Weak Grid Connection (CIGRE TB 671)

pandapower

A wind or PV plant behind a long cable, or on a grid with little synchronous generation, sees a weak grid: its voltage moves strongly with its own power, its converters can lose synchronism, and generic models stop being reliable. This study follows the method of CIGRE Technical Brochure 671, Connection of wind farms to weak AC networks (2016). It calculates the short-circuit ratios at the connection point, the MV collector and every turbine, for the system intact and with outages, classifies the connection as strong, weak or very weak, and tells you which studies the connection needs next.

Where to find it

Simulate → Weak Grid Connection (CIGRE TB 671). Requires a subscription. The diagram needs an External Grid and the plant's units as Static Generators. The offshore wind farm template (File → New → Renewables) is a ready example.

When to use

  • Early in a wind, PV or battery connection, to find out whether the grid is strong enough and how deep the dynamic studies must go
  • Offshore plants with a long AC export cable, where the MV collector is much weaker than the connection point
  • Several converter-based plants in one area that share the same fault level
  • Checking the minimum SCR the turbine manufacturer asks for at the turbine terminals
  • A first size of a synchronous condenser when the connection is weak

How to run

  1. Model the plant from the grid to the turbines, with the units as Static Generators in service.
  2. Click Simulate → Weak Grid Connection (CIGRE TB 671).
  3. Enter the minimum and maximum short-circuit power the grid operator gives for the connection point, and their R/X. Empty fields use the values of the External Grid.
  4. Enter the lowest SCR the turbines are designed for (ask the manufacturer).
  5. Optionally add nearby converter-based plants, the kinetic energy of the system for the rate of change of frequency, and the data of a synchronous condenser. Click Calculate.

Enter the grid's minimum fault level. The classification uses the minimum short-circuit power, so a placeholder value in the External Grid makes every plant look very weak. The offshore wind template has 100 MVA there; enter the value from the grid operator. The study warns when the minimum is below 1.5 times the plant rating.

Inputs

Group Inputs
Plant and connection point Connection point bus (automatic: the control bus of the Plant Supervisory Controller, else the External Grid bus), MV collector bus (automatic: for every turbine, the bus after its unit transformer), plant rated power (empty: the Static Generators in service), SCR base (MW as in TB 671, or MVA), frequency.
Grid strength External Grid, S"k min and max [MVA], R/X at minimum and maximum, and whether to include credible outages.
Technology Lowest SCR at the turbine terminals, the minimum SCR (MSCR) of this plant for the fault level allocation, and the power factor for the voltage sensitivity.
Nearby plants Up to three converter-based plants: bus, rated power, MSCR, and reactance to their own MV collector in % on their rating.
Rate of change of frequency Kinetic energy of the synchronous system [MW s], largest infeed loss (empty: this plant), kinetic energy lost with it, and the RoCoF the plant must ride through.
Synchronous condenser Where it connects (the main MV collector, the connection point, or any bus), x"d, step-up transformer reactance and inertia constant H.

Short-circuit ratios and limits

Fault levels are calculated as in IEC 60909 with the turbines and every other converter-based unit out of service, because converters add little fault current. With the converters out, S"k is what the Short Circuit study gives when you switch them off.

Index Where Definition
SCR Connection point S"k / P of the plant
ESCR Each MV collector bus and each turbine terminal S"k / (own P + Σ WPIF · P of every other unit), with the interaction factor WPIF = |Zij| / |Zii|. A unit shares the grid with every unit electrically close to it.
CSCR MV collector Fault level with all collector buses tied together, divided by the plant power
WSCR Group of plants Σ S"ki · Pi / (Σ Pi)², when nearby plants are entered

The connection is classified by the lowest value over the system intact (minimum fault level) and every outage that keeps the plant connected:

Category Connection point MV collector Next step
Strong SCR ≥ 5 ESCR, CSCR ≥ 4 Generic models are fine for the stability studies.
Weak 3 to 5 2 to 4 Detailed RMS models from the manufacturer.
Very weak < 3 < 2 Manufacturer models and EMT simulation, control tuning, mitigation.

