With the increasing integration of industrial production systems, multi-stage processes become more complex and exhibit strong stage heterogeneity and inter-stage error propagation, which limits conventional approaches for anomaly detection and root cause analysis. This paper proposes a causal graph neural framework for anomaly detection and root cause analysis in multi-stage industrial processes. The process is modeled as a hierarchical causal structure composed of a global graph and stagelevel subgraphs which are intended for interpretable fault propagation tracing under explicit process assumptions. Process prior knowledge is encoded as feasibility gates to constrain candidate edges, upon which learnable masks select effective connections. Global graph captures inter-stage causal dependencies, while masked subgraph learns intra-stage dynamic relationships. Then, the spatio-temporal layers with temporal encoding and graph convolution are employed for representation learning and forecasting. Residual-based anomaly detection then activates a hierarchical diagnosis pipeline, including stage-level localization via the global graph, intra-stage attribution of key parameters via dynamic subgraphs, and fault propagation path inference. Experiments on a multistage industrial dataset show that the proposed method demonstrates the feasibility of residual-based anomaly triggering and structured root-cause tracing and provides a structured and interpretable root cause analysis solution for complex manufacturing systems.

The internal dynamics of grid-forming modular multilevel converters (GFM-MMCs) are coupled with the external output characteristics. However, the destabilizing potential of these internal dynamics is often underestimated, posing a growing threat to the reliable operation of GFM-MMCs, especially as the submodule count increases. To address this risk, this paper employs vector analysis to investigate this internal dynamic and characterizes it as negative damping, subsequently developing a submodule capacitor voltage fluctuation–negative damping (SVF–ND) model. The developed model indicates that the negative damping is proportional to the number of submodules, implying an increased risk of instability as the submodule count increases. Given this challenge, an internal dynamic shaping (IDS) method is proposed to mitigate the negative damping effect. The method reduces the negative-damping region, thereby improving the stability margin and enhancing overall system stability. Furthermore, it increases the equivalent submodule capacitance and reduces capacitor voltage fluctuation, thereby extending capacitor lifetime in a cost-effective manner, and maintains the original operating point without requiring additional parameter tuning. Finally, both simulation and experimental results validate the proposed analysis and control strategy.

This paper proposes a two-step scheme for estimating the frequency and its rate of change of non-stationary signals in modern power electronics-dominated power systems. In the first step, a finite-time identification scheme based on the Volterra integral operators is designed to track the frequency of the non-stationary signals with fast convergence. A novel augmented framework of signal identification is proposed to eliminate the effects of the harmonics and the measurement noise. In the second step, the estimated frequency is injected into a non-asymptotic numerical differentiator to track its rate of change. It has been proven the proposed estimator possesses fast convergence and robustness against the disturbance. Finally, simulation and experiments are conducted, and the estimation results are compared with known algorithms, demonstrating the effectiveness of the proposed method.

The convergence stability and accuracy of PMSM parameter estimation are fundamentally governed by model observability. While fundamental-frequency approaches are established, the theoretical observability of emerging models relying on high-frequency current derivative information remains largely unexplored. This paper proposes a real-time quantitative framework for evaluating the observability of PMSM parameter-estimation basis models within a unified multiscale perspective. The proposed method introduces a condition-number-based metric derived from the information matrix, enabling real-time assessment of model excitation and conditioning. Most existing parameter estimation methods are summarized into three basis models, the fundamental-frequency time-scale model (FFTM), the switching-time-scale full model (STFM), and the virtual voltage vector model (VVVM). These models are comparatively analyzed within this framework. Analytical results reveal the excitation-dependent conditioning characteristics of each model and identify their weakly observable conditions. An adaptive observer guided by the proposed metric is utilized specifically to demonstrate the effectiveness of the observability assessment. Both numerical and experimental results confirm that the proposed criterion provides a reliable and real-time index for model observability evaluation and parameter-identification stability.

Restricting bus voltage deviation is crucial for normal operation of multi-bus DC microgrids, yet it has received insufficient attention due to the conflict between two main control objectives in DC microgrids, i.e., voltage regulation and current sharing. By revealing a necessary and sufficient condition for achieving these two objectives, this paper proposes a novel consensus-based current sharing control law that can achieve the compromised control objective, balancing both current sharing and voltage deviation restriction. Additionally, we examine the effectiveness of the proposed control scheme for DC Microgrids that include both critical nodes and ordinary nodes, where there is a simultaneous requirement for voltage deviation limits on critical nodes and accurate current sharing among ordinary nodes. Theoretical results are verified by simulations, and the effectiveness in handling plug-and-play operations of distributed generators is also illustrated.

