This With the increasing penetration of wind power in power systems, the volatility and randomness of its output pose severe challenges to grid stability. To address this core challenge, this study breaks through conventional modeling paradigms by proposing and implementing an innovative modeling framework that integrates environmental reconstruction and parameter identification. This paper employs the DIgSILENT PowerFactory software as a platform, not merely applying its built-in basic models, but constructing a high-precision, full-system model that encompasses the entire process from wind energy capture to power grid feeding. The model meticulously integrates the wind turbine aerodynamic module, pitch control, drivetrain mode. The core innovation of this study lies in two key steps: First, an innovative preprocessing approach was developed to reconstruct field-sampled wind speed data while accounting for wake and turbulence effects. This method overturns the conventional practice of directly using raw wind speed data as model inputs by introducing environmental intelligence, generating equivalent wind speed sequences that accurately reflect the complex loads acting on the wind turbine, thereby ensuring the physical authenticity of simulation input conditions at the source. Second, building on this foundation, the parameter identification functionality of DIgSILENT was deeply developed and utilized to autonomously construct a control parameter identification module specifically for doubly-fed induction generators. This module can drive the optimization and tuning of key control parameters of the unit using preprocessed high-confidence wind data, achieving a leap from “structural simulation” to “behavioral reproduction” in the model. The results indicate that, compared to traditional models, the model constructed using this method can more accurately simulate the dynamic response and electromechanical transient processes of wind turbines under complex wind conditions, demonstrating significant theoretical value and broad engineering application prospects.
To address the power fluctuation and frequency stability issues caused by the high proportion of new energy sources connected to the grid, this paper proposes an electro-hydrogen collaborative system architecture that integrates the electrolysis water hydrogen production process. This system converts excess wind and solar power into green hydrogen, thereby enhancing the consumption capacity and comprehensive utilization efficiency of clean energy. In terms of control strategy, a hierarchical coordinated control scheme is adopted: the upper layer, in response to the grid's frequency regulation requirements, employs virtual synchronous generator (VSG) control in the energy storage converter, simulating the inertia and damping characteristics of synchronous generators to provide active frequency support to the grid; the lower layer innovatively introduces multi-objective adaptive model predictive control (MPC), which, through a dynamic weight adjustment mechanism, achieves multi-objective collaborative optimization of current tracking accuracy, switching loss, harmonic suppression, and power smoothing. This adaptive MPC controller not only retains the traditional advantages of online rolling optimization and low dependence on model accuracy but also automatically adjusts control priorities based on frequency deviation status, significantly improving control accuracy and adaptability while ensuring the system's dynamic response speed. Simulation results show that the proposed control strategy can effectively coordinate system operation, enhance grid frequency stability, and achieve efficient consumption of renewable energy and hydrogen energy conversion, providing an effective technical path for building a new power system.
In recent years, frequent extreme events have seriously threatened the stable operation of the power system, which in turn has triggered extensive research interest in the resilience capability of novel urban distribution grids. This work gives an overview of the conceptual definition of resilient grids and the current status of research on capability enhancement, which summarizes the potentials of novel distribution grids in comparison with transmission grids, and elaborates the definition of urban resilient distribution grids. Subsequently, based on the resilience curve, the performance of resilient distribution grids is characterized in detail during the occurrence of extreme events, while the corresponding evaluation index system is proposed. Finally, the current status of existing research is comprehensively summarized from the perspective of performance characteristics enhancement, and possible future research directions are envisioned.
Against the backdrop of intelligent manufacturing, the network environment and data architecture of digital workshops are crucial for efficient operation, yet related research remains underexplored. This paper takes an aviation thermometer workshop as the research object. To address challenges like its multi-networks operating independently but interdependent in business, demands for high transmission and low latency, and existing issues in network and data processing, we define the concept of intelligent manufacturing-oriented digital workshops, design a four-layer system architecture, and adopt the architecture inheriting Lambda's advantages and adding edge computing and data feedback loops to build the data architecture considering multiple elements. Results show the multi-network environment meets high integration and real-time requirements, and this architecture solves problems of high data concurrency and diversity, providing a basic data processing solution for the workshop's digital and intelligent development.
Gas insulated switchgear (GIS) is critical for power grid safety, but internal faults like partial discharge, contact overheating, and mechanical failures pose significant risks if not detected promptly. Traditional rigid sensors suffer from poor surface conformity and signal distortion, while conventional diagnostic methods depend heavily on manual feature extraction, limiting generalization. This paper proposes a non-invasive monitoring technique using high-sensitivity flexible acoustic sensing. A four-layer polyvinylidene fluoride (PVDF) flexible sensor is designed to enhance surface adhesion and weak signal (5–50 mV) capture. An adaptive feature extraction method combining ensemble empirical mode decomposition (EEMD) and Hilbert transform is introduced to automatically derive key features such as fundamental frequency ratio, multi-scale entropy, and odd-even harmonic ratio. These features are fed into an artificial neural network (ANN) for fault diagnosis. Simulations on a 110 kV GIS show the system effectively captures fault signals with periods of 0.02–0.1 s. The ANN model achieves recognition accuracies of 95
