PaperlayerView exact revision on arXiv

Source-linked AI summary

Fixed-Time Voltage Regulation in Distribution Networks with Impedance Awareness

Nilanjan Roy Chowdhury, Venkatesh Sarangan

arXiv:2610.01449v1eess.SYcs.SY

TL;DR

Voltage regulation with uncertain impedance is challenging, and many existing approaches provide only asymptotic or exponential recovery. The paper derives fixed-time stability conditions and implements them through a quadratic-programming controller. Under stated parameter conditions, the analysis guarantees convergence in fixed time, while bounded control can make recovery finite-time for large disturbances.

  • Problem

    Voltage regulation is non-convex under uncertain network impedance, and many existing methods provide only asymptotic or exponential recovery.

    Source: §I · p. 1

  • Method

    The approach combines fixed-time stability and control Lyapunov analysis to derive gain and design-parameter bounds, then uses a QP framework to compute control set-points without knowing the exact system matrix B.

    Source: §I · pp. 1–2, §III · p. 3

  • Results

    When min{α1, α2} > ξδβ2√β1 and η ∈ (1, 2], the analysis guarantees convergence toward a sub-level set of diameter below 1 in fixed time.

    Source: A. Stability Studies of Fixed-Time… · p. 4

  • Takeaways & Limitations

    The derived conditions provide a basis for fixed-time voltage regulation under impedance uncertainty, with feasible set-points computed through the QP framework.

    Source: §I · pp. 1–2

    When a disturbance starts outside the safe operating envelope, bounded control degrades fixed-time convergence to finite-time recovery until voltage re-enters the envelope.

    Source: B. Limitation of Fixed-Time… · p. 5

Abstract

from arXiv · show

This letter introduces an optimization-based fixed time control algorithm for solving the voltage regulation problem of a radial and balanced power distribution network. The proposed algorithm requires no prior knowledge of the exact network impedance (resistance and reactance), yet guarantees voltage convergence to the predefined safe limit within a fixed time-window. We embrace results from Fixed-time stability (FxTs) and Control Lyapunov function (CLF) to analyze the stability and robustness of the underlying algorithm and then synthesize it leveraging the Quadratic Programming (QP) approach. We first analytically provide sufficient conditions on the control gains and the design parameters, that can ensure voltage regulation in fixed-time. Thereafter we transform the existing voltage regulation problem into an equivalent QP-based optimization framework and translate the aforesaid conditions to find feasible solutions for voltage stability. We empirically verify the efficacy of our proposed algorithm on the IEEE-33 bus distribution network and compare its performance against other existing robust control methods.

I. INTRODUCTION

Voltage regulation must address non-convexity and uncertain network impedances, while many existing methods guarantee only asymptotic or exponential recovery. This letter proposes a robust fixed-time controller that integrates stability analysis with a QP formulation.

I. INTRODUCTION

  • Voltage recovery time is the time required to return network voltages to the predefined safe limits after a sudden, large external disturbance.The safe limits are ±5% around nominal voltage.Source: §I · p. 1
  • Existing voltage-regulation methods often guarantee only asymptotic or exponential recovery, despite impedance uncertainty and non-convexity.Some data-driven and learning-based approaches also depend on data quality or ample historical training data,,.Source: §I · p. 1
  • The proposed controller targets real-time regulation in balanced distribution networks under impedance uncertainty, with a pre-calculable, uniformly bounded recovery time.It uses a lossy linearization that analytically approximates network losses rather than traditional lossless formulations,,,,,.Source: §I · p. 1
  • Fixed-time stability and control Lyapunov functions derive gain and design-parameter bounds, while quadratic programming provides feasible control set-points.The framework also accounts for impedance uncertainty and reactive-power limits while optimizing transient performance and guaranteeing closed-loop stability.Source: §I · p. 2

II. FOUNDATIONAL FRAMEWORKS

The section introduces the lossy distribution-flow model and formulates voltage regulation as a fixed-recovery-time problem.