X/R below 3 (impedance angle below 70°) at the connection point is reported as weak too, and the lowest turbine ESCR is compared with the manufacturer's minimum.

Results

  • Verdict: strong, weak or very weak, with the index and the case that decide it.
  • Planning flowchart: the study steps of TB 671 (fault levels, short-circuit ratios, lowest SCR, model and study depth), with the path of your plant highlighted.
  • Summary tiles: SCR at the connection point, ESCR / CSCR at the MV collector, lowest turbine ESCR, X/R, the largest plant the connection point can take, and the RoCoF.
  • Studies to run next: required, recommended and optional studies with the reason and the TB 671 clause. Open starts that study in Electrisim (Transient Stability, Eigenvalue Analysis, Grid Code P-Q and U-Q, Voltage Stability, Contingency, Harmonics, Short Circuit, Grid-Forming STATCOM Sizing).
  • Fault levels and short-circuit ratios: one row per case (system intact at minimum and maximum, each credible outage), and the outages that disconnect the plant.
  • MV collector buses and turbine terminals: S"k, SCR and ESCR per bus and per unit, weakest first. Click a turbine to show it on the diagram.
  • Voltage sensitivity: the voltage at the connection point and at the turbines for full and half output, injecting and absorbing reactive power; the power at voltage collapse; the power factor with the least voltage change; and P-V curves for several reactive power levels.
  • Fault level allocation: each converter plant uses up MSCR × its rating of the fault level. The study shows what is left at each plant before and after this plant, and the largest plant that fits.
  • Rate of change of frequency after the largest infeed loss, and the kinetic energy needed to stay below 1 Hz/s, 2 Hz/s and the plant's ride-through limit.
  • Synchronous condenser: the rating that lifts the weakest index out of "very weak" and out of "weak", per criterion, with the indices after it and the inertia it adds.
  • Download CSV and Save image of the P-V curves.

Example: offshore wind farm template

450 MW in 30 units, two 66 kV collectors, a 275 kV export system of about 100 km and a 400 kV connection point. With a grid minimum of 10 000 MVA the connection point is strong (SCR 22), but the MV collector is weak (CSCR 2.2) and the turbines see an ESCR of 1.7, below a manufacturer minimum of 2.0. The export system is radial, so its outages disconnect the plant rather than weaken it. A synchronous condenser of about 56 MVA at a 66 kV collector lifts the turbines to 2.0.

How it works
  • The bus impedance matrix is built from the IEC 60909 network of pandapower (minimum and maximum case, voltage factor c, line end temperature) with the converter-based units out of service. The impedances between the connection point, the collectors and the turbines come from it, so the fault levels equal pandapower's short-circuit results.
  • Credible outages are N-1 of lines, transformers, generators and, when there is more than one, External Grids on the grid side and in the export system. A line or transformer with parallel circuits loses one circuit. The collector system inside the plant is not outaged.
  • Nearby plants are modelled as an MV node behind the reactance you enter. For the fault level allocation each plant is a source that takes up MSCR × its rating.
  • The voltage sensitivity solves the two-bus equation exactly for the grid impedance seen at the bus, with the grid source at 1.0 p.u.
  • The synchronous condenser is added as a source behind x"d plus its transformer, and its rating is found by bisection for each target.
  • The calculations reproduce the worked examples of TB 671 (Appendix B voltage sensitivity, the ESCR / CSCR / WSCR table of Chapter 6, the fault level allocation example and the RoCoF example); these are part of the automated tests.

Limits of the study

  • These are screening indices. They decide how deep the dynamic studies must go; they do not replace them.
  • The voltage sensitivity uses a Thevenin equivalent: line charging and shunt reactors are not in it.
  • The limits are those of TB 671. Grid operators and manufacturers may use other limits; follow theirs where they differ.

Last updated: November 2026

Electrisim on YouTube · Frontend on GitHub