The shift to renewable-dominated power systems has produced low-inertia grids, undermining system stability. In this context, grid-forming inverters (GFMs) have emerged as a promising solution. However, GFMs challenge conventional analysis techniques, especially those relying on small-signal or root-mean-square (RMS) models. Such models rely on linearization and sinusoidal steady-state assumptions, which fail in large-signal cases. Stability of GFM-based systems therefore becomes operating-point dependent, and a feasible operating point may not even exist. While large-signal analyses are available, decentralized certification of operating-point convergence with explicit transient guarantees, such as rate and overshoot, remains rare. This paper proposes an algebraic, decentralized contraction-based framework. The proposed contraction stability analysis certifies system stability and convergence to desired operating points. The method works in the time domain and captures nonlinear, large-signal behavior of synchronization and power-sharing mechanisms. Moreover, the contraction rate provides an explicit bound on transient time: trajectories converge exponentially to the new operating point at a controlled rate, yielding computable contraction regions that certify stability and large-signal convergence across operating-point changes. These regions directly guide parameter tuning for heterogeneous GFMs.

This paper proposes a distributed observer for linear time-invariant (LTI) non-autonomous systems. The observer can handle time-varying topologies. Based on multi-hop communication and Volterra operator, all nodes can reconstruct the global state within a fixed time. The concept of “freshness” is introduced to update the information used for estimation in real-time, so that the observer can instantaneously respond to topology changes. When the changes prevent nodes from obtaining sufficient information from their neighbors, the algorithm automatically switches to open-loop prediction to ensure reconstruction. Compared with existing methods, the algorithm proposed in this paper relaxes the common joint strong-connectivity assumption and is applicable to more general non-autonomous systems, significantly improving its practicality. Finally, the effectiveness and robustness of the proposed observer were verified through numerical simulations in both noise-free and noisy scenarios.

Accurate online maps of the incremental inductances L-d, L-q are critical to model-based drives, sensorless observers, and real-time fault diagnosis. Existing techniques either rely on slow, equipment-intensive offline tests or inject auxiliary signals that disturb the drive, depend on precise rotor position, and are difficult to calibrate against inverter nonlinearities. This article sidesteps those limitations by recasting the fundamental voltage vector sequence as a differential virtual-voltage vector that inherently rejects dead-time distortion and device voltage drops. A piecewise multisample linear regression-applied only to the quasi-linear segment of each current trace-then yields a dead-time-immune, noise-resilient estimate of current slopes. The proposed approach introduces three mutually reinforcing advances: 1) a virtual-voltage-vector-oriented framework that inherently accounts for inverter nonlinearities; 2) a multisampling-based current-slope estimation strategy that is immune to dead-time effects and suppresses noise; and 3) a signal-injection-free, rotor-position-independent inductance observer that is explicitly built on the first two advances, harnessing their accurate voltage modeling and derivative measurements. The proposed estimator is evaluated against the standard ac-standstill test and a current-slope online method; the mean error remains below 5% in every case. Validation spans the full operating range, including extreme load and speed transients.

Accurate incremental inductance maps of the d - and q-axes underpin model-based predictive control, sensorless operation, and fault diagnosis of permanent-magnet synchronous motors (PMSMs). This article presents a noninvasive offline measurement technique that exploits the current-derivative transients naturally generated by inverter switching events. By reformulating the discrete switching model in a virtual-voltage-vector-oriented frame, the inductance measurement task becomes a rotor-angle-independent geometric fitting problem. A robust fitting solver with outlier rejection mitigates nonlinear effects such as sensor drift and noise. The method operates on a standard finite-control-set (FCS) drive and requires no shaft lock, additional signal injection, or extra sensors, yet yields a full-range inductance map. Experimental results on an interior PMSM demonstrate that the calibrated inductance map is in close agreement with finite-element analysis (FEA), conforms to the IEEE Std 1812-2023 offline benchmark, and remains consistent with values obtained from a current-derivative-based online estimator.

碳中和目标正在推动现代能源系统由集中、 单一和确定的运行模式向分布式、 多能耦合和高度不确定的形态加速演进。 风能、 光伏、 储能、 氢能及供热网络的大规模发展对能源系统的预测、 规划、 调度、 控制和全生命周期管理提出了更高要求。 人工智能可通过数据驱动预测、 物理信息学习、 强化学习、 智能优化、 数字孪生和生成式模型加快高保真模拟, 提升可再生能源预测与消纳能力, 促进电-热-气-氢协同调度, 并服务于设备设计、 故障诊断、 材料发现、 碳核算和排放监测。 本专辑聚焦人工智能与低碳能源工程的深度融合, 强调将物理机理、 不确定性量化、 安全约束和可解释决策嵌入智能方法, 推动能源研究由黑箱预测迈向可信决策、 由单一设备优化迈向系统与全生命周期协同。 未来需进一步解决数据共享、 模型泛化、 网络安全、 隐私保护及工程验证等问题, 为构建安全、 经济、 韧性和可验证低碳的未来能源系统提供科学依据与技术支撑。