II. FOUNDATIONAL FRAMEWORKS

  • Section II-A reviews the linearized lossy distribution-flow model, then Section II-B formulates voltage regulation with a fixed recovery time.The model is referenced to.Source: §II · p. 2

A. Linear approximation of branch flow model

The paper represents radial distribution networks as trees and uses a linearized lossy flow model that generalizes the lossless model. Accounting for losses is presented as improving accuracy across loading levels.

A. Linear approximation of branch flow model

  • A radial distribution network is represented as a tree with the substation as root and known, fixed root voltage.The edge set represents feeder lines, and removing the root-node column yields a non-singular reduced incidence matrix.Source: A. Linear approximation of branch flow… · p. 2
  • The linearized voltage approximation follows a lossy distribution-flow model and includes resistance, reactance, and a voltage-sensitivity parameter.The sensitivity parameter approximates feeder-line losses using endpoint voltages expressed through squared magnitudes.Source: A. Linear approximation of branch flow… · p. 2
  • Setting σ_ij = 0.5 on every feeder line recovers the standard lossless model as a special case, so subsequent results also apply to lossless networks.The lossy formulation is therefore presented as a generalized linear approximation framework,,,.Source: A. Linear approximation of branch flow… · p. 2
  • For radial networks with positive line impedances, the lossless model overestimates voltage magnitudes and underestimates power flows.Its accuracy also degrades outside a tightly bounded, lightly loaded operating region, whereas accounting for losses yields high state-estimation accuracy across varying loading levels,.Source: A. Linear approximation of branch flow… · p. 2

B. Problem setup

The setup casts voltage regulation as a fixed-time optimal-control problem with bounded reactive-power rate inputs. It uses continuous-time dynamics with unknown real-time external injections and allows the formulation to extend to networks with only some controllable nodes.

B. Problem setup

  • The objective jointly minimizes voltage-violation and reactive-power costs by a predefined fixed time, with coefficients balancing the two costs.The optimization is subject to the voltage dynamics and safe-voltage limits.Source: B. Problem setup · p. 2
  • The decision variable is the rate of reactive-power injection, bounded by the compact set Xu = [−umax, umax].The continuous-time voltage dynamics use a drift matrix built from network resistance and reactance, while vext(t) represents uncontrollable active injections and substation voltage.Source: B. Problem setup · p. 2
  • The external-injection term vext(t) is assumed unknown in real time, precluding feedforward optimization for computing control set-points.The goal is to return network voltages to the safe set at the specified fixed time after a disturbance.Source: B. Problem setup · p. 2
  • Although the setup initially presumes controllers at every node, the authors state that the formulation also applies when only a subset of nodes is controllable.They partition nodes into controllable and uncontrollable subsets and state that the resulting dynamics retain the relevant structure for applying the fixed-time algorithm.Source: B. Problem setup · pp. 2–3

III. CHARACTERISTICS AND BEHAVIOUR OF FIXED-TIME VOLTAGE REGULATORS

The voltage-regulation dynamics are reformulated using voltage error without requiring exact knowledge of the network matrix B. The analysis treats impedance mismatch and external disturbances as perturbations subject to stated bounds.

III. CHARACTERISTICS AND BEHAVIOUR OF FIXED-TIME VOLTAGE REGULATORS

  • The voltage violation vector e(t) = v(t) − v_nom yields a continuous-time model that can be analyzed without knowing B precisely.B depends on network resistance and reactance, which may be unavailable or outdated.Source: §III · p. 3
  • The model represents the difference between estimated and actual impedance effects as a vanishing perturbation χ(·).B̂ denotes the estimate of B.Source: §III · p. 3
  • The analysis assumes persistent disturbance w(t) is bounded by γ and momentary disturbance d(t) by a known d_max.The momentary disturbance is treated as an impulsive transient; the text relates the persistent-disturbance bound to active bus power not varying abruptly.Source: §III · p. 3
  • The stated disturbance bounds constrain the setting in which the impedance-unaware regulation analysis applies.The text says bounded impedance estimates are plausible in existing algorithms [2],.Source: §III · p. 3

A. Stability Studies of Fixed-Time Regulation Paradigm

The stability analysis gives conditions for steering voltage error to a goal set in fixed time, then relates that set to the safe voltage range. Its convergence bound is independent of the initial disturbance magnitude, but depends on disturbance and impedance-estimation bounds.

