Learnings

Outcomes

The project has developed:

• A multi-Party Agreement which enables the ESO and other party to acquire and validate a new system model that will enhance, as well as de-risk the integration of new technologies.    
• PowerNova have completed the build of the PSCAD models for South East Coast part of GB network using manufacture provided models.  
• A variety of validation tests were undertaken to show the suitability of this model for the assessment of South East Coast converter stability This includes Comparing simulated and actual grid performance during a fault, Comparing a load flow study of the South East Coast undertaken by ESO using DigSilent PowerFactory with the pre-fault steady-state power flows within the PSCAD grid model.
• PowerNova have completed various stability studies assessment of the South East Coast network by applying worst double circuit fault in the region under different scenarios of import level from converter based technologies. Previous studies have shown that this near-instantaneous phase-angle jump can be a challenge for converters with slow PLLs.  
• Plan drafted for PowerNova to support NGESO in applying the learnings from stability study and replicate the analysis at another region with high convertors penetration (e.g. East Angelia) and also develop a methodology for root cause analysis and identification of the suspected convertor(s) in cases of instability.
• Work Package 1-4 (WP1-4) Reports have been completed .   

Lessons Learnt

Model Provision and sharing

The process of requesting EMT models retrospectively has proven to be quite challenging especially for equipment that has been commissioned a long time ago and for which the appropriate EMT models have not been developed. Manufacturer models require a considerable amount of time and effort to be produced and validated and should ideally be requested at the point of connection to the network. This approach is currently being implemented more regularly as part of the Bilateral Connection Agreement (BCA) during the connections process. The Grid Code Working Group (GC0141) is also progressing a modification that should hopefully allow easier access to EMT models in the future. As the system becomes more complex, there will be a growing need to conduct EMT simulations and having accurate models that have been fully validated to represent the performance of equipment is a key component of the process.  

The additional challenge which was faced in this project is that even though some EMT models were available to NGESO, it was not possible to pass these on to a third-party consultant without an agreed NDA in place. Going forward, there may be a greater need to share black-box models across network connectees to verify the interaction between equipment and the use of standard NDAs where possible would help facilitate the process. 

In general, we have taken it for granted that the PSCAD model we receive from a manufacturer contains an accurate representation of the control system installed in the field. However, this may not always be the case in some cases: it seems that the software in the actual equipment can be updated (including updates specifically designed to change stability performance) without the PSCAD model being similarly updated. Presumably as the converter gets older the discrepancy between PSCAD and reality will widen. This is a matter of considerable concern and may require grid code changes to ensure the models will be kept up to date. A comparison of modelled converter behaviour to actual measured converter behaviour during faults or other system disturbances would also be desirable. 

While the manufacturer models were provided with documentation, the quality of the documentation could be quite variable. Frequently the documentation focused on how the user could get the model to run, but provided little insight into the myriad of user-accessible control settings that could be varied. Where the model is site-specific it is probably reasonable to assume that the settings it is delivered with are appropriate. However, the absence of documentation means that even for generic manufacturer models  it is necessary to make the same assumption. 

EMT Simulations

The outcome of WP2 provides useful recommendations on when and how to conduct EMT studies. The use of PSCAD has been recommended for conducting EMT simulations mainly due to the fact that most manufacturers have their equipment modelled in the PSCAD environment, that the software package allows simulations to be run using multiple CPU cores in parallel, and is used by at least two TSOs worldwide. This reinforces the need for the TOTEM project (NIA_SHET_0032) which is looking to develop an EMT model of the GB transmission system in PSCAD. 

WP2 also recommends the use of short circuit ratio (SCR) to screen when EMT studies are required based on the recommendation from CIGRE TB 671 “Connection of Wind Farms to Weak AC Networks” – which sets out the current international consensus on studying converter-dominated grids.  

System Instability is more likely to occur under high levels of non-synchronous generation and WP2 recommends the use of automated systems that can run through multiple generation and demand scenarios as well as contingencies. In this project, a script has been developed that allows an automatic dispatch of generation, running of simulations and extraction of results from PSCAD. 

Review of benefits case  
 
Valuable lessons shared on model sharing and the best practices for running wider EMT studies are very useful for the ESO and TOs. This is the first time a wide region of the GB system has been modelled in detail with actual users model for stability analysis in EMT environment.