Unplanned grid-connected (GC)/stand-alone (SA) transitions commonly occur in AC microgrids during protection trips, manual breaker operation, or low-bandwidth supervisory communication. Under such unplanned transitions, a grid-forming inverter must support the local-load voltage in stand-alone operation and regulate the desired power/current injection in grid-connected operation. Existing P–Q droop-based seamless-transfer methods often rely on planned transition commands, supervisory islanding detection, or pre-synchronization interval, which may prevent timely voltage/current support during unplanned bidirectional transitions. To address this problem, this paper proposes a seamless contraction-control (SCC) framework for target dynamics. Using the SCC, contraction-based grid-connected current-control and stand-alone voltage-control laws are proposed. With the new control laws, the inverter achieves transient stability and converges to the target trajectory with a prescribed convergence rate. Furthermore, a breaker-status observer is proposed to infer the grid-connected/stand-alone mode from voltage measurements on both sides of the breaker, eliminating the need for a dedicated pre-synchronization interval or supervisory islanding detection process and enabling timely voltage/current support during unplanned transitions. Experimental results validate that the proposed method achieves stand-alone voltage support, stable grid-connected current injection under symmetrical/unsymmetrical grid-voltage sag and phase-jump disturbances, and unplanned bidirectional transitions.

Injecting appropriate circulating currents can reduce the submodule capacitor voltage ripple in the modular multilevel converters (MMCs) and improve their internal performance. However, few studies have explored the impact of circulating current injection control (CCIC) on the system stability of grid-forming MMCs (GFM-MMCs), especially across a wide range of grid strengths. In this article, sequence impedance responses are modeled for stability analysis. A direct link is established between low-frequency oscillations and the negative damping region in the sequence impedance model. By applying CCIC based on instantaneous information, the sequence impedance of the MMC is reshaped, which reduces the negative damping region and enhances system stability. This behavior differs from its effect in grid-following MMCs (GFL-MMCs). However, the circulating current injected based on the output instantaneous is deterministic, which limits its ability to improve the system stability. To overcome this limitation, a power secondharmonic injection control (PSHIC) is proposed. By tuning the control coefficient, the stability margin can be further extended. The proposed impedance model and stability assessment are validated through simulations.

Voltage regulation under conventional grid-forming controllers is tightly coupled to power sharing and dc-link dynamics. Consequently, its tracking accuracy deteriorates during grid faults, sudden power sharing changes, or dc-bus voltage varying. To address this issue, a symmetric sliding-mode control (SSMC) method is developed and its voltage precision region is derived. It illustrates how much ac-side power dynamics and dc-link voltage varying can be decoupled from the voltage regulation task, which helps predict when an abnormal entangling appears. While conventional sliding-mode controls address voltage-tracking error through complex sliding surface designs, repetitive correction techniques or special reaching laws, this work identifies that the error at power-line frequency primarily stem from the asymmetry property of inverters with the delay effect and the computational inaccuracy. Guided by this insight, an asymmetry compensation structure is proposed, which avoids added design complexity and directly mitigates voltage tracking error. Furthermore, the control design is supported by a physical and quantitative explanation, aiding in parameter tuning. Simulation and experimental results demonstrate that the proposed method achieves faster tracking responses while maintaining robust and more accurate tracking under both dc-link voltage and ac-side current variations. Conventional grid-forming and classical sliding-mode controllers, which handle these variations separately, cannot match this combined speed and robustness. Furthermore, the voltage precision region is explicitly verified.

State estimation for linear time-invariant systems with unknown inputs is a fundamental problem in various research domains. In this article, we establish conditions for the design of unknown input observers (UIOs) from a geometric approach perspective. Specifically, we derive a necessary and sufficient geometric condition for the existence of a centralized UIO. Compared to existing results, our condition offers a more general design framework, allowing designers the flexibility to estimate partial information of the system state. Furthermore, we extend the centralized UIO design to distributed settings. In contrast to existing distributed UIO approaches, which require each local node to satisfy the rank condition regarding the unknown input and output matrices, our method accommodates cases where a subset of nodes does not meet this requirement. This relaxation significantly broadens the range of practical applications. Simulation results are provided to demonstrate the effectiveness of the proposed design.

To achieve accurate and stable online identification of inductance parameters for PMSM(permanent magnet synchronous motor),an online inductance observation method based on virtual voltage vector excitation and current differential response was proposed,which required no additional test signal injection and was decoupled from rotor position,stator resistance,and permanent magnet flux linkage.By introducing the concept of a virtual voltage vector-oriented coordinate system,it was analytically derived and proven that the d-and q-axis inductances of a PMSM can be observed independently of the angular position in the conventional d-q synchronous reference frame.Building on this,the implementation procedure for extracting virtual voltage vectors and current differential information was discussed in detail,enabling non-intrusive inductance identification without any signal injection.The effectiveness and accuracy of the proposed method were validated by comparison with offline test procedures in IEEE standards.

