Next Generation Grid Codes

• Next Generation Grid Codes are regulatory frameworks that specify device-level frequency-domain envelopes to guarantee dynamic ancillary service performance in inverter-dominated grids.
• They use decentralized passivity inequalities and quantitative performance limits—covering metrics like frequency nadir and RoCoF—to certify closed-loop stability across distributed resources.
• The transfer-function-based specification enables seamless integration with grid-forming converter control architectures, supporting plug-and-play compliance and robust testing.

Next Generation Grid Codes (NGGCs) denote a paradigm shift in the specification and verification of dynamic ancillary service requirements for power systems with high penetrations of converter-based generation. NGGCs supplant traditional, ad hoc time-domain prescriptions with formal, device-level frequency-domain envelopes for both stability and quantitative performance. They enable plug-and-play interoperability, scalability to large and inverter-based networks, and streamlined compliance testing. The following sections detail the core theoretical underpinnings, specification methodologies, emerging implications for converter design and control, and open challenges in the field.

1. Formal Definition and System-Level Objectives

NGGCs are regulatory or framework constructs that specify dynamic-ancillary-service requirements by imposing device-level frequency-domain envelopes on each resource’s transfer function, in place of historical step-response profiles. Their two core system-level objectives are:

• Guaranteed closed-loop stability: Ensured for the entire multi-device network under all admissible disturbances by verifying device-local frequency-domain conditions.
• Explicit performance guarantees: Achieved via quantitative bounds on critical time-domain metrics such as frequency nadir, rate-of-change-of-frequency (RoCoF), steady-state deviation, and oscillation damping for active-power–frequency (“pf”) dynamics, as well as analogous bounds (voltage peak, steady-state deviation, damping) for reactive-power–voltage (“qv”) dynamics (Häberle et al., 11 Jan 2026).

These objectives are achieved non-parametrically (model-agnostically): they require no centralized state-space models or system-level parameter sharing, but rather enforce compliance via well-defined, locally testable frequency-domain envelopes.

2. Decentralized Stability Certification Framework

Stability within the NGGC framework is certified using per-device frequency-domain passivity inequalities, enforced after suitable loop-shifting of the network and controller transfer functions. Let $D(s)$ denote the block-diagonal stack of device transfer functions in both “pf” and “qv” channels, and $N(s)$ the Kron-reduced network model,

$$D(s) = \mathrm{diag}(D_i^{\rm pf}(s),\, D_i^{\rm qv}(s))_{i=1}^n, \quad N(s) = \begin{bmatrix} N^{\rm pf}(s) & 0 \ 0 & N^{\rm qv}(s) \end{bmatrix}$$

The network $N^{\rm qv}(s)$ is loop-shifted via a diagonal correction $\Gamma^{\rm qv}$: $c_i = \sum_{j\neq i} b_{ij} 0.8/(1+\rho^2)$, resulting in a passive equivalent network $N'(s) = N(s)(I+\Gamma N(s))^{-1}$. The corresponding device transfer functions are shifted to $D'(s)$, which must satisfy:

• pf-channel (active power/frequency): Each $D_i^{\rm pf}(s)$ must be stable, strictly proper, and satisfy $\Re\{D_i^{\rm pf}(\j\omega)\} > 0\,\, \forall \omega \geq 0$.
• qv-channel (reactive power/voltage): Each $D_i^{\rm qv}(s)(1-c_i D_i^{\rm qv}(s))^{-1}$ must be strictly passive, or, equivalently, $\Re\{(D_i^{\rm qv}(\j\omega))^{-1}\} > c_i$.

These conditions are directly verifiable on each device’s Nyquist plot and are independent of other device or network parameters (Häberle et al., 11 Jan 2026).

3. Quantitative Performance Envelopes

NGGCs move beyond stability to deliver explicit performance guarantees aligned with system-level ancillary service metrics.

Frequency Dynamics

The “average mode” frequency response to aggregated active power disturbances is

$$\Delta f_{\rm avg}(s) = D_{\rm avg}(s) \Delta p_{\rm d}^\Sigma(s), \quad D_{\rm avg}(s) = \left(\sum_{i=1}^n (D_i^{\rm pf}(s))^{-1}\right)^{-1}$$

Spectral constraints on each $D_i^{\rm pf}(s)$ ensure:

• Nadir bound: $\|D_i^{\rm pf}(\j\omega)\|_\infty \leq \Delta f_{\max} / (2.5 \Delta p_{\rm d}^\Sigma)$ over all $\omega$.
• Steady-state error: $$|D_i^{\rm pf}(0)| \leq \Delta f_{\rm ss,max} / \Delta p_{\rm d}^\Sigma$$.
• RoCoF: $\lim_{\omega \to \infty} |\j\omega D_i^{\rm pf}(\j\omega)| \leq \Delta\dot{f}_{\max} / \Delta p_{\rm d}^\Sigma$.
• Damping: $\Re\{(D_i^{\rm pf}(\j\omega))^{-1}\} \geq \rho_{\rm f} > 0$ (Häberle et al., 11 Jan 2026).

