9.2 Inverter-Based Resources (IBRs), Grid-Forming & Grid-Following Controls

9.2 Inverter-Based Resources (IBRs), Grid-Forming & Grid-Following Controls

Executive Overview: The large-scale displacement of conventional thermal synchronous generators by Inverter-Based Resources (IBRs)—including solar photovoltaic (PV) plants, Battery Energy Storage Systems (BESS), and Type 4 wind turbines—represents a fundamental paradigm shift in power system dynamics. Unlike synchronous machines governed by rotor inertia and electromagnetic flux dynamics, IBRs are decoupled from the grid by power electronic converters governed entirely by millisecond-scale software control loops. The PE Power exam increasingly emphasizes the distinct control architectures (Grid-Following vs. Grid-Forming), weak grid stability constraints (Short Circuit Ratio), and the unique protection challenges arising from semiconductor fault current limitations.

---

1. IBR Generation Systems Architecture

Modern renewable generation and energy storage plants utilize distinct converter configurations to interface DC or variable-frequency AC energy sources to the high-voltage utility grid:

+---------------------------------------------------------------------------------------------------+
|                                 COMMON IBR SYSTEM ARCHITECTURES                                   |
+---------------------+-------------------------------+---------------------------------------------+
| Technology          | Front-End Power Source        | Converter Interface Topology                |
+---------------------+-------------------------------+---------------------------------------------+
| **Solar PV**        | DC Photovoltaic Arrays        | DC-DC Boost (with MPPT) + 3-Phase DC-AC VSI |
| **BESS**            | DC Battery Racks (Li-ion/Flow)| Bi-Directional 4-Quadrant DC-AC VSI         |
| **Type 4 Wind**     | Variable-Speed AC Generator   | Full-Scale Back-to-Back (AC-DC-AC) VSI      |
| **Type 3 Wind**     | Doubly-Fed Induction (DFIG)   | Partial-Scale Rotor Converter (~30% rating) |
+---------------------+-------------------------------+---------------------------------------------+

Maximum Power Point Tracking (MPPT)

Solar PV arrays exhibit nonlinear $I-V$ and $P-V$ output curves with a single unique Maximum Power Point (MPP) that shifts continuously with solar irradiance ($G$ in $\text{W/m}^2$) and cell temperature ($T$ in $^\circ\text{C}$). The front-end DC-DC converter executes MPPT algorithms—predominantly Perturb and Observe (P&O) or Incremental Conductance ($dI/dV = -I/V$ at the MPP)—to continuously adjust converter duty cycle and maximize harvested power.

---

2. Synchronous Reference Frame Control ($d-q$ Decomposition)

To control three-phase sinusoidal voltages and currents with zero steady-state error using standard Proportional-Integral (PI) controllers, three-phase stationary quantities ($a-b-c$) are transformed into a synchronously rotating reference frame ($d-q-0$) using Park's Transformation:

$\begin{bmatrix} x_d \\ x_q \\ x_0 \end{bmatrix} = \frac{2}{3} \begin{bmatrix} \cos(\theta) & \cos(\theta - 120^\circ) & \cos(\theta + 120^\circ) \\ -\sin(\theta) & -\sin(\theta - 120^\circ) & -\sin(\theta + 120^\circ) \\ 1/2 & 1/2 & 1/2 \end{bmatrix} \begin{bmatrix} x_a \\ x_b \\ x_c \end{bmatrix}$

Where $\theta(t) = \omega t = 2\pi f t$ is the instantaneous rotor or grid voltage phase angle.

Decoupled Active and Reactive Power Equations

When the $d$-axis is aligned with the grid voltage vector such that $v_d = V_m$ (peak voltage magnitude) and $v_q = 0$, instantaneous active power ($P$) and reactive power ($Q$) simplify into decoupled linear expressions:

$P = \frac{3}{2} (v_d i_d + v_q i_q) = \frac{3}{2} v_d i_d$ $Q = \frac{3}{2} (v_q i_d - v_d i_q) = -\frac{3}{2} v_d i_q$

• Direct-axis current ($i_d$): Directly controls Active Power ($P$) or DC-link voltage regulation.
• Quadrature-axis current ($i_q$): Directly controls Reactive Power ($Q$) or terminal AC voltage regulation.

---

3. Grid-Following (GFL) Controls & Weak Grid Instability

Historically, virtually all commercial solar, battery, and wind inverters have been deployed using Grid-Following (GFL) control architectures.

Operational Principles of GFL

• Current-Source Behavior: A GFL inverter behaves as a controlled current source ($I\angle\theta$) synchronized to the existing grid voltage.
• Phase-Locked Loop (PLL): The inverter measures the three-phase AC terminal voltage at the Point of Interconnection (POI) and uses a software PLL to track the grid phase angle $\theta_{pll}$. The calculated angle is fed into the Park transformation blocks to orient $i_d$ and $i_q$.
• Fundamental Assumption: GFL assumes the grid is an infinitely "stiff" AC voltage source whose voltage magnitude and frequency are established by external synchronous generators.