A. Stability Studies of Fixed-Time Regulation Paradigm

  • An FxTS-CLF is a positive function h on compact Z with a decay condition using two powers of h and positive gains α1, α2.The resulting convergence time is bounded by t_FX as specified in the definition.Source: A. Stability Studies of Fixed-Time… · p. 3
  • With min{α1, α2} > ξδβ2√β1 and η ∈ (1, 2], Theorem 1 guarantees fixed-time convergence to a goal set S_G of diameter below 1.The guarantee applies when the stated assumptions hold and the initial voltage lies outside X_v.Source: A. Stability Studies of Fixed-Time… · pp. 3–4
  • Theorem 1 uses a two-stage mechanism: one exponent drives error toward S, then the other guides it into S_G.S is defined by h(e) ≤ 1; the exponents dominate in the respective outside- and inside-S regions.Source: A. Stability Studies of Fixed-Time… · p. 4
  • The convergence bound t_FX does not depend on initial disturbance magnitude, but depends on γ, impedance-estimate bound δ, and Lipschitz bound ξ.Corollary 1 bounds convergence time by t_FX.Source: A. Stability Studies of Fixed-Time… · p. 4
  • Maintaining diam(S_G) < 1 requires α̌ to dominate the stated disturbance-related terms, potentially demanding a very large gain.The text says α̌ must be arbitrarily large to restrict the goal-set diameter below 1.Source: A. Stability Studies of Fixed-Time… · p. 4
  • Lemma 1 connects the voltage-error set S_v to the safe voltage range X_v: membership in S_v implies v(t) ∈ X_v.The analysis then bounds the gain α̌ so S_G ⊆ S_v, yielding the chain S_G ⊆ S_v ⊆ X_v and fixed-time voltage regulation.Source: A. Stability Studies of Fixed-Time… · pp. 4–5

B. Limitation of Fixed-Time Convergence under Bounded Control

Bounded control can interrupt fixed-time convergence: outside the safe operating region, saturation produces a finite-time return whose duration scales with the initial violation, after which fixed-time convergence is restored.

B. Limitation of Fixed-Time Convergence under Bounded Control

  • The first two theorems omit control-input bounds, although practical controllers face apparent-power and thermal limits that can cause saturation.Large, rapid disturbances can then turn fixed-time convergence into finite-time recovery until the violation returns to the safe envelope.Source: B. Limitation of Fixed-Time… · p. 5
  • Ω0 is the safe operating envelope: emax is the largest violation norm for which the fixed-time law stays within the control limit Xu.Theorem 3 defines a saturated update for initial violations outside Ω0.Source: B. Limitation of Fixed-Time… · p. 5
  • If σmin(B)umax > dmax, the saturated update brings any initial violation outside Ω0 into Ω0 within finite settling time.The theorem provides an upper bound on that settling time, but its formula is not included in the passage.Source: B. Limitation of Fixed-Time… · p. 5
  • Outside Ω0, saturation adds a linear temporal penalty proportional to ||e(0)|| until the trajectory re-enters Ω0, where fixed-time convergence is restored.Theorem 3 describes this return as finite-time; the unconstrained fixed-time mechanism resumes upon entry.Source: B. Limitation of Fixed-Time… · p. 5

C. Synthesizing Fixed-Time Control under Input Constraints

The section synthesizes constrained voltage-regulation control by translating the problem into a QP and imposing conditions derived from the fixed-time stability analysis. A slack variable supports feasibility, while the framework also accommodates parameter estimators with bounded error.

