The distributed generation and Low Carbon Technologies (LCTs) connected to the electricity distribution network have changed the way the network is operating, causing the voltage profiles of the various substation feeders to vary significantly depending on the type of connections they have. This variation in the utilisation of the system, increases the need to manage the voltages in the network locally and dynamically in order to make the most of the available network capacity. This project, through power system studies will explore the capability of releasing network capacity by controlling the power factor of existing generators and therefore optimising the voltages locally. As part of these studies a “Virtual Statcom” will be implemented. This will be an algorithm that will be coordinating the reactive power output of generators in order to release network capacity.
Benefits
The project will provide knowledge on whether a Virtual Statcom can be used to control the output of existing generators in order to increase network capacity. It will provide recommendations on the design of the Virtual Statcom algorithm, its implementation and it will also provide learning on how much capacity could potentially be released. The algorithm will be designed such that it can be applied on any network model, so that any other DNO could implement this solution to evaluate what capacity could be released using a Virtual Statcom in their network.
Learnings
Outcomes
The Virtual Statcom project was delivered in five work packages (WP), with the findings from each WP presented in a report:
- WP1 “Virtual Statcom design, study zone selection and data validation” Report– This deliverable presented; the findings from the literature review, the networks selected as study zones for the project, and data validation results for the selected study zones. The deliverable also presented the initial design for the Virtual Statcom hosting capacity and optimisation algorithms.
- WP2 “Power flow simulations and Virtual Statcom algorithms implementation” Report– This deliverable presented; the initial power system analysis of the study zones, the implementation of the Virtual Statcom algorithms and the results of the Virtual Statcom studies for edge case generation and load scenarios.
- WP3 “Graphical User Interface” Report – This deliverable presented the development of the GUI, the final GUI implemented as well as lessons learnt from the development of the GUI. The deliverable also provided a user guide for the Virtual Statcom GUI.
- WP4 “Time series comparison studies” Report – This deliverable presented the updates made to the Virtual Statcom algorithms and the results from time series studies using the Virtual Statcom algorithms.
- WP5 “Virtual Statcom feasibility study reporting” Report – This deliverable presented a summary of all work packages and conclusions from the project regarding the technical feasibility of the Virtual Statcom concept. The deliverable also provided functional specifications for a real time control system and planning time tool to utilise the Virtual Statcom concept.
The Virtual Statcom project delivered the following technical outputs:
- The Virtual Statcom hosting capacity algorithm to determine the available generation or load hosting capacity for a given network.
- The Virtual Statcom optimisation algorithm to resolve or reduce network constraints and optimise the reactive dispatch for existing generators.
- The Virtual Statcom GUI to run the Virtual Statcom Algorithms
The Virtual Statcom algorithms were used to investigate the feasibility of increasing generation and load hosting capacity for the following three 33 kV BSP networks and one 11 kV Primary network.
The Virtual Statcom hosting capacity algorithm delivered can be used to determine the available generation or load hosting capacity for a given network in each contingency configuration. The algorithm is also able to be used to determine the contingency configuration with the lowest hosting capacity, termed the traditional planning hosting capacity.
The Virtual Statcom optimisation algorithm has demonstrated the ability to resolve or reduce constraints in networks through optimised reactive power dispatch, this can reduce the amount of active power curtailment required to manage network constraints. There is the potential for a real time tool to use the Virtual Statcom optimisation algorithm to dispatch reactive power to resolve network constraints.
In networks with no network constraints the optimisation algorithm is able to minimise an objective function based on network power flows and bus voltages. The structure of the optimisation algorithm also allows different objective functions to be easily incorporated.
The Virtual Statcom algorithm has shown that little generation capacity can be released by optimising the reactive power output of existing generators. A trade‑off exists when optimising to increase thermal headroom and optimising to increase voltage headroom for generation. Due to this trade‑off, network complexities and interactions between different feeder groups, minimising objective functions developed in the project does not always provide an increase in generation hosting capacity for all network configurations, system generation and load profiles.
The Virtual Statcom has shown that load hosting capacity can be released by optimising the reactive power output of existing generators. The trade‑off that exists when increasing generation in a network is not present when increasing load. More specifically, optimising to increase thermal headroom and optimising to increase voltage headroom for load have the same outcome of higher system voltages and lower network losses.
There is the potential for a real time Virtual Statcom tool to use optimised reactive power dispatch to reduce losses and improve the load hosting capacity of a network to serve the expected increases in load from LCTs. Reduced losses in power system networks also reduces the environmental impact of supplying electricity and lowers costs to the end customer.
Lessons Learnt
The project was planned in a way that allowed significant amount of time for the Design phase. This proved to be a beneficial approach as it enabled a comprehensive assessment of the various methodologies to be made before choosing the most suitable solution for the project.
Additionally, as part of the Design phase, WPD’s experts in Primary System Design were involved in the project to ensure that their expertise and knowledge of the network was considered in the various design decisions. This influenced the selection of the network areas for the detailed studies and ensured that the right assumptions are made when running the network studies, applying a consistent approach that agrees with WPD’s Business As Usual practices.