+-----------------------------------------------------------------------------------+
|                         GRID-FOLLOWING (GFL) CONTROL LOOP                         |
|
|  Grid V_abc ---> [ PLL ] -----------------------------> Phase Angle theta_pll ---+
|                    |                                                             |
|  P_ref, Q_ref ---> [ Current Reference Gen ] -> (i_d*, i_q*)                     |
|                                                        |                         |
|  Measured I_abc -> [ abc -> dq ] --------------------->+-> [ Current PI Loops ]  |
|                                                                 |                 |
|                                                          [ PWM Modulator ]        |
|                                                                 |                 |
|                                                         [ 3-Phase IGBT VSI ]      |
+-----------------------------------------------------------------------------------+

Weak Grids and the Short Circuit Ratio (SCR)

Grid strength at a POI is quantified by the Short Circuit Ratio (SCR), defined as the ratio of available three-phase short-circuit MVA ($S_{sc}$) to the rated MW capacity ($P_{rated}$) of the IBR plant:

$\text{SCR} = \frac{S_{sc}}{P_{rated}} = \frac{V_{LL,nom}^2}{|Z_{th}| \cdot P_{rated}}$

Where $Z_{th} = R_{th} + j X_{th}$ is the Thevenin grid equivalent impedance seen from the POI.

Mechanism of GFL Instability in Weak Grids

In a weak grid with high series reactance $X_{th}$, any step change in active current $\Delta i_d$ causes an instantaneous phase angle shift across the grid impedance: $\Delta \delta \approx \frac{\Delta P \cdot X_{th}}{V_{POI} \cdot V_{grid}}$ Because the GFL inverter's PLL attempts to track this shifting terminal voltage angle, a positive feedback loop is established between the PLL angle tracker and the current injection loop. When the grid SCR falls below $\sim 2.0$, this feedback loop becomes unstable, causing large sub-synchronous oscillations, control hunting, and inverter tripping.

---

4. Grid-Forming (GFM) Controls & Virtual Synchronous Machines

To overcome weak grid limitations and operate reliably in high-penetration renewable grids (or 100% inverter grids), Grid-Forming (GFM) control topologies have emerged.

Operational Principles of GFM

• Voltage-Source Behavior: A GFM inverter acts as an ideal, controllable AC voltage source behind a virtual coupling impedance ($E\angle\delta$).
• Independent Angle Generation: GFM does not rely on a fast PLL to track terminal voltage. Instead, it internally integrates its own internal frequency $\omega(t)$ to generate the phase reference angle $\theta(t) = \int \omega, dt$.
• Instantaneous Power Exchange: If a sudden load step occurs, the phase angle difference between the internal voltage $E\angle\delta$ and the grid voltage causes an instantaneous, intrinsic injection of active power without waiting for control sensor feedback: $P = \frac{E \cdot V_{POI}}{X_{virtual}} \sin(\delta)$

Droop Control Mechanisms in GFM

GFM inverters emulate the self-regulating power-frequency ($P-f$) and reactive power-voltage ($Q-V$) droop characteristics of synchronous governors and exciters:

1. Active Power - Frequency ($P-f$) Droop: $\omega - \omega_0 = -k_p (P - P_0) \implies f - f_0 = -R_p \cdot \frac{f_0}{P_{rated}} (P - P_0)$ Where $R_p$ is the per-unit droop percentage (typically 3% to 5%). If system frequency drops ($f < f_0$), the inverter automatically increases active power output $P$.
2. Reactive Power - Voltage ($Q-V$) Droop: $V - V_0 = -k_q (Q - Q_0)$ If terminal voltage drops below nominal ($V < V_0$), the inverter automatically exports reactive power ($Q$) to support grid voltage.

Virtual / Synthetic Inertia

GFM inverters can emulate the mechanical Swing Equation of a synchronous machine rotor in software:

$2H \frac{d\omega}{dt} = P_m - P_e - D(\omega - \omega_0)$

Where $H$ is the synthetic inertia constant (in seconds), $P_m$ is the virtual mechanical power order, $P_e$ is the measured electrical power output, and $D$ is the damping factor. By releasing or absorbing energy stored in the DC-link capacitors or battery cells, GFM inverters provide instantaneous inertial response that arrests the Rate of Change of Frequency (RoCoF) during major generation loss events:

$\text{RoCoF} = \left. \frac{df}{dt} \right|_{t=0^+} = -\frac{f_0 \cdot \Delta P_{loss}}{2 H_{total} \cdot S_{sys}}$

GFL vs. GFM Comparison Matrix

---

5. IBR Short-Circuit Fault Behavior & Protection System Impacts

The replacement of synchronous generators with IBRs fundamentally disrupts traditional utility protection schemes.

