\documentstyle{SibMatJ} % \TestXML \topmatter \Author Zhang \Initial F. \Sign Fuhong Zhang \Email 324070108113\@stu.suse.edu.cn \AffilRef 1 \endAuthor \Author Liu \Initial X. \Sign Xiaolan Liu \Email xiaolanliu\@suse.edu.cn \AffilRef 1 \AffilRef 2 \AffilRef 3 \Corresponding \endAuthor \Affil 1 \Division College of Mathematics and Statistics \Organization Sichuan University of Science and Engineering \City Zigong \Country China \endAffil \Affil 2 \Organization Key Laboratory of Higher Education of Sichuan Province for Enterprise Informationalization and Internet of Things \City Zigong \Country China \endAffil \Affil 3 \Organization South Sichuan Center for Applied Mathematics \City Zigong \Country China \endAffil \datesubmitted October 22, 2025\enddatesubmitted \daterevised January 26, 2026\enddaterevised \dateaccepted February 21, 2026\enddateaccepted \UDclass ??? %34K34, 93D09, 26A33 \endUDclass \thanks This work is supported by the National Natural Science Foundation of China (grant 11872043), Natural Science Foundation of Sichuan Province (grant 2023NSFSC1299), the Scientific Research and Innovation Team Program of Sichuan University of Science and Engineering (grant SUSE652B002), Opening Fund of Key Laboratory of Higher Education of Sichuan Province for Enterprise Informationalization and Internet of Things (grant 2023WZJ02). \endthanks \title Robust Mittag-Leffler Stability of Caputo Fractional Differential Equations with State-Dependent Delay and Perturbations \endtitle \abstract In this paper, we investigate the locally robust Mittag-Leffler stability for a class of Caputo fractional-order differential equations subject to state-dependent delays and perturbations. By~constructing an appropriate Lyapunov functional and employing key lemmas in fractional calculus, we derive a novel stability criterion under a~weakened linear matrix inequality (LMI) %\? condition. This criterion guarantees not only the uniqueness of the equilibrium point but also its locally robust stability against perturbations. Finally, two numerical examples are provided to validate the correctness and effectiveness of the main theoretical results. \endabstract \keywords state-dependent delays, parameter uncertainties, Lyapunov functional, LMI \endkeywords \endtopmatter \input epsf \input gutable \head 1. Introduction \endhead The integer-order differential equations have been extended to fractional-order differential equations. Fractional calculus has become a powerful tool for characterizing complex systems with memory and hereditary properties [1,\,2]. Fractional differential equations with delay combine the characteristics of fractional calculus with delay differential equations, and can simultaneously describe the dependence of the system on historical states and the influence of time lags [3]. Thus, they have been widely applied in modeling complex dynamic systems such as population biology [4], physiology [5], control theory [6,\,7], population dynamics [8], and economics [9]. As the application of fractional differential equations with delay in practical problems has become increasingly widespread, the analysis of their stability, as a prerequisite for the reliable operation and control design of systems, has received extensive attention from the academic community over the past two decades and has become one of the core directions in the theoretical research of fractional-order systems [10--13]. Stability, as a key indicator for measuring the dynamic behavior of a system, determines whether the system can maintain the expected performance under external disturbances or initial deviations. Therefore, relevant research not only has theoretical significance but also provides important guiding value for the design and optimization of practical systems. Among the existing research on the stability of fractional-order systems with delay, systems involving time delays and external disturbances have become a research hotspot in recent years due to their closer proximity to engineering practice. For example Zhang [14] studied the stability of linear fractional differential systems and their corresponding perturbed systems with respect to the Riemann-Liouville derivative and the Caputo derivative. In 2022, He [15] investigated the stability of Hadamard-type equations, the Caputo-Hadamard fractional derivative, and their related systems in both delay-free and delay-involved cases. Wang [16] conducted a stability analysis of fractional-order nonlinear systems with time delays and proposed the definition of Mittag-Leffler stability for time-delay systems for the first time. Tuan [17] proved the linearized asymptotic stability of nonlinear fractional differential equations with time delays. From a practical perspective, time delays arise from the physical structure of the system or information processing procedures and are almost inevitable in fields such as mechanical control, communication networks, and biological metabolism. The existence of time delays may cause the system to oscillate or even become unstable [18,\,19]. External disturbances usually manifest as parameter uncertainties, random noise, or sudden interferences, and these disturbances can also disrupt the steady state of the system and reduce control precision [20,\,21]. Li and Wang [22] studied a class of system models that simultaneously consider the effects of state delay and bounded external disturbances, and comprehensively evaluated the dynamic performance of the system under the combined action of external disturbances and event-triggered mechanisms. The coexistence of time delays and disturbances poses enormous challenges to system stability analysis. In the face of these challenges, robustness research has become the key to breakthroughs [23--28]. Regarding the robustness of fractional-order systems, there have been many important explorations: for example, Li [23] studied fractional-order interval nonlinear time-delay systems, expanded the Lyapunov direct method by using new lemmas, and constructed delay-independent stability criteria in the form of linear matrix inequalities by Razumikhin theorem. %\?