Experimental Technology and Management

针对电网数字化转型下电力专用传感器的供能问题以及相关教学手段不足问题，提出基于能量流的变压器振动压电取能仿真与实验教学平台构建方案。设计了变压器振动电磁-机械振动耦合仿真与实验、串并联压电阵列力-电耦合仿真与实验、能量管理电路仿真与实验、储能电容充电实验等实验功能及实验案例，探讨了该仿真与实验平台在跨学科教学和交叉学科研究方面的用途。该平台为电气工程、智能电网信息工程等专业学生参与科研训练和工程实践提供了重要的实践平台，促进了问题式教学模式的教学方法创新，同时也为电力设备低功耗传感器供能问题研究提供了有力支撑。

[Objective] To address critical energy supply challenges for power-specific sensors and pedagogical limitations exposed during power grid digitalization, this study presents a holistic simulation and experimental platform for piezoelectric vibration energy harvesting, targeting transformer vibrations. [Methods] Designed to bridge theoretical knowledge and practical engineering applications, the proposed platform comprehensively integrates four synergistic modules spanning the entire energy flow pathway.(1) The electromagnetic– mechanical vibration coupling module employs COMSOL Multiphysics for the finite element modeling of power transformer dynamics, simulating core vibrations governed by magnetostriction principles(where acceleration scales proportionally with voltage squared) and winding vibrations(where acceleration correlates with current squared), predicting dominant frequencies at 100 and 200 Hz. These results are experimentally validated using CT1010 LF piezoelectric accelerometers mounted magnetically on transformer surfaces, capturing realworld vibrations such as the measured 0.42 g acceleration amplitude at a 200-Hz frequency from a 240-kVA/220-kV SFSZ10-240000/220 transformer under operational conditions.(2) The force–electric coupling module focuses on series–parallel piezoelectric arrays utilizing Sm-doped lead magnesium niobate–lead titanate bimorph cantilevers. It incorporates detailed simulations to analyze resonance shifts when adjusting tungsten mass block positions along the cantilever axis and establishing Multisim-compatible equivalent circuit models parameterized through laser displacement testing. It shows a free-end vibration displacement U_d of 0.021 65 mm and derived sinusoidal current sources I_p = I_M sin(ωt) with clamped capacitance C_p values of 27.32 nF. Experimental validation under 1g acceleration excitation demonstrates optimized configurations, such as nine parallel arrays delivering maximum power outputs of 2.341 mW at a load resistance of 150 kΩ.(3) The energy management module simulates, designs, and implements power conditioning circuits exemplified by the synchronous electric charge extraction topology. Multisim simulations guide printed circuit board layouts fabricated using Altium Designer to achieve efficient AC–DC conversion and voltage regulation.(4) The energy storage and delivery module evaluates charging–discharging cycles for capacitors powering sensors. It documents 16 daily operational cycles for a 5-V vibration monitoring sensor, consuming 0.6 V per measurement event with systematic charging time metrics quantifying energy autonomy performance. [Results] Extensive case studies centered on the SFSZ10 transformer validate platform efficacy: electromagnetic simulations predict vibration hotspots, whereas experimental measurements confirm spatial variations in surface acceleration. Meanwhile, piezoelectric array simulations reveal that parallel configurations boost current summation and power transfer to lower resistance loads, whereas experiments identify optimal 150 kΩ matching points. SECE circuit comparisons demonstrate enhanced efficiency over basic rectifiers. Moreover, sensor charging tests verify sustainable operation under realistic duty cycles. Pedagogically, this integrated framework enables problem-based learning where students execute vibration spectral analysis using Fourier transforms, design cantilever structures by tuning mass positions to match 100–300 Hz vibration spectra, construct electrical models from laser-measured displacement and voltage data, compare array topologies through output power curves, and evaluate circuit efficiency using capacitor charging rates. These cultivate competencies in multiphysics simulation, experimental instrumentation, and energy system optimization. For research, the platform's modular architecture facilitates parametric studies on piezoelectric material selection, array scalability, circuit topologies, and storage integration—directly advancing solutions for autonomous sensor networks in digitalized power equipment. Quantitative outcomes include achieving 4.354 mW simulated power at 21 kΩ for nine parallel piezoelectric patches versus 2.341 mW experimentally at 150 kΩ, identifying material and impedance matching challenges; transformer vibration mapping showing 71.43% of measurement points exhibiting 200-Hz center frequencies; and capacitor charging sequences enabling multicycle sensor operation without external power. [Conclusions] By unifying high-fidelity simulations with hands-on experiments across the electromagnetic, mechanical, electrical, and energy domains, this platform not only revolutionizes engineering education for electrical and smart grid disciplines but also establishes an extensible research testbed for developing self-powered systems critical to grid digitalization under China's dual carbon strategy. This effectively closes the gap between academic concepts and field-deployable energy harvesting technologies for next-generation power infrastructure.