This addresses the consensus problem of a leader-following multi-agent system (MAS) in the presence of attacked nodes. A distributed observer is proposed deploying the subsequence-reduced algorithm (SRA) based on a thorough analysis of the attack pattern and scope of the cyber-attack model. By exploiting the robustness of the network communication graph and the redundant information, the proposed observer effectively eliminates cyber-attack impacts and achieves consensus convergence. The consensus property and attack rejection capability are analyzed by Lyapunov theory for both subsequence-reduced linear distributed observer (SRLDO) and the subsequence-reduced adaptive linear distributed observer (SRALDO). The effectiveness of the proposed observer is verified through a two-link robotic arm cluster experiment with comparisons to recent methods.

Thyristor controlled phase shifting transformer (TCPST) is a new type of power flow control equipment. Due to its series and shunt coupling topology and special connection to the power grid, it is necessary to consider the handling strategy in case of power grid faults. Based on the structure and principle of TCPST, the influence of grid fault on TCPST is analyzed in this paper. It is clarified the distribution of fault current and characteristics of overcurrent and overexcitation within TCPST. On this basis, the control strategy of TCSPT for deal with grid fault impact is proposed, and the fault state identification criterion and control sequences for the TCPST is constructed. By identifying the faults within and outside the zone of TCPST and the recovery state of power grid faults with the electrical quantities, the out of service and auto-restart strategy for TCPST are realized. The temporarily withdraw method for TCPST is used to protect equipment safety in the event of severe power grid faults. Finally, some simulations are conducted for the proposed fault ride-through strategy of TCPST. It is shown that the reliable ride-through of TCPST during power grid faults can be achieved by using the strategy, which has a good application for feasibility and effectiveness of TCPST. It is benefit to take advantage of the potential of power flow regulation function of TCPST.

Most previously proposed controllers are analyzed in the small-signal/quasi-steady regime rather than large-signal or transient stability for grid-forming inverters (GFMI). Additionally, methods that presume system-wide data–global measurements and complete grid-model knowledge–are challenging to realize in practice and unsuitable for large-scale operation. Moreover, proportional current sharing is rarely embedded into them. The whole system is a high-order, nonlinear differential system, making analysis intractable without principled simplifications. Hence, contraction stability analysis in GFMI is proposed to guarantee the large-signal stability. Furthermore, a contraction-based controller is proposed to synchronize GFMI. Additionally, this paper proposes integrating an auxiliary virtual-impedance layer into the contraction-based controller to achieve proportional current sharing, while the GFMI retains global stability and voltage synchronization. A dispatchable virtual oscillator control (dVOC), also known as the Andronov–Hopf oscillator (AHO) is used to validate the proposed contraction stability analysis and contraction-based controller with virtual-impedance. It is proved that the complex multi-converter system can achieve output-feedback contraction under large-signal operation. Therefore, without requiring system-wide data, the proposed method offers voltage synchronization, decentralized stability conditions for the transient stability of AHO and proportional current sharing, beyond prior small-signal, quasi-steady analysis.

Estimation of fundamental frequency and sinusoidal components is required for the regulation of modern power electronics-dominated power systems. Most of the existing estimation methods are designed for signals with stationary frequency. Hence, their accuracy could significantly degrade in the face of non-stationary frequencies, which is common in low-inertia power systems. In this paper, we propose a novel scheme for real-time estimation of a time-varying power frequency and the resulting fundamental signal. This is a time-domain method for (1) estimating non-stationary frequency and (2) fundamental signal reconstruction. It has the advantage of tracking the fundamental frequency component and treating all harmonics and subharmonics as noise. The method is based on a kernel-based estimation scheme and characterized by high accuracy, fast response, and noise immunity because of the inclusion of non-asymptotic kernel functions. The effectiveness of the proposed estimation scheme for non-stationary frequency tracking and fundamental signal reconstruction is verified by simulation and experimental results, which explore the use of the proposed scheme for frequency extraction of power signals appear in real world low-inertia systems.

This paper proposes a robust observer for nonlinear systems. A transformation is designed to convert the nonlinear system into a linear one, utilizing the Volterra integral operator to mitigate the effects of initialization errors. Neural network techniques are then applied to learn both the transformation and its inverse. As a result, the state variables of the nonlinear system can be accurately tracked using the estimated transformation, providing enhanced robustness against measurement noise and model uncertainty. Specifically, the proposed approach reduces the trade-off between convergence speed and noise immunity, which is often influenced by the choice of observer eigenvalue. The paper provides comprehensive error analysis and validation through simulations.