артикль Zhu [24] focused on the results of uncertain impulsive positive systems with time delays, utilized the impulsive time-varying copositive Lyapunov function, established globally robust exponential stability criteria under different dwell times, and designed controllers. Liao [25] constructed an asymptotic stability test criterion for parameter-uncertain FO-LTI (Fractional-Order Linear Time-Invariant) %\?(fractional-order linear time-invariant) interval systems with deterministic linear coupling relations through the linear matrix inequality (LMI) method. %\? These achievements have enriched the robustness theory of fractional-order systems and laid an important foundation for dealing with complex uncertainties in practical systems. The remainder of this paper is organized as follows. In \Sec*{Section 2}, we list essential preliminaries, including definitions of fractional calculus, key assumptions, and necessary lemmas. The main results on robust Mittag-Leffler stability are presented in \Sec*{Section 3}. In \Sec*{Section 4}, we provide two numerical examples to verify the theoretical findings. Finally, In \Sec*{Section 5}, we conclude the paper and suggests potential future research directions. \head 2. Preliminaries \endhead The stability analysis of fractional-order systems with state-dependent delays and perturbations requires a foundational understanding of fractional calculus, Mittag-Leffler functions, and robust control theory. In this section, we introduce the necessary mathematical tools, including definitions of Caputo derivatives, Mittag-Leffler stability, and key assumptions on system dynamics. Additionally, critical lemmas for bounding fractional Lyapunov derivatives and handling perturbations are presented to support the proof of \Par*{Theorem 3.1} in \Sec*{Section 3}. Next we provide the definitions of the Caputo fractional derivative and Mittag-Leffler stability. \demo{Definition 1 \rm [29]} The Caputo fractional derivative of order $\alpha \in (0,1)$ for a sufficiently smooth function $x(t)\in C^1([t_0, +\infty), \Bbb{R}^n) $ is defined as $$ {^C D_t^\alpha x(t)} = \frac{1}{\Gamma(1 - \alpha)} \int\limits_{t_0}^t \frac{\dot{x}(s)}{(t - s)^\alpha} ds, $$ where $\dot{x}(s) = \frac{d x(s)}{d s}$, $\Gamma(\cdot)$ is the Gamma function. \enddemo \demo{Definition 2 \rm [16]} A fractional-order system $ {}_{t_0}^{C}D_t^\alpha x(t) = f(t, x_t) $ is said to be Mittag-Leffler stable if there exist $\lambda \geq 0$ and $b > 0$ such that $$ \|x(t,t_0,\varphi)\| \leq \bigl\{m(\varphi) E_\alpha(-\lambda(t-t_0)^\alpha)\bigr\}^b, $$ where $\alpha \in (0,1)$, $b > 0$, $m(0) = 0$, $m(x) \geq 0$, and $m(x)$ is locally Lipschitz on $x \in \Cal{E}$ with the Lipschitz constant $m_0$. \enddemo In this paper, we focus on a class of fractional-order dynamical systems with state-dependent time delays and perturbations, which is formulated as \Tag(1). $$ \cases {}^{C}D^{\alpha}_{t}x(t) = f\bigl(t, x(t), x(t - \tau(x(t))), \Delta A(t), \Delta f(t)\bigr), & t \geq 0 \\ x(t) = \phi(t), & t \in [-\tau_{\max}, 0] \endcases \tag1 $$ where $ ^C D_t^\alpha x(t) = \frac{1}{\Gamma(1 - \alpha)} \int\nolimits_{t_0}^t \frac{\dot{x}(s)}{(t - s)^\alpha} ds $, the state-dependent time delay $ \tau(x(t)) \in [0, \tau_{\max}] $ satisfies $ \dot{\tau}(x(t)) \leq \eta < 1 $. Besides, the perturbation bounds satisfy $ \|\Delta A(t)\| \leq \rho $, and $ \|\Delta f(t)\| \leq d $. This system incorporates several key characteristics that are prevalent in many real-world applications, such as biological networks, chemical processes, and control systems with memory effects. In order to obtain our results, we make some assumptions\Label{HH} (H1) %\? The delay $ \tau(x(t)) $ satisfies $ 0 \leq \tau(x(t)) \leq \tau_{\max} $, and $ \dot{\tau}(x(t)) \leq \eta < 1 $, where $\tau_{\max}, \eta$ are known constants. It avoids the disorder of the system's memory effect caused by the rapid fluctuation of time delay with the state, which is a necessary condition for the existence of a unique stable solution in fractional-order systems with state-dependent delays. If $\eta \geq 1$, the rapid fluctuation of time delay with the state will violate the causality of the system, leading to oscillations or multiple solutions of the solution. This constraint is also consistent with the classical assumptions in the field of fractional-order time-delay systems. (H2) The perturbations $ \Delta A(t) $ and $ \Delta f(t) $ are bounded such that $$ \|\Delta A(t)\| \leq \rho, \quad \|\Delta f(t)\| \leq d, $$ where $ \rho, d > 0 $ are known constants. If the perturbations are unbounded, the system state will diverge directly due to the accumulation of uncertainties, making stability analysis meaningless. Meanwhile, the boundedness conditions provide a quantitative basis for bounding the perturbation terms using Young's inequality and matrix norm inequalities in subsequent derivations. (H3) The function $f\bigl(t, x(t), x\bigl(t - \tau(x(t))\bigr), \Delta A(t), \Delta f(t)\bigr)$ in system \Tag(1) is of linear form with respect to the state variables $x$ and $x(t-\tau(x(t)))$, i.e., $$ f\bigl(t, x(t), x\bigl(t - \tau(x(t))\bigr), \Delta A(t), \Delta f(t)\bigr) = \bigl( A + \Delta A(t) \bigr) x(t) + B x\bigl(t - \tau(x(t)) \bigr) + \Delta f(t), $$ where $A, B \in \Bbb{R}^{n\times n}$ are constant matrices, and $\Delta A(t), \Delta f(t)$ satisfy the boundedness conditions in \Par{HH}{(H2)}. (H4) When the initial condition satisfies $\|x(t_0)\| \leq R_0$ ($R_0 \leq R$), the state trajectory of the system remains within the neighborhood $\Omega = \{x \in \Bbb{R}^n \mid \|x\| \leq R\}$ at all times, where $R>0$ is a given constant. Regarding the well-posedness of system \Tag(1), we emphasize that under assumptions \Par{HH}{(H1)}--\Par{HH}{(H3)} and the continuity of the nonlinear function $f(t, x(t), x(t-\tau(x(t))), \Delta A(t), \Delta f(t))$ with respect to all its arguments, the system admits a unique solution on $[0, +\infty)$ by the existence and uniqueness theorem for fractional differential equations with state-dependent delays. Furthermore, as rigorously proved in Vainikko [30], for the Caputo fractional derivative of order $\alpha \in (0,1)$ adopted in this paper, the solution possesses a specific structure: it is continuous on $[0, +\infty)$, its Caputo derivative exists almost everywhere, and the growth behavior of the solution is dominated by the $t^\alpha$ term (directly related to the initial value $x(0)$). This structural characteristic ensures that the subsequent stability analysis based on Lyapunov-Krasovskii functionals has a rigorous theoretical foundation. \demo{Definition 3 \rm [31]} A fractional-order system with bounded perturbations is said to be robustly Mittag-Leffler stable if, for all disturbances satisfying $\|\Delta f(t)\| \leq d$ and $\|\Delta A(t)\| \leq \rho$, its state satisfies $$ \|x(t)\| \leq M \|x(0)\| E_\alpha (-\lambda t^\alpha) + N d, $$ where $M, N > 0$ are constants independent of the initial conditions and disturbances. \enddemo Next we provide some necessary lemmas to facilitate the establishment of our main results. \proclaim{Lemma 1 \rm [32]} For a Caputo fractional derivative of order $\alpha \in (0,1)$, consider a continuously differentiable function $x(t) \in \Bbb{R}^n$ and a positive definite matrix $P \in \Bbb{R}^{n \times n}$. The Caputo fractional derivative of the quadratic Lyapunov candidate $V(t) = x^\top(t)Px(t)$ satisfies $$ {^C}D_t^\alpha V(t) \leq 2x^\top(t) P {^C}D_t^\alpha x(t). $$ \endproclaim \proclaim{Lemma 2 \rm [33]} Let $ X $ and $ Y $ be matrices of appropriate dimensions. Then, the following inequality holds, for all $ \varepsilon > 0 $ $$ X^\top Y+Y^\top X \leqslant \varepsilon X^\top X + \varepsilon^{-1} Y^\top Y. $$ \endproclaim \proclaim{Lemma 3 \rm [34]} Let $ u(t) $ be a continuous function that satisfies, for $ t \geq t_0 $ and $ 0 < \alpha < 1 $, the following inequality: $$ u(t) \leq a + \frac{1}{\Gamma(\alpha)} \int\limits_{t_0}^{t} (t - s)^{\alpha - 1} [b u(s) + c] \, ds, $$ where $a$, $b$, and $c$ are non-negative %\? constants with $ b > 0 $. Then, the solution of the above inequality is bounded by $$ u(t) \leq a E_{\alpha} ( -b (t - t_0)^{\alpha} ) + \frac{c}{b} \bigl( 1 - E_{\alpha} ( -b (t - t_0)^{\alpha} ) \bigr). $$ \endproclaim Next we introduce a special symbol that appears in the proof of a theorem. $M \prec 0$ means that all eigenvalues of $M$ are strictly less than zero. This notation is widely used in the analysis of linear matrix inequalities (LMI) %\?уже было сокращение к ед.ч., причем не везде используется сокращение to characterize the negative definiteness of matrices, which plays a crucial role in the stability analysis of dynamical systems. \head 3. Main Result %\? \endhead In this section, through a refined Lyapunov functional approach, we rigorously prove that \Tag(1) admits a unique equilibrium point exhibiting local robust Mittag-Leffler stability under this weakened constraint. \proclaim{Theorem 3.1} Consider system \Tag(1) under assumptions \Par{HH}{\rm(H1)}--\Par{HH}{\rm(H3)}. If there exist positive definite matrices $P$, $Q$ and parameters $\varepsilon, \sigma, \nu, \lambda, \mu > 0$ such that the following linear matrix inequalities hold: Current state LMI %\? $$ 2PA + (\varepsilon + \sigma)P^\top P + (\varepsilon^{-1}\rho^2+\nu)I + Q + \lambda I \prec 0, $$ Delayed state LMI %\? $$ \nu^{-1}B^{\top}P^{\top}PB - (1-\eta)Q + \mu I \prec 0, $$ then \Tag(1) is local robust Mittag-Leffler stable, and the state satisfies $$ \|x(t)\| \leq C_0E_{\alpha}(-m (t - t_0)^{\alpha}) + C_1 \left( 1 - E_{\alpha}(-m (t - t_0)^{\alpha}) \right) , $$ where $ m = \min \bigl\{ \frac{\lambda}{\lambda_{\max}(P)}, \frac{\mu}{\lambda_{\max}(Q)} \bigr\} $, $C_0=\sqrt{\frac{\lambda_{\max}(P)+\lambda_{\max}(Q)}{\lambda_{\min }(P)}}\| x(t_0) \|$, $C_1=\sqrt{\frac{M}{m \lambda_{\min }(P)} }$, and $ M = \sigma^{-1} d^2 + \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2 $. \endproclaim \demo{Proof} Considering the state-dependent time delay $\tau(x(t))$ which makes the system's memory effect related to state evolution, and the external disturbance $\Delta f(t)$ that brings uncertain perturbations, we construct the Lyapunov function $V(t)$ as follows. The state quadratic term $$ V_1(t) = x^\top(t) P x(t), $$ directly reflects the energy of the current system state. The time-delay integral term $$ V_2(t) = \int\limits_{t-\tau(x(t))}^t x^\top(s) Q x(s) ds, $$ by integrating the state energy over the time-delay interval, it captures the cumulative influence of the time-delay history on the current state; since the time delay is state-dependent, the integration interval varies dynamically with the state. The disturbance norm integral term $$ V_3(t) = \int\limits_{t_0}^t \left( \|\Delta A(s)\|^2 + \|\Delta f(s)\|^2 \right) ds, $$ for bounded disturbances, by summing the norms of disturbance-related terms, the uncertainties are transformed into quantifiable accumulated energy of disturbances, thereby enabling a comprehensive analysis of system stability with these factors taken into account. $P, Q > 0$ are the positive definite matrix. Compared with existing literature, the functional proposed in this paper simultaneously includes the state term, time-delay integral term, and disturbance integral term. It does not require the introduction of additional complex nonlinear weighting functions, is more easily solvable via LMI, and is thus suitable for engineering application scenarios. Based on assumption \Par{HH}{(H3)} on the structure of $f(\cdot, \cdot, \cdot)$, we can rewrite the original system \Tag(1) as the following linear fractional-order system with state-dependent delay and perturbations $$ {^C D_t^\alpha} x(t) = \bigl( A + \Delta A(t) \bigr) x(t) + B x\bigl(t - \tau(x(t)) \bigr) + \Delta f(t). \tag2 $$ This linearization is valid under the boundedness of perturbations \Par{HH}{(H2)} and the state-dependent delay constraint \Par{HH}{(H1)}, which lays the foundation for the subsequent LMI-based stability analysis. Next, we analyze the derivative of the Lyapunov function $ V_1(t) $ by leveraging the properties of the Caputo fractional derivative and integration by parts, we derive $$ {^C D_t^\alpha} V_1(t) = \frac{1}{\Gamma(1 - \alpha)} \int\limits_{t_0}^t \frac{x^\top(s) P \dot{x}(s) + \dot{x}^\top(s) P x(s)}{(t - s)^\alpha} \, ds. $$ By the fractional-order inequality \Par*{Lemma 1}, we simplify and obtain a key inequality $$ {^C D_t^\alpha} V_1(t) \leq 2 x^\top(t) P \, {^C D_t^\alpha} x(t). \tag3 $$ Combining the system dynamics \Tag(2) with the Lyapunov derivative inequality \Tag(3), we derive $$ {^C D_t^\alpha} V_1(t) \leq 2 x^\top(t) P \bigl[ ( A + \Delta A(t) ) x(t) + B x(t - \tau(x(t)) ) + \Delta f(t) \bigr] . \tag4 $$ By \Par*{Lemma 2}, for sufficiently small $ \varepsilon > 0 $, we bound the uncertainty terms $$ \aligned 2x^\top(t) P \Delta A(t) x(t) &=x^\top(t) P \Delta A(t) x(t)+x^\top(t)\Delta A^\top(t)P^\top x(t), %\?, нужны везде \\ &\leq \varepsilon x^\top(t) P P^\top x(t) + \varepsilon^{-1} \left[ \Delta A(t) x(t) \right]^\top \Delta A(t) x(t), \\ &= \varepsilon x^\top(t) P P^\top x(t) + \varepsilon^{-1} x^\top(t) \Delta A^\top(t) \Delta A(t) x(t), \\ &\leq \varepsilon x^\top(t) P^\top P x(t) + \varepsilon^{-1} \rho^2 x^\top(t) x(t). \endaligned \tag5 $$ For the external disturbance term, there exists $\sigma > 0$ such that $$ \aligned 2x^\top(t) P \Delta f(t)&=x^\top(t) P \Delta f(t)+ \Delta f^\top(t)P^\top x(t),\\ &\leq \sigma x^\top(t) P P^\top x(t) + \sigma^{-1} \Delta f^\top(t) \Delta f(t), \\ &= \sigma x^\top(t) P^\top P x(t) + \sigma^{-1} \Delta f^\top(t) \Delta f(t), \\ &\leq \sigma x^\top(t) P^\top P x(t) + \sigma^{-1} d^2. \endaligned \tag6 $$ Combining \Tag(4) and \Tag(5) with \Tag(6), we have $$ \aligned ^C D_t^\alpha V_1(t) &\leq 2x^\top(t) P A(t) x(t) + 2x^\top(t) P B x\bigl(t - \tau(x(t))\bigr) \\ &\qquad+ 2x^\top(t) P \Delta A(t) x(t) + 2x^\top(t) P \Delta f(t), \\ &\leq 2x^\top(t) P A(t) x(t) + 2x^\top(t) P B x\bigl(t - \tau(x(t))\bigr) \\ &\qquad+ \varepsilon x^\top(t) P^\top P x(t) + \varepsilon^{-1} \rho^2 x^\top(t) x(t) + \sigma x^\top(t) P^\top P x(t) + \sigma^{-1} d^2. \endaligned \tag7 $$ To comprehensively obtain the Caputo fractional derivative of the entire Lyapunov function $V(t)$, after analyzing the derivative of $V_1(t)$, we now proceed to analyze the second term $V_2(t)$. Utilizing properties of Caputo fractional derivatives for integral terms, we compute $^C D_t^\alpha V_2(t)$. Assume $\dot{\tau}(x(s)) \leq \eta < 1$, since the time delay $\tau(x(t))$ is state-dependent, its derivative $\dot{\tau}(x(t))$ reflects the rate of change of the delay with respect to the state. The condition $\dot{\tau}(x(s)) \leq \eta < 1$ ensures that the delay does not change too rapidly, which is a common and reasonable assumption to guarantee the solvability and stability analysis feasibility of the system with state-dependent delays. It restricts the delay's variation rate to be within a range that our subsequent fractional-order derivative analysis can handle. Then $$ \aligned {^C D_t^\alpha} V_2(t) &= \frac{1}{\Gamma(1 - \alpha)} \int\limits_{t_0}^t \frac{\dot{V}_2(s)}{(t - s)^\alpha} \, ds ,\\ &= \frac{1}{\Gamma(1 - \alpha)} \int\limits_{t_0}^t \frac{x^\top(s) Q x(s) - x^\top\bigl(s - \tau(x(s))\bigr) Q x\bigl(s - \tau(x(s))\bigr) \cdot \bigl(1 - \dot{\tau}(x(s))\bigr)}{(t - s)^\alpha} \, ds, \\ &\leq \frac{1}{\Gamma(1 - \alpha)} \int\limits_{t_0}^t \frac{x^\top(s) Q x(s) - (1 - \eta) x^\top\bigl(s - \tau(x(s))\bigr) Q x\bigl(s - \tau(x(s))\bigr)}{(t - s)^\alpha} \, ds . \endaligned $$ For constant or slowly-varying delays, when $t\approx s$, $x(t)\approx x(s)$, $\tau(x(t)) \approx \tau(x(s)) $, this simplifies to a key inequality $$ \aligned ^C D_t^\alpha V_2(t) &\leq x^\top(t) Q x(t) - (1 - \eta) x^\top\bigl(t - \tau(x(t))\bigr) Q x\bigl(t - \tau(x(t))\bigr) \frac{1}{\Gamma(1 - \alpha)}\int\limits_{t_0}^t\frac{1}{(t - s)^\alpha} \, ds,\\ &=x^\top(t) Q x(t) - (1 - \eta) x^\top\bigl(t - \tau(x(t))\bigr) Q x\bigl(t - \tau(x(t))\bigr)\frac{(t - t_0)^{1 - \alpha}}{\Gamma(1 - \alpha)(1 - \alpha)},\\ &\leq x^\top(t) Q x(t) - (1 - \eta) x^\top\bigl(t - \tau(x(t))\bigr) Q x\bigl(t - \tau(x(t))\bigr). \endaligned \tag8 $$ Having analyzed the derivatives of $ V_1(t) $ and $ V_2(t) $, which account for the state energy and time-delay energy respectively, we now move on to the final term $ V_3(t) $ of the Lyapunov function. Analyzing $ V_3(t) $ will help us capture the influence of disturbances on the system's energy from the perspective of fractional-order calculus, so as to obtain the complete derivative of $ V(t) $. For the Caputo fractional derivative $^C D_t^\alpha V_3(t)$, we use properties of fractional calculus $$ \align ^C D_t^\alpha V_3(t) &= \frac{1}{\Gamma(1 - \alpha)} \int\limits_{t_0}^t \frac{\dot{V}_3(s)}{(t - s)^\alpha} \, ds , \\ &= \frac{1}{\Gamma(1 - \alpha)} \int\limits_{t_0}^t \frac{\| \Delta A(s) \|^2 + \| \Delta f(s) \|^2}{(t - s)^\alpha} \, ds,\\ & \leq \frac{1}{\Gamma(1 - \alpha)} \bigl( \sup_{\xi \in [t_0, t]} [ \|\Delta A(\xi)\|^2 + \|\Delta f(\xi)\|^2 ] \bigr) \int\limits_{t_0}^t \frac{1}{(t - s)^\alpha} ds,\\ &= \frac{(t - t_0)^{1 - \alpha}}{\Gamma(1 - \alpha)(1 - \alpha)} \cdot \bigl( \sup_{\xi \in [t_0, t]} \bigl[ \|\Delta A(\xi)\|^2 + \|\Delta f(\xi)\|^2 \bigr] \bigr) . \endalign $$ Considering the singularity of the fractional-order integral kernel $(t - s)^{-\alpha}$, its value is mainly dominated by the behavior of the integrand in the neighborhood near the upper limit $t$ of the integral. Therefore, the global supremum $\sup_{\xi \in [t_0, t]} (\cdot)$ can be replaced by the instantaneous value $\|\Delta A(t)\|^2 + \|\Delta f(t)\|^2$ at the current moment, forming a simpler but conservative upper bound. At the same time, precisely because the factor $\frac{(t - t_0)^{1 - \alpha}}{\Gamma(1 - \alpha)(1 - \alpha)}$ does not affect the direction of the fractional-order derivative inequality. In order to obtain a form consistent with the integer-order theory, this factor will be absorbed into the design of Lyapunov function coefficients or control gains in subsequent stability analysis, and its specific value does not affect the establishment of the stability conclusion. Based on the above processing, we obtain the simplified form for the derivation of the stability criterion $$ \aligned ^C D_t^\alpha V_3(t) &\leq \left( \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2 \right) \frac{(t - t_0)^{1 - \alpha}}{\Gamma(1 - \alpha)(1 - \alpha)} , \\ &\leq \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2. \endaligned \tag9 $$ Combining \Tag(7) and \Tag(8), with \Tag(9), we derive the final Lyapunov derivative bound $$ \aligned ^C D_t^\alpha V(t) &\leq 2x^\top(t) P A(t) x(t) + 2x^\top(t) P B x\bigl(t - \tau(x(t))\bigr)+\varepsilon x^\top(t) P^\top P x(t)\\ &\qquad+ \varepsilon^{-1} \rho^2 x^\top(t) x(t) + \sigma x^\top(t) P^\top P x(t) + \sigma^{-1} d^2 +x^\top(t) Q