C. Synthesizing Fixed-Time Control under Input Constraints

  • The voltage-regulation problem is transformed into an equivalent QP that combines FxTS-CLF conditions with input constraints to synthesize control actions.The formulation solves pointwise for u(t) to regulate voltage toward Xv.Source: C. Synthesizing Fixed-Time Control… · pp. 5–6
  • The design selects µ1 and µ2 and requires min {α1, α2} ≥ β2 for fixed-time voltage stability without affecting QP feasibility.The bounded set Xu restricts the control inputs throughout the optimization window.Source: C. Synthesizing Fixed-Time Control… · p. 6
  • A slack variable ν preserves feasibility when setting ν = 0 could make the input-constrained QP infeasible; Theorem 4 states feasibility for every v(t) outside Xv.For each such voltage, the theorem guarantees a pair (u(t), ν) satisfying the stated constraints.Source: C. Synthesizing Fixed-Time Control… · p. 6
  • The control framework can incorporate offline or online estimators that provide bounded parameter-estimation error.The cited offline least-squares approach remains sensitive to data quality and quantity; the online convex-body-chasing approach is more robust to modeling errors and adversarial noise but has lower optimal performance in large-scale systems.Source: C. Synthesizing Fixed-Time Control… · p. 6

IV. NUMERICAL CASE STUDIES

The IEEE-33 bus case studies test voltage regulation under time-varying PV and load profiles, then assess performance against baselines and robustness to uncertainty. The FxTS-CLF approach stabilizes voltages where the baselines do not and maintains safe voltages under the tested impedance and measurement uncertainties.

Network setup

  • The test uses the standard IEEE 33-bus network over 48 hours, with PV and load profiles sampled every 15 minutes.Four 4 MW PV panels are placed at buses 2, 5, 12, and 18.Source: §IV · p. 6, Figure 3 · p. 6
  • PV integration produces both over- and under-voltage violations in the uncontrolled network, motivating the regulation test.The plotted uncontrolled voltages are for buses 11, 16, 24, and 32.Source: §IV · p. 6, Figure 4 · p. 6

Regulation with time-varying load and PV generation

  • Across design-parameter values, FxTS-CLF has significantly faster bus-16 voltage recovery than the baselines, and only the proposed approach stabilizes network voltages.All methods respect reactive power limits; practical convergence is faster than the theoretical recovery time from Corollary 1.Source: §IV · p. 7, Figure 5 · p. 7, Table I · p. 7

Robustness verification

  • Under doubled feeder resistance and reactance and 10% voltage-measurement noise, the proposed algorithm keeps voltages within safe limits while the baselines fail.The uncertainty test uses Gaussian measurement noise with a 10% standard deviation.Source: §IV · p. 7, Figure 6 · p. 7

V. CONCLUDING REMARKS

The conclusion presents an impedance-cognizant fixed-time voltage controller for radial, balanced distribution networks. It derives stability conditions using FxTS and CLF, translates them into a QP, and reports IEEE-33 bus validation under uncertainties, with a finite-time reaching phase when saturation occurs.

Concluding remarks

  • FxTS and CLF analysis yields bounds on control gains and design parameters, which are translated into a QP to find control set points.The stated setting is real-time voltage regulation in radial, balanced distribution networks under external disturbances.Source: §V · p. 7
  • When initial voltage violations exceed safe boundaries, controller saturation temporarily changes fixed-time convergence to finite-time reaching, with a linear penalty proportional to violation magnitude.Unconstrained FxTS convergence recovers after the voltage returns to the safe operating envelope.Source: §V · p. 7
  • IEEE-33 bus validation confirms voltage stability against measurement and modelling uncertainties.Source: §V · p. 7
Loading 2610.01449v1…