+---------------------------------------------------------------------------------------------------+
|                     SYNCHRONOUS GENERATOR VS. IBR FAULT CURRENT COMPARISON                        |
+----------------------------+-----------------------------------+----------------------------------+
| Parameter                  | Synchronous Generator             | Inverter-Based Resource (IBR)    |
+----------------------------+-----------------------------------+----------------------------------+
| **Initial Peak Current**   | $5.0 - 8.0\text{ pu}$ (subtransient $I_k''$) | $1.1 - 1.3\text{ pu}$ (software clamped) |
| **Steady-State Current**   | $1.5 - 2.5\text{ pu}$             | $1.0 - 1.2\text{ pu}$            |
| **Current Angle**          | Highly inductive (lagging $\sim 85^\circ$) | Variable (governed by control mode) |
| **Negative-Sequence ($I_2$)** | Natural physical injection ($I_2 = V_2 / X_2$) | Near zero (unless programmed)    |
| **Thermal Withstand Time** | Several seconds (rotor iron/copper)| Microseconds (IGBT junction limits)|
+----------------------------+-----------------------------------+----------------------------------+

Protection System Degradation Mechanisms

1. Desensitization of Instantaneous & Time-Overcurrent (50/51) Relays: Legacy overcurrent pickup thresholds are typically set at $1.5 - 2.0\text{ pu}$ of normal load. Because an IBR can supply no more than $1.1 - 1.3\text{ pu}$ fault current, overcurrent relays fail to detect downstream bolted faults.
2. Distance Relay (21) Underreach / Overreach: Distance relays compute apparent impedance $Z_{app} = V / I$. During faults, fast converter current limiters alter the phase angle between $V$ and $I$. If the inverter injects reactive current to support voltage while active current is capped, the apparent impedance measured by a line relay can rotate outside the mho circle, causing false tripping or failure to trip.
3. Failure of Directional Overcurrent (67/67Q) Elements: Traditional directional relays rely on negative-sequence current ($I_2$) or zero-sequence current to determine whether a fault is forward or reverse. Standard GFL inverters suppress negative-sequence current injection to protect IGBTs from unbalanced overcurrents, starving 67Q relays of the polarizing quantity needed for directional discrimination.

---

6. Worked Grid Integration Scenario

Problem: Short Circuit Ratio and GFM Droop Response Calculation

Scenario: A $150\text{ MW}$, $0.95$ power factor rated solar-plus-storage plant is interconnected to a $230\text{ kV}$ transmission substation. The utility Thevenin equivalent impedance at the $230\text{ kV}$ POI bus is $Z_{th} = 3.2 + j 28.5\ \Omega$.

Calculate:

1. The Short Circuit MVA ($S_{sc}$) at the POI and determine the plant's Short Circuit Ratio (SCR).
2. Classify the grid strength at the POI.
3. The plant's BESS utilizes GFM control with a $4.0%$ active power-frequency ($P-f$) droop on a $60.0\text{ Hz}$ base. If a sudden generation outage causes grid frequency to drop to $59.80\text{ Hz}$, calculate the resulting active power increase ($\Delta P$) delivered by the BESS (assume nominal rating $P_{rated} = 150\text{ MW}$ and sufficient headroom).

Calculation Workflow:
Step 1: Compute POI Short Circuit MVA and SCR
  Nominal Voltage V_nom = 230 kV (line-to-line)
  Impedance Magnitude |Z_th| = sqrt(3.2^2 + 28.5^2) = sqrt(10.24 + 812.25) = sqrt(822.49) = 28.68 ohms
  
  Three-Phase Short Circuit MVA:
  S_sc = V_nom^2 / |Z_th| = (230 kV)^2 / 28.68 ohms = 52,900 / 28.68 = 1,844.5 MVA
  
  Short Circuit Ratio (SCR):
  SCR = S_sc / P_rated = 1,844.5 MVA / 150 MW = 12.30

Step 2: Classify Grid Strength
  Since SCR = 12.30 >> 3.0, the POI is classified as a STRONG grid.

Step 3: Compute GFM Droop Frequency Response
  Base frequency f_0 = 60.00 Hz
  Disturbed frequency f = 59.80 Hz
  Frequency deviation Delta_f = f - f_0 = 59.80 - 60.00 = -0.20 Hz
  
  Droop equation: (Delta_f / f_0) = -R_p * (Delta_P / P_rated)
  Rearranging for Delta_P:
  Delta_P = - (Delta_f / f_0) * (P_rated / R_p)
  Delta_P = - (-0.20 Hz / 60.00 Hz) * (150 MW / 0.04)
  Delta_P = (0.003333) * (3,750 MW) = +12.50 MW
  
  The GFM BESS increases its active power output by 12.50 MW to support grid frequency.

---

7. Common Exam Traps & Strategic Pitfalls

• Assuming IBRs Provide High Subtransient Current: Never use traditional generator subtransient reactance ($X_d'' \approx 0.15\text{ pu}$) when calculating fault currents from solar or battery plants. IBR fault current is strictly limited to $1.1 - 1.3\text{ pu}$ by control software.
• Confusing GFL and GFM Synchronization: GFL requires a pre-existing AC voltage waveform to lock its PLL onto; GFM internally synthesizes its own voltage waveform and can operate in an isolated, dead-bus (black start) island.
• Droop Sign Inversion: A decrease in frequency ($\Delta f < 0$) must result in an increase in active power output ($\Delta P > 0$). Ensure proper negative sign application in the droop equation: $\Delta P = -\frac{\Delta f / f_0}{R_p} \cdot P_{rated}$.