With the increasing demand for high reliability and availability in power conversion equipment within power electronics systems,the fault diagnosis of neutral-point-clamped(NPC) three-level inverters has garnered wide...With the increasing demand for high reliability and availability in power conversion equipment within power electronics systems,the fault diagnosis of neutral-point-clamped(NPC) three-level inverters has garnered widespread attention.To address the challenges of fault feature extraction,this article proposes an end-to-end diagnostic approach based on a wavelet kernel convolutional neural network (WKCNN),capable of extracting multi-scale features from current signals to significantly enhance diagnostic accuracy.This method directly uses raw three-phase current signals as input,applying wavelet kernel convolution to automatically capture frequency-domain fault features,combined with a Softmax classifier optimized by the Adam algorithm to achieve fault diagnosis for NPC threelevel inverters.Experimental results under various operating conditions demonstrate that this approach maintains robust diagnostic accuracy across multiple fault scenarios,with comparative analysis further confirming its advantages in diagnostic efficiency and performance over traditional machine learning and other deep learning methods.展开更多
With the rapid integration of renewable energy sources,modern power systems are increasingly challenged by heightened volatility and uncertainty.Doubly-fed variable-speed pumped storage units(DFVS-PSUs)have emerged as...With the rapid integration of renewable energy sources,modern power systems are increasingly challenged by heightened volatility and uncertainty.Doubly-fed variable-speed pumped storage units(DFVS-PSUs)have emerged as promising technologies for mitigating grid oscillations and enhancing system flexibility.However,the excitation converters in DFVS-PSUs are prone to significant issues such as elevated common-mode voltage(CMV)and neutral-point voltage(NPV)fluctuations,which can lead to electromagnetic interference and degrade transient performance.To address these challenges,an optimized virtual space vector pulse width modulation(OVSVPWM)strategy is proposed,aiming to suppress CMV and NPV simultaneously through coordinated multi-objective control.Specifically,a dynamic feedback mechanism is introduced to adjust the balancing factor of basic vectors in the synthesized virtual small vector in real-time,achieving autonomous balancing of the NPV.To address the excessive switching actions introduced by the OVSVPWM strategy,a phase duty ratio-based sequence reconstruction method is adopted,which reduces the total number of switching actions to half of the original.A zero-level buffering scheme is employed to reconstruct the single-phase voltage-level output sequence,achieving peak CMV suppression down to udc/6.Simulation results demonstrate that the proposed strategy significantly improves electromagnetic compatibility and operational stability while maintaining high power quality.展开更多
This paper proposes a novel SVPWM （space vector pulse width modulation） strategy for the three-level neutral-point-clamped voltage source inverter, based on the particular disposition of all the redundant voltage ve...This paper proposes a novel SVPWM （space vector pulse width modulation） strategy for the three-level neutral-point-clamped voltage source inverter, based on the particular disposition of all the redundant voltage vectors. The new modulation approach shows superior performance for harmonic voltage and balancing control of neutral-point potential compared to the popular eight-stage centered SVPWM. It realizes suppression of inverter neutral-point potential variation by accurately modifying redundant factor of small vectors pairs, only requiring information of DC-link capacitor voltages and three-phase load currents. This is convenient to apply and is compatible of digital computer realization. Feasibility of the proposed control approach is verified by simulation and experimental results.展开更多
In the traditional three-level space vector pulse width modulation(SVPWM)algorithm,the sector judgment is computationallycomplex since the sector is divided into triangles and hexagons.In addition,the switching freque...In the traditional three-level space vector pulse width modulation(SVPWM)algorithm,the sector judgment is computationallycomplex since the sector is divided into triangles and hexagons.In addition,the switching frequency is high becausethe seven-segment switching sequence is adopted.For this reason,a new SVPWM control algorithm for three-level inverteris proposed,in which the sector judgment is simplified by dividing the sector into quasi hexagons?and the new four-segmentswitching sequence is adopted to reduce the switching frequency.Simulation results show that the total harmonic distortiongrows down with the switching frequency decreasing,moreover,the algorithm runtime is also decreased.展开更多
基金supported in part by Zhejiang Provincial“Pioneer”and“Leading Goose”R&D Program of China under Grant 2024C01014the National Natural Science Foundation of China under Grant52177055。
文摘With the increasing demand for high reliability and availability in power conversion equipment within power electronics systems,the fault diagnosis of neutral-point-clamped(NPC) three-level inverters has garnered widespread attention.To address the challenges of fault feature extraction,this article proposes an end-to-end diagnostic approach based on a wavelet kernel convolutional neural network (WKCNN),capable of extracting multi-scale features from current signals to significantly enhance diagnostic accuracy.This method directly uses raw three-phase current signals as input,applying wavelet kernel convolution to automatically capture frequency-domain fault features,combined with a Softmax classifier optimized by the Adam algorithm to achieve fault diagnosis for NPC threelevel inverters.Experimental results under various operating conditions demonstrate that this approach maintains robust diagnostic accuracy across multiple fault scenarios,with comparative analysis further confirming its advantages in diagnostic efficiency and performance over traditional machine learning and other deep learning methods.
文摘With the rapid integration of renewable energy sources,modern power systems are increasingly challenged by heightened volatility and uncertainty.Doubly-fed variable-speed pumped storage units(DFVS-PSUs)have emerged as promising technologies for mitigating grid oscillations and enhancing system flexibility.However,the excitation converters in DFVS-PSUs are prone to significant issues such as elevated common-mode voltage(CMV)and neutral-point voltage(NPV)fluctuations,which can lead to electromagnetic interference and degrade transient performance.To address these challenges,an optimized virtual space vector pulse width modulation(OVSVPWM)strategy is proposed,aiming to suppress CMV and NPV simultaneously through coordinated multi-objective control.Specifically,a dynamic feedback mechanism is introduced to adjust the balancing factor of basic vectors in the synthesized virtual small vector in real-time,achieving autonomous balancing of the NPV.To address the excessive switching actions introduced by the OVSVPWM strategy,a phase duty ratio-based sequence reconstruction method is adopted,which reduces the total number of switching actions to half of the original.A zero-level buffering scheme is employed to reconstruct the single-phase voltage-level output sequence,achieving peak CMV suppression down to udc/6.Simulation results demonstrate that the proposed strategy significantly improves electromagnetic compatibility and operational stability while maintaining high power quality.
文摘This paper proposes a novel SVPWM （space vector pulse width modulation） strategy for the three-level neutral-point-clamped voltage source inverter, based on the particular disposition of all the redundant voltage vectors. The new modulation approach shows superior performance for harmonic voltage and balancing control of neutral-point potential compared to the popular eight-stage centered SVPWM. It realizes suppression of inverter neutral-point potential variation by accurately modifying redundant factor of small vectors pairs, only requiring information of DC-link capacitor voltages and three-phase load currents. This is convenient to apply and is compatible of digital computer realization. Feasibility of the proposed control approach is verified by simulation and experimental results.
基金National Natural Science Foundation of China(No.61261029)
文摘In the traditional three-level space vector pulse width modulation(SVPWM)algorithm,the sector judgment is computationallycomplex since the sector is divided into triangles and hexagons.In addition,the switching frequency is high becausethe seven-segment switching sequence is adopted.For this reason,a new SVPWM control algorithm for three-level inverteris proposed,in which the sector judgment is simplified by dividing the sector into quasi hexagons?and the new four-segmentswitching sequence is adopted to reduce the switching frequency.Simulation results show that the total harmonic distortiongrows down with the switching frequency decreasing,moreover,the algorithm runtime is also decreased.