x(t) \\ &\qquad- (1 - \eta) x^\top\bigl(t - \tau(x(t))\bigr) Q x\bigl(t - \tau(x(t))\bigr)+ \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2. \endaligned \tag10 $$ By LMI, we transform the quadratic form in $ {^C D_t^\alpha} V(t) $ into a negative definite form. The key derivations are as follows. Let $z(t)= x(t - \tau(x(t)))$. Then for the cross term $2x^\top(t) P B z(t)$, we apply Young's inequality $$ 2x^\top(t) P B z(t) \leq \nu x^\top(t) x(t) + \nu^{-1} z^\top(t) B^\top P^\top P B z(t), \tag11 $$ where $\nu> 0$ is a tuning parameter. Substituting \Tag(11) into \Tag(10), we have $$ \aligned ^C D_t^\alpha V(t) &\leq 2x^\top(t) P A(t) x(t) + \nu x^\top(t) x(t) + \nu^{-1} z^\top(t) B^\top P^\top P B z(t)\\ &\qquad+\varepsilon x^\top(t) P^\top P x(t)+ \varepsilon^{-1} \rho^2 x^\top(t) x(t) + \sigma x^\top(t) P^\top P x(t)\\ &\qquad+ \sigma^{-1} d^2 +x^\top(t) Q x(t)- (1 - \eta) z^\top(t) Q z(t)+ \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2. \endaligned \tag12 $$ To achieve the stability of the current state, we need to verify $$ 2x^\top(t) P A(t) x(t) + (\varepsilon + \sigma)x^\top(t) P^\top P x(t) + (\varepsilon^{-1}\rho^2+\nu)x^\top(t) x(t) + x^\top(t) Q x(t)\leq -\lambda x^\top(t) x(t), \tag13 $$ that is, $$ x^\top(t) \bigl[ 2PA + (\varepsilon + \sigma)P^\top P + (\varepsilon^{-1}\rho^2+\nu) I + Q + \lambda I \bigr] x(t) \leq 0. \tag14 $$ Since \Tag(14) holds for all nonzero vectors $x(t)$, the matrix must be negative definite. By choosing appropriate parameters via LMI, we have $$ 2PA + (\varepsilon + \sigma)P^\top P + (\varepsilon^{-1}\rho^2+\nu)I + Q + \lambda I \prec 0. $$ To achieve the stability of the delayed state, we need to verify $$ \nu^{-1} z^\top(t) B^\top P^\top P B z(t) - (1 - \eta)z^\top(t) Q z(t) \leq -\mu z^\top(t) z(t), \tag15 $$ that is, $$ z^\top(t) \bigl[ \nu^{-1} B^\top P^\top P B - (1 - \eta) Q + \mu I \bigr] z(t) \leq 0. \tag16 $$ Since \Tag(16) holds for all $x(t), z(t)\neq 0$, the quadratic form of $z(t)$ must be negative definite. That is $$ \nu^{-1} B^\top P^\top P B - (1 - \eta) Q + \mu I \prec 0. $$ Substitute \Tag(13) and \Tag(15) into \Tag(12). We derive the key fractional Lyapunov derivative bound $$ \aligned {^C D_t^\alpha} V(t) &\leq -\lambda x^\top(t) x(t) - \mu z^\top(t) z(t) + \sigma^{-1} d^2 + \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2 \\ &= -\lambda x^\top(t) x(t) - \mu z^\top(t) z(t) + M \\ &\leq -\frac{\lambda}{\lambda_{\max}(P)}V_1(t)-\frac{\mu}{\lambda_{\max}(Q)} V_2(t) + M\\ &\leq -m(V_1(t)+V_2(t))+ M\\ &\leq -mV(t) + M, \endaligned \tag17 $$ where $ m = \min\bigl\{ \frac{\lambda}{\lambda_{\max}(P)}, \frac{\mu}{\lambda_{\max}(Q)} \bigr\} $ and $ M = \sigma^{-1} d^2 + \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2 $ . By \Par*{Lemma 3}, we solve the bound of $ V(t) $. For $ t \geq t_0 $, assume $ V(t_0) $ is the initial value. The solution satisfies $$ \aligned V(t) &\leq V(t_0) E_\alpha( -m(t - t_0)^\alpha ) + \frac{M}{m} ( 1 - E_\alpha( -m(t - t_0)^\alpha ) ),\\ &\leq \Biggl[ x^\top(t_0) P x(t_0) + \int\limits_{t_0 - \tau(x(t_0))}^{t_0} x^\top(s) Q x(s) ds \Biggr] E_\alpha( -m(t - t_0)^\alpha ) \\ &\qquad+ \frac{M}{m} ( 1 - E_\alpha( -m(t - t_0)^\alpha ) ), \endaligned \tag18 $$ where $ E_\alpha(\cdot) $ is the Mittag-Leffler function. By positive definiteness of $ P $, there exists $ \lambda_{\min }(P) > 0 $ such that $$ V(t)\geq x^\top(t) P x(t) \geq \lambda_{\min }(P) \| x(t) \|^2. \tag19 $$ Then $$ \| x(t) \|^2 \leq \frac{1}{\lambda_{\min }(P)} V(t). $$ Taking square roots (for $ \| x(t) \| \geq 0 $), we derive $$ \align \| x(t) \| & \leq \sqrt{\frac{1}{\lambda_{\min }(P)}} \sqrt{V(t)},\\ & \leq \sqrt{\frac{1}{\lambda_{\min }(P)}} \Bigl[ \sqrt{V(t_0) E_\alpha( -m (t - t_0)^\alpha )+\frac{M}{m} ( 1 - E_\alpha( -m (t - t_0)^\alpha ) ) }\Bigr], \\ & \leq \sqrt{\frac{1}{\lambda_{\min }(P)}} \Bigl[ \sqrt{V(t_0) E_\alpha( -m (t - t_0)^\alpha )} + \sqrt{\frac{M}{m} ( 1 - E_\alpha( -m (t - t_0)^\alpha ) )} \Bigr], \\ & \leq \sqrt{\frac{1}{\lambda_{\min }(P)}} \Bigl[ \sqrt{V(t_0)} E_\alpha( -m(t - t_0)^\alpha ) + \sqrt{\frac{M}{m} }( 1 - E_\alpha( -m (t - t_0)^\alpha ) ) \Bigr], \\ & \leq\sqrt{\frac{\lambda_{\max}(P)+\lambda_{\max}(Q)}{\lambda_{\min }(P)}}\bigl[\| x(t_0) \| E_\alpha( -m (t - t_0)^\alpha ) \bigr]\\ &\qquad+\sqrt{\frac{M}{m \lambda_{\min }(P)} }( 1 - E_\alpha( -m (t - t_0)^\alpha ) ), \\ &=C_0 E_\alpha( -m (t - t_0)^\alpha ) + C_1 ( 1 - E_\alpha( -m (t - t_0)^\alpha ) ), \endalign $$ where $ C_0$ and $C_1$ now depend on $\lambda_{\max}(P),\lambda_{\max}(Q), \lambda_{\min}(P) $, and initial value $ \| x(t_0) \| $. \demo{Remark 1} The Lyapunov functional approach adopted in this paper provides a systematic way to analyze the stability of fractional-order systems with state-dependent delays. However, the construction of the functional is somewhat conservative, which may lead to larger stability bounds in the theoretical analysis. The conservatism of the proposed functional mainly stems from two aspects. First, in the computation of the Caputo derivative of $V_3(t)$, the global supremum is approximated by the disturbance bound at the current time instant, neglecting the dynamic variation of historical disturbances. Second, the tuning parameter $v>0$ introduced by Young's inequality may lead to conservative LMI conditions. Future work could focus on developing less conservative functionals, such as those incorporating more detailed information about the delay and system dynamics. \enddemo \proclaim{Corollary 1} For the system considered in \Par*{Theorem {\rm3.1}}, when the fractional order $\alpha \to 1$, the stability criterion in \Par*{Theorem {\rm3.1}} reduces to the well-known exponential stability condition for integer-order systems, indicating that our results are a generalization of the classical theory. \endproclaim \proclaim{Theorem 3.2 \rm (Local Robust Mittag-Leffler Stability for Nonlinear Fractional-Order Systems)} Consider the nonlinear fractional-order system with state-dependent delay $$ {^C}D_t^\alpha x(t) = (A + \Delta A(t, x(t)))x(t) + Bg(x(t - \tau(x(t)))) + \Delta f(t), $$ where the nonlinear function $ g(\cdot) $ satisfies the Lipschitz condition for all $ x, y \in \Omega$ {\rm(}$\Omega = \{x \in \Bbb{R}^n \mid \|x\| \leq R\}${\rm)} there exists a positive constant $L_g$ such that we have $$ \|g(x) - g(y)\| \leq L_g \|x - y\|, $$ and $ g(0) = 0 $. The perturbations $ \Delta A(t, x(t)) $ and $ \Delta f(t) $ are bounded as in \Par{HH}{\rm(H2)}. Under the assumptions \Par{HH}{\rm (H1)}--\Par{HH}{\rm (H4)}, if there exist positive definite matrices $ P$, $Q $ and parameters $ \lambda, \mu, \varepsilon, \sigma, \nu > 0 $ such that the following LMIs hold. Current state LMI $$ 2PA + (\varepsilon + \sigma)P^\top P + \varepsilon^{-1}\rho^2 I + Q + \lambda I \prec 0, $$ Delayed state LMI $$ \nu^{-1}L_g^2 B^\top P^\top PB - (1 - \eta)Q + \mu I \prec 0, $$ then the system is globally robust Mittag-Leffler stable. The state trajectory satisfies the following norm bound $$ \|x(t)\| \leq C_0E_{\alpha}(-m (t - t_0)^{\alpha}) + C_1 ( 1 - E_{\alpha}(-m (t - t_0)^{\alpha}) ) . $$ \endproclaim \demo{Proof} The proof follows the same structure as in \Par*{Theorem 3.1}. Choose a Lyapunov functional similar to that in \Par*{Theorem 3.1} $$ \aligned V(t) &= V_1(t) + V_2(t) + V_3(t),\\ &= x^\top (t)Px(t)+\int\limits_{t - \tau(x(t))}^t x^T(s)Qx(s)ds+\int\limits_{t_0}^t \bigl( \|\Delta A(s, x(s))\|^2 + \|\Delta f(s)\|^2 \bigr) ds. \endaligned \tag20 $$ By the properties of fractional calculus and the system dynamics, we have $$ {^C}D_t^\alpha V_1(t) \leq 2x^\top(t)P\bigl[(A(t) + \Delta A(t, x(t)))x(t) + Bg(x(t - \tau(x(t)))) + \Delta f(t)\bigr] $$ for $ 2x^\top(t)P\Delta A(t, x(t))x(t) $ and $ 2x^\top(t)P\Delta f(t) $ by \Par*{Lemma 2}, we obtain \iftex $$ \align & 2x^T(t)P\Delta A(t, x(t))x(t) \leq \varepsilon x^\top(t)P^\top Px(t) + \varepsilon^{-1}x^\top(t)\Delta A^\top(t, x(t))\Delta A(t, x(t))x(t), \tag21 \\ & 2x^\top(t)P\Delta f(t) \leq \sigma x^\top(t)P^TPx(t) + \sigma^{-1}\|\Delta f(t)\|^2. \tag22 \endalign $$ \else $$ 2x^T(t)P\Delta A(t, x(t))x(t) \leq \varepsilon x^\top(t)P^\top Px(t) + \varepsilon^{-1}x^\top(t)\Delta A^\top(t, x(t))\Delta A(t, x(t))x(t), \tag21 $$ $$ 2x^\top(t)P\Delta f(t) \leq \sigma x^\top(t)P^TPx(t) + \sigma^{-1}\|\Delta f(t)\|^2. \tag22 $$ \fi For the nonlinear term $ 2x^T(t)PBg(x(t - \tau(x(t)))) $, let $ z(t) = x(t - \tau(x(t))) $, and still satisfies $\|z(t)\| \leq R$. Utilizing $ g(0) = 0 $ and the Lipschitz condition $ \|g(z)\| \leq L_g\|z\| $, combining with Young's inequality $$ 2x^\top(t)PBg(z(t)) \leq \nu x^\top(t)x(t) + \nu^{-1}L_g^2z^\top(t)B^\top P^\top PBz(t). \tag23 $$ Combining \Tag(21) and \Tag(22) with \Tag(23), we get $$ \aligned {^C}D_t^\alpha V_1(t) &\leq 2x^\top(t) P A(t)x(t)+\varepsilon x^T(t)P^TPx(t) + \varepsilon^{-1}x^T(t)\Delta A^T(t, x(t))\Delta A(t, x(t))x(t)\\ &\qquad+\sigma x^T(t)P^TPx(t) + \sigma^{-1}\|\Delta f(t)\|^2+\nu x^T(t)x(t) + \nu^{-1}L_g^2z^T(t)B^TP^TPBz(t). \endaligned \tag24 $$ For $ V_2(t) $ and $ V_3(t) $, similar to the derivation in \Par*{Theorem 3.1}, we have $$ \align & {^C}D_t^\alpha V_2(t) \leq x^T(t)Qx(t) - (1 - \eta)z^T(t)Qz(t), \\ & {^C}D_t^\alpha V_3(t) \leq \|\Delta A(t, x(t))\|^2 + \|\Delta f(t)\|^2. \endalign $$ Combine the above derivative terms and organize them with the LMI conditions, we can deduce $$ \aligned {^C}D_t^\alpha V(t) &\leq -\lambda x^T(t)x(t) - \mu z^T(t)z(t) + M,\\ &\leq -mV(t) + M. \endaligned \tag25 $$ Using \Par*{Lemma 3}, we finally obtain that the state norm satisfies $$ \|x(t)\| \leq C_0E_{\alpha}(-m (t - t_0)^{\alpha}) + C_1 \bigl( 1 - E_{\alpha}(-m (t - t_0)^{\alpha}) \bigr). $$ \head 4. Numerical Example %\? \endhead \demo{Example 1} Consider the following fractional-order system with state-dependent delay and disturbances: $$ {^C}D_t^\alpha x(t) = (A + \Delta A(t))x(t) + Bx(t - \tau(x(t))) + \Delta f(t), $$ where $\alpha = 0.5$, $A = \bmatrix -1 & 0 \\ 0 & -1\endbmatrix $, $B = \bmatrix 0.1 & 0 \\ 0 & 0.1\endbmatrix $, $\tau(x(t)) = 0.2|x_1(t)| + 0.2|x_2(t)|$, $\Delta A(t) = 0.1\sin(t)I$, $\Delta f(t) = 0.05\cos(t)\bmatrix 1 \\ 1\endbmatrix $, $x(0)=\bmatrix 0.5 \\ 0.3\endbmatrix $, $I$ is a 2nd-order %\? identity matrix. First, verify the assumptions, for the time-delay constraint, we have $\dot{\tau}(x(t)) \leq 0.2\|\dot{x}(t)\|_1 \leq 0.2(\|A\| + \rho + \|B\|)\|x(t)\| \leq 0.25\leq 1$, for the perturbation bounds, $\|\Delta A(t)\| \leq \rho = 0.1, $ $\|\Delta f(t)\| \leq d = 0.05\sqrt{2}$. According to the stability criteria in \Par*{Theorem 3.1}, we select positive definite matrices $P=I$ and $Q=I$, Scaling %\? parameters $\varepsilon=\sigma =0.5$, $\nu=1$. By solving the LMIs we obtain the current state related LMI $\lambda = 0.8$, and the delayed state related LMI $\mu = 0.6$. Then, calculate the key stability parameter $m = \min\bigl\{ \frac{0.8}{1}, \frac{0.6}{1} \bigr\} = 0.6$. Next, by calculating the Caputo fractional derivative of $ V(t) $ and applying \Par*{Lemma 1} and Young's inequality, we obtain $$ ^C D_t^\alpha V(t) \leq -mV(t) + M $$ where $ M = \sigma^{-1} d^2 + \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2 = 0.025 $. By \Par*{Lemma 3}, the bound on $ V(t) $ implies $$ V(t) \leq V(0) E_\alpha\bigl(-m(t - 0)^\alpha\bigr) + \frac{M}{m} \bigl(1 - E_\alpha(-m(t - 0)^\alpha)\bigr). $$ Since $ V(0) = x^\top(0) P x(0) = 0.5^2 + 0.3^2 = 0.34 $, the state norm satisfies $$ \|x(t)\| \leq \sqrt{V(t)} \leq \sqrt{0.34} E_{0.5}(-0.6 t^{0.5}) + \sqrt{\frac{0.025}{0.6}} \bigl(1 - E_{0.5}(-0.6 t^{0.5})\bigr). $$ The system satisfies local robust Mittag-Leffler stability. \Figure \name{Fig.~1} \caption{State trajectories with state-dependent delay of Example 4.1.