Practical System Modelling

This step is beginning of engineering part. Nowadays, it is common and reasonable to use power system software to perform analysis.  Most common in use are SKM, ETAP, Easypower, DigSilent, CYME and others. It still can be done by hand or with use of Excel, however complexity of electrical systems will make it not efficient and costly. CAE software is a great and helpful tool, but it is only as good as the person that is using it. We need to remember that what comes IN goes OUT. Proper modelling is a must. 

System modelling requires us to establish few areas: 

• System topology and operations modes 
• System prospective short circuit current  
• Equipment data 
• Electrodes configurations (new from 2018) 
• Required scenarios   
• Assumptions  

System topology and scope of work 

Once we collected all the data or enough data, modelling can be started.  It can vary from one software to another, but a common part is to create a digital model as base for further modifications and adjustments. In Step 1 we focused on scope of work and here is why. We can evaluate: 

• complete system (in electrical terms) 
• part of the system  

Complete system study allows to determine multiple scenarios and finding min and max short circuit current. If we evaluate part of the electrical system, we need precise data about short circuit current in the evaluated part. It is very common for short circuit studies to focus only on max values for equipment sizing. The first tricky thing behind arc flash is that max current doesn’t always lead to the highest incident energy. 

Operation modes 

When system model is ready, attention is needed for scenarios, proper standards for calculations and adjusting analysis options. While software generally helps, it is only tool that can be used good or bad.  We are following mostly IEEE 1584 method for calculating arc flash in AC systems. This for European market requires some adjustments: 

• short circuit analysis method acc. IEC 60909 
• busbar gaps to follow IEC 61439 and IEC 62271-1 (recommended) 
• working distance is another topic, however IEEE 1584 standard options are reasonable  

Scenarios

Our base model is ready, so it is time to create scenarios.

We need to keep in mind that our arc flash analysis focus is to identify and provide recommendations for improvements. This means that scenarios need to take into account :

• Short circuit study results 
• Protective device coordination (selectivity) 
• Arc flash analysis results 
• *Load flow study (if included) 

Optionally, depending on our system we can consider: 

• BESS system connected 
• PV/WIND, etc. system connected 
• UPS, RUPS, DRUPS systems  
• CLR (Current Limiting Reactor) in operation 
• Specific switching arrangements (parallel transformers, multiple generators, etc.) 
• Protection specific functions modes (ARMS, arc fibre optic detection, arc quenching devices, differential protection, etc.) 

Equipment and configuration 

Data retrieved from site visit, documentation and provided sources need to be carefully implemented into digital model. Sometimes it requires verification of database and validation of available models for errors. Those issues have different weight, and it is related with installation size.  Most recommended is to use data verified by site visit, especially for settings of protective devices. Electrode configurations are a new topic from latest 2018 release of IEEE 1584, and it will be discusses separately in next chapter. 

Assumptions 

From my experience so far, assumptions are almost impossible to avoid but need to be minimised. This happens because often it takes too much time to retrieve data from: 

• utility operator  
• locations that can’t be de-energised due to ongoing process (sometimes it can take months to get the data) 
• restricted access to locations (eg. transformer rooms or utility switch rooms) 
• old equipment (missing data sheets, etc.) 

Usually some reasonable assumptions can be made based on experience. It is always project specific, but some general assumption list can be made, like: 

• transformer short circuit voltage/impedance rating 
• fuse types and tripping curves  
• low voltage breakers types and settings (similar model based) 
• system short circuit power  (depends on configuration, sometimes it has little effect for end results) 
• induction motors data 
• switchgears and panels ratings 
• cable lengths and type (this is tricky and requires results review).