Voltage Dynamics

Analogous constraints are enforced for each $D_i^{\rm qv}(s)$, guaranteeing peak and steady-state regulation and sufficient damping, with possible high-frequency roll-off to limit bandwidth (Häberle et al., 11 Jan 2026).

4. Transfer Function–Based Specification and Implementation

A central innovation of NGGCs is the formal mapping of historic piecewise-linear, time-domain service requirements into rational transfer-function templates for converter control.

• Piecewise step-response curves (e.g., FCR, FFR) are decomposed into Laplace-domain segment representations, assembled, and approximated as rational transfer functions using Padé approximations for time delays.
• The resulting transfer function

$$G(s) = \frac{a_n s^n + \ldots + a_0}{b_m s^m + \ldots + b_0}$$

is directly embeddable in converter outer-loop PI or robust MPC architectures via matching control schemes (Häberle et al., 2023).

• Compliance is validated both in the time-domain (by verifying step responses against profiles) and frequency-domain (by checking $G(s)$ magnitude and phase envelopes) (Häberle et al., 2023).

Specification Step	Description	Reference
Time-domain curve extraction	Define breakpoints $(t_i, y_i)$ for the needed response curve	(Häberle et al., 2023)
Transfer function synthesis	Map curve to $G(s)$ via Laplace segment construction, Padé delay approx.	(Häberle et al., 2023)
Controller integration	Insert $G(s)$ in outer PI/robust control loop	(Häberle et al., 2023)
Compliance test	Enforce frequency and time-domain matches to $G_{\rm req}(s)$	(Häberle et al., 2023)

5. NGGCs and Grid-Forming Converter Control

The NGGC framework is compatible with advanced GFM designs, such as Emulated Synchronous Condenser (ESC) plus current-source architectures, which inherently satisfy both stability and advanced grid code requirements including:

• Severe phase jumps,
• Balanced and unbalanced fault ride-through (FRT),
• Main grid disconnection (islanding),
• Black start capability.

A unified control structure (ESC with virtual swing equation, parallel active power current source, virtual impedance, and inner-loop elliptical saturation) enforces NADIR, ride-through, and steady-state metrics without explicit mode switching or PLLs (Freytes et al., 2023). In time-domain simulations of a 2-MW BESS, such architectures meet or exceed voltage and frequency ride-through envelopes, provide firm inertial support, and maintain stability under worst-case events, aligning with NGGC objectives (Freytes et al., 2023).

6. Field Size and Coding Requirements for Distributed NGGCs

For data and distributed control applications, NGGCs interface with grid-like code topologies where robust erasure recovery is critical. Maximally recoverable (MR) codes for $T_{m\times n}(1, b, 0)$ topologies have polynomial upper and lower field size bounds. Recent results establish:

• Upper bounds: $q > C_0(m, b) n^{2b(m-1)} + n(b-1)$ for general $m, b$, with explicit improvements for $m=3,4$.
• Lower bounds: $q = \Omega(n^2)$ or $q = \Omega(n)$ for small $m$, via Sidon set arguments (Kong et al., 2019).
• Polynomial field-size requirements render scaling to large $n$ (long codes, large system instances) practical and resource-efficient, supporting NGGC data anchoring and recovery under correlated failures.

MR Code Topology	Sufficient Field Size $q$	Necessary Field Size $q$	Reference
$T_{m\times n}(1, b, 0)$ general	$q > C_0(m, b) n^{2b(m-1)}$	---	(Kong et al., 2019)
$T_{4\times n}(1,2,0)$	$q > C_1 n^5\log n$	$q \ge (n-3)^2/4 + 2$	(Kong et al., 2019)
$T_{3\times n}(1,3,0)$	$q > C_2 n^5\log n$	$q \ge \sqrt{n^2-11n+34}$	(Kong et al., 2019)

7. Implications, Open Questions, and Future Directions

NGGCs constitute a rigorous, scalable foundation for dynamic ancillary service specification and device-level compliance in inverter-dominated and low-inertia grids. Principal implications and open problems include:

• Implications: Unified passivity-plus-performance envelopes enable manufacturers to offer plug-and-play resources and system planners to obtain explicit performance metrics. Automation of certification and compliance testing is facilitated by transfer-function-based specifications (Häberle et al., 11 Jan 2026, Häberle et al., 2023).
• Open questions: Sharper combinatorial characterization of erasure patterns for distributed code topologies, explicit deterministic MR code constructions at minimal field sizes, and generalization to $T_{m\times n}(a, b, h)$ topologies remain outstanding (Kong et al., 2019). For converter control, integrating parameter-varying transfer function templates to adapt to real-time grid conditions and formalizing higher-layer supervisory frameworks are fertile areas for future research.

NGGCs unify local device-level, frequency-domain certification with global system performance, recasting ancillary service regulation for the era of high renewable and inverter penetration (Häberle et al., 11 Jan 2026, Häberle et al., 2023, Freytes et al., 2023, Kong et al., 2019).