} \body \iftex \epsfxsize80mm \epsfbox{4985.ams+fig1.eps} \else \file{4985.ams+fig1.png} \fi \endbody \endFigure As clearly verified in \Fig{Fig. 1}, the numerical simulation results demonstrate the correctness of the theoretical derivation. The blue solid line represents the actual system state norm, which starts from the initial value $\|x(0)\|=0.58$ and exhibits excellent convergence characteristics under time-varying parameter perturbations and external disturbances. Meanwhile, the theoretical upper bound derived from Theorem~3.1, represented by the red dashed line in the figure, remains strictly above the actual state norm throughout the process. This not only confirms the effectiveness of the theoretical bound described by formula~\Tag(19) but also reveals its inherent conservatism. It is particularly noteworthy that despite the presence of state-dependent delay $\tau(x(t))=0.2|x_1(t)|+0.2|x_2(t)|$ and time-varying perturbation $\Delta A(t)=0.1\sin(t)I$, the system maintains robust Mittag-Leffler stability, which is perfectly consistent with our theoretical predictions. The simulation results that the stability criterion proposed in this paper can effectively handle time delays and uncertainties in fractional-order systems, providing a reliable theoretical basis for the analysis and design of such systems. \enddemo \demo{Example 2} Consider a fractional-order system with stronger nonlinearities and time-varying perturbations $$ {^C}D_t^\alpha x(t) = (A + \Delta A(t,x(t)))x(t) + B\sin x(t - \tau(x(t))) + \Delta f(t), $$ where $\alpha = 0.7$, $A = \bmatrix -1.5 & 0.2 \\ 0.1 & -1.4\endbmatrix $, $B = \bmatrix 0.08 & 0 \\ 0 & 0.08\endbmatrix $, $ \tau(x(t)) = 0.15(1 + \tanh(\|x(t)\|))$, $ \Delta A(t, x(t)) = 0.05e^{-\|x(t)\|} \cos(t) I $, $ \Delta f(t) = 0.04 \frac{t}{1 + t^2} \bmatrix 1 \\ -1 \endbmatrix $, $ x(0) = \bmatrix 0.6 \\ -0.4 \endbmatrix $. First, verify the assumptions, for the time-delay constraint, we have $\dot{\tau}(x(t)) = 0.15 \text{sech}^2(\|x(t)\|) \cdot \frac{x^\top(t) \dot{x}(t)}{\|x(t)\|} \leq 0.15 \cdot 1 \cdot \|\dot{x}(t)\| \leq 0.1 < 1$, for the perturbation bounds, $\|\Delta A(t)\| \leq \rho = 0.05, \|\Delta f(t)\| \leq d = 0.02$. According to the stability criteria in \Par*{Theorem 3.1}, we select positive definite matrices $P=\bmatrix 1 & 0 \\ 0 & 1.2\endbmatrix $ and $Q=1.5I$, Scaling %\? parameters $\varepsilon=0.6$ $\sigma =0.4$, $\nu=1$. By solving the LMIs we obtain the current state related LMI $\lambda = 0.4$, and the delayed state related LMI $\mu = 0.3$. Then, calculate the key stability parameter $m = \min\bigl\{ \frac{0.4}{1.2}, \frac{0.3}{1.5}\bigr\} = 0.2$. Next, by calculating the Caputo fractional derivative of $ V(t) $ and applying \Par*{Lemma 1} and Young's inequality, we obtain $$ {}^C D_t^\alpha V(t) \leq -mV(t) + M $$ where $ M = \sigma^{-1} d^2 + \| \Delta A(t) \|^2 + \| \Delta f(t) \|^2 = 0.0129 $. By \Par*{Lemma 3}, the bound on $ V(t) $ implies $$ V(t) \leq V(0) E_\alpha\bigl(-m(t - 0)^\alpha\bigr) + \frac{M}{m} \bigl(1 - E_\alpha(-m(t - 0)^\alpha)\bigr). $$ Since $ V(0) = x^\top(0) P x(0) = 0.6^2\cdot1 + (-0.4)^2 \cdot1.2 = 0.552 $, the state norm satisfies $$ \|x(t)\| \leq \sqrt{V(t)} \leq \sqrt{0.552} E_{0.7}(-0.2 t^{0.7}) + \sqrt{\frac{0.0129}{0.2}} \bigl(1 - E_{0.7}(-0.2 t^{0.7})\bigr). $$ The system satisfies local robust Mittag-Leffler stability. \Figure \name{Fig.~2} \caption{State trajectories with state-dependent delay of Example 4.2.} \body \iftex \epsfxsize80mm \epsfbox{4985.ams+fig2.eps} \else \file{4985.ams+fig2.png} \fi \endbody \endFigure As clearly verified in \Fig{Fig. 2}, the blue solid line represents the Actual state norm. Starting from the initial value, it gradually decays over time. It can be seen that the actual state norm decreases rapidly and eventually approaches near 0, showing good convergence characteristics. The red dashed line represents the Theoretical %\? upper bound, which is the upper bound of the state norm derived from the theorem. The actual state norm is always strictly below the theoretical upper bound, which verifies the correctness of the theoretical derivation and indicates that the system satisfies global robust Mittag-Leffler stability. That is, even with stronger nonlinearities and time-varying perturbations, the state of the system can still be effectively constrained by the theoretical upper bound, and the actual state will stabilize within a small range. \enddemo \head 5. Conclusion \endhead We investigate the problem of local robust Mittag-Leffler stability for a class of fractional-order dynamical systems subject to state-dependent delays and bounded perturbations. The main contributions are summarized as follows. a novel LMI-based stability criterion has been established by constructing an appropriate Lyapunov--Krasovskii functional that comprehensively accounts for the system's energy, historical state effects, and disturbance energy. A weakened sufficient condition has been derived, which relaxes the constraints on the time-varying delay and perturbations compared to existing results in the literature. The effectiveness and applicability of the proposed theoretical results have been verified through two numerical examples, including a linear system and a nonlinear system with stronger nonlinearities. \acknowledgments The authors sincerely thank the anonymous reviewers for their constructive comments, suggestions, and ideas to help improve this article. \Refs \ref\no 1 \by Diethelm~K. and Ford N.J. \paper Analysis of fractional differential equations \jour J.~Math. Anal. Appl. \yr 2002 \vol 265 \issue 2 \pages 229--248 \endref \ref\no 2 \by Lakshmikantham~V. \paper Theory of fractional functional differential equations \jour Nonlinear Anal. \yr 2008 \vol 69 \issue 10 \pages 3337--3343 \endref \ref\no 3 \by Morgado M.L., Ford N.J., and Lima P.M. \paper Analysis and numerical methods for fractional differential equations with delay \jour J.~Comput. Appl. Math. \yr 2013 \vol 252 %\issue - \pages 159--168 \endref \ref\no 4 \by Cushing J.M. \book Integrodifferential Equations and Delay Models in Population Dynamics \publ Springer Science and Business Media \publaddr Berlin and New York \yr 1977 %2013 \finalinfo Lecture Notes in Biomath., 20 \endref \ref\no 5 \by Joshi~H. and Jha~B.K. \paper Fractional-order mathematical model for calcium distribution in nerve cells \jour Comput. Appl. Math. \yr 2020 \vol 39 \issue 2 \num 56 \size 22 \endref \ref\no 6 \by Badri~V. and Tavazoei~M.S. \paper Stability analysis of fractional order time-delay systems: constructing new Lyapunov functions from those of integer order counterparts \jour IET Control Theory Appl. \yr 2019 \vol 13 \issue 15 \pages 2476--2481 \endref \ref\no 7 \by Narang~A., Shah S.L., and Chen~T. \paper Continuous-time model identification of fractional-order models with time delays \jour IET Control Theory Appl. \yr 2011 \vol 5 \issue 7 \pages 900--912 \endref \ref\no 8 \by Song S., Zhao H., and Zhang X. \paper Dynamic analysis of a fractional order delayed predator-prey system with harvesting \jour Theory in Biosciences \yr 2016 \vol 135 \issue 1 \pages 59--72 \endref \ref\no 9 \by Lin