only the DSSE accuracy but also many other functions that rely on load flow calculation and need accurate load estimates or forecasts.
Distribution system state estimation (DSSE) is needed to find the most likely state of the network when the number and accuracy of measurements are limited. Typically, two types of measurements are used in DSSE: real-time measurements and pseudomeasurements. In recent years, finding cost-efficient ways to improve the DSSE accuracy has been a popular subject in the literature. While others have focused on optimizing the type, amount and location of real-time measurements, the main hypothesis of this thesis is that it is possible to enhance the DSSE accuracy by using interval measurements collected with automatic meter reading (AMR) to improve the load profiles used as pseudo-measurements.
The work done in this thesis can be divided into three stages. In the first stage, methods for creating new AMR-based load profiles are studied. AMR measurements from thousands of customers are used to test and compare the different options for improving the load profiling accuracy. Different clustering algorithms are tested and a novel twostage clustering method for load profiling is developed. In the second stage, a DSSE algorithm suited for smart grid environment is developed. Simulations and real-life demonstrations are conducted to verify the accuracy and applicability of the developed state estimator. In the third and final stage, the AMR-based load profiling and DSSE are combined. Matlab simulations with real AMR data and a real distribution network model are made and the developed load profiles are compared with other commonly used pseudo-measurements.
The results indicate that clustering is an efficient way to improve the load profiling accuracy. With the help of clustering, both the customer classification and customer class load profiles can be updated simultaneously. Several of the tested clustering algorithms were suited for clustering electricity customers, but the best results were achieved with a modified k-means algorithm. Results from the third stage simulations supported the main hypothesis that the new AMR-based load profiles improve the DSSE accuracy.
The results presented in this thesis should motivate distribution system operators and other actors in the field of electricity distribution to utilize AMR data and clustering algorithms in load profiling. It improves not only the DSSE accuracy but also many other functions that rely on load flow calculation and need accurate load estimates or forecasts.
| Original language | English |
|---|---|
| Publisher | Tampere University of Technology |
| Number of pages | 91 |
| ISBN (Electronic) | 978-952-15-4105-6 |
| ISBN (Print) | 978-952-15-4094-3 |
| Publication status | Published - 23 Feb 2018 |
| Publication type | G5 Doctoral dissertation (articles) |
Publication series
| Name | Tampere University of Technology. Publication |
|---|---|
| Volume | 1529 |
| ISSN (Print) | 1459-2045 |
Access to Document
- mutanen 1529
In reference to IEEE copyrighted material which is used with permission in this thesis, the IEEE does not endorse any of Tampere University of Technology's products or services. Internal or personal use of this material is permitted. If interested in reprinting/republishing IEEE copyrighted material for advertising or promotional purposes or for creating new collective works for resale or redistribution, please go to http://www.ieee.org/publications_standards/publications/rights/rights_link.html to learn how to obtain a License from RightsLink.