In contrast to the large number of buses and loads in a distribution network, a very limited number of measurements are available. In practice, measurements are only available at distribution substations, transformers and some important loads. The small number of direct measurements is generally insufficient to make the system observable and hence obtaining accurate state estimation in distribution networks is a major challenge.
Although the installation of Automatic Meter Reading (AMR) and Remote Terminal Units (RTUs) could provide more direct measurements, the cost of implementing large number of such measurement devices to improve the observability of the system is significant and their widespread installation is therefore economically not justifiable. This will lead to a tool which enables best possible performance at minimum measurement system complexity and cost.
Dynamic State Estimation (DSE) of the static state of the network leverages the information in the time history of few available measurement points to improve estimation robustness and accuracy. DSE furthermore possesses the ability to predict the state of the system one step ahead. This forecasting ability has tremendous advantages in performing security analysis and allows more time for the operator to take control actions. Furthermore, DSE can easily integrate any type of measurement information (e.g. voltage, current, power measurements), and can be applied on any network type (e.g. looped, meshed, radial).
The project aims to develop methodologies to determine the minimum number of measurement elements to obtain a proper estimation of the grid state. The previous tools will be used for a set of show cases, and the aim with be to show the benefit of these techniques to a) ensure stable behaviour of the electrical grid, and b) optimize integration of energy sources into the grid.
- Optimal automatic meter reading & remote technical unit placement to enhance observability of electrical network.
- Dynamic state estimation to leverage sensor information for estimating the state of the network and state prediction.
- Show cases to demonstrate the benefits of the previous monitoring tools.
- Tools to enable the best possible performance at minimum measurement system complexity and cost.