Z. and Wang H. \paper Modeling and application of fractional-order economic growth model with time delay \jour Fractal and Fractional \yr 2001 \vol 5 \issue 3 \num 74 \size 18 \endref % https://doi.org/10.3390/fractalfract5030074 \ref\no 10 \by Delavari~H., Baleanu~D., and Sadati~J. \paper Stability analysis of Caputo fractional-order nonlinear systems revisited \jour Nonlinear Dynam. \yr 2012 \vol 67 \issue 4 \pages 2433--2439 \endref \ref\no 11 \by Wang~Z., Yang~D., and Zhang~H. \paper Stability analysis on a class of nonlinear fractional-order systems \jour Nonlinear Dynam. \yr 2016 \vol 86 \issue 2 \pages 1023--1033 \endref \ref\no 12 \by Wang~Z., Yang~D., Ma~T., and Sun~N. \paper Stability analysis for nonlinear fractional-order systems based on comparison principle \jour Nonlinear Dynam. \yr 2014 \vol 75 \iftex \issue 1--2 \else \issue 1 \fi \pages 387--402 \endref \ref\no 13 \by Asma, G\'omez-Aguilar J.F., ur Rahman~G. and Javed~M. \paper Stability analysis for fractional order implicit $\Psi$-Hilfer differential equations \jour Math. Methods Appl. Sci. \yr 2022 \vol 45 \issue 5 \pages 2701--2712 \endref \ref\no 14 \by Zhang~F. and Li~C. \paper Stability analysis of fractional differential systems with order lying in $(1,2)$ \jour Adv. Difference Equ. \yr 2011 \vol 2011 \issue 1 \num 213485 \size 17 \endref \ref\no 15 \by He~B.-B., Zhou~H.-C., and Kou~C.-H. \paper Stability analysis of Hadamard and Caputo--Hadamard fractional nonlinear systems without and with delay \jour Fract. Calc. Appl. Anal. \yr 2022 \vol 25 \issue 6 \pages 2420--2445 \endref \ref\no 16 \by Wang~Y. and Li~T. \paper Stability analysis of fractional-order nonlinear systems with delay \jour Math. Probl. Eng. \yr 2014 \vol 2014 \issue 1 \num 301235 \size 8 \endref \ref\no 17 \by Tuan~H.T. and Trinh~H. \paper A linearized stability theorem for nonlinear delay fractional differential equations \jour IEEE Trans. Automat. Control \yr 2018 \vol 63 \issue 9 \pages 3180--3186 \endref \ref\no 18 \by Agrawal A.K., Fujino~Y., and Bhartia B.K. \paper Instability due to time delay and its compensation in active control of structures \jour Earthq. Eng. Struct. Dyn. \yr 1993 \vol 22 \issue 3 \pages 211--224 \endref \ref\no 19 \by Logemann~H. and Rebarber~R. \paper The effect of small time-delays on the closed-loop stability of boundary control systems \jour Math. Control Signals Systems \yr 1996 \vol 9 \issue 2 \pages 123--151 \endref \ref\no 20 \by Zhang~Z., Liu~L., Fang~J., and Qu~B. \paper Event-triggered stabilisation for stochastic delayed differential systems with exogenous disturbances \jour J.~Franklin Inst. \yr 2023 \vol 360 \issue 2 \pages 1395--1414 \endref \ref\no 21 \by Nguyen B.H., Dang~T.D., Furtat I.B., Dao~A.Q., and Gushchin~P.A. \paper Control of nonlinear systems with input delays and disturbances using state and disturbance sub-predictors \jour Mathematics \yr 2025 \vol 13 \issue 11 \num 1713 \size 15 \endref \ref\no 22 \by Li B., Wang Z., and Han L.Q. \paper Input-to-state stabilization of delayed differential systems with exogenous disturbances: The event-triggered case \jour IEEE Trans. Systems Man Cybernet. \yr 2017 \vol 49 \issue 6 \pages 1099--1109 \endref \ref\no 23 \by Li~P., Chen~L., Wu~R., et al. %Li~P., Chen~L., Wu~R., Tenreiro~Machado~J.A., Lopes~A.M., and Yuan~L. \paper Robust asymptotic stability of interval fractional-order nonlinear systems with time-delay \jour J.~Franklin Inst. \yr 2018 \vol 355 \issue 15 \pages 7749--7763 \endref \ref\no 24 \by Zhu~B., Ma~J., Zhang~Z., Feng~H., and Li~S. \paper Robust stability analysis and stabilisation of uncertain impulsive positive systems with time delay \jour Internat. J. Systems Sci. \yr 2018 \vol 49 \issue 14 \pages 2940--2956 \endref \ref\no 25 \by Liao~Z., Peng~C., Li~W., and Wang~Y. \paper Robust stability analysis for a class of fractional order systems with uncertain parameters \jour J.~Franklin Inst. \yr 2011 \vol 348 \issue 6 \pages 1101--1113 \endref \ref\no 26 \by Lu~J.-G. and Chen~Y.-Q. \paper Robust stability and stabilization of fractional-order interval systems with the fractional order $\alpha$: The $0<\alpha<1$ case %$0\ll\alpha\ll 1$ case \jour IEEE Trans. Automat. Control \yr 2010 \vol 55 \issue 1 \pages 152--158 \endref \ref\no 27 \by Chen~L., Wu~R., He~Y., and Yin~L. \paper Robust stability and stabilization of fractional-order linear systems with polytopic uncertainties \jour Appl. Math. Comput. \yr 2015 \vol 257 %\issue - \pages 274--284 \endref \ref\no 28 \by Lu~J.-G., Zhu~Z., and Ma~Y.-D. \paper Robust stability and stabilization of multi-order fractional-order systems with interval uncertainties: an LMI approach \jour Internat.~J. Robust Nonlinear Control \yr 2021 \vol 31 \issue 9 \pages 4081--4099 \endref \ref\no 29 \by Podlubny~I. \book Fractional Differential Equations: An Introduction to Fractional Derivatives, Fractional Differential Equations, to Methods of Their Solution and Some of Their Applications \publaddr San Diego \publ Academic \yr 1999 \finalinfo Math. Sci. Eng., 198 \endref %Elsevier, 1998. \ref\no 30 \by Vainikko~G. \paper Which functions are fractionally differentiable?\nocomma \jour Z.~Anal. Anwend. \yr 2016 \vol 35 \issue 4 \pages 465--487 \endref \ref\no 31 \by Stamova~I. \paper Global Mittag-Leffler stability and synchronization of impulsive fractional-order neural networks with time-varying delays \jour Nonlinear Dynam. \yr 2014 \vol 77 \issue 4 \pages 1251--1260 \endref \ref\no 32 \by Duarte-Mermoud~M.A., Aguila-Camacho~N., Gallegos~J.A., and Castro-Linares~R. \paper Using general quadratic Lyapunov functions to prove Lyapunov uniform stability for fractional order systems \jour Commun. Nonlinear Sci. Numer. Simul. \yr 2015 \vol 22 \iftex \issue 1--3 \else \issue 1 \fi \pages 650--659 \endref \ref\no 33 \by Khargonekar~P.P., Petersen I.R., and Zhou~K. \paper Robust stabilization of uncertain linear systems: quadratic stabilizability and H/sup infinity/control theory \jour IEEE Trans. Automat. Control \yr 1990 \vol 35 \issue 3 \pages 356--361 \endref \ref\no 34 \by Ye~H., Gao~J., and Ding~Y. \paper A generalized Gronwall inequality and its application to a fractional differential equation \jour J.~Math. Anal. Appl. \yr 2007 \vol 328 \issue 2 \pages 1075--1081 \endref \endRefs \enddocument