Wind power generation is a relatively mature new energy source at present, and the mainstream wind power generation technologies are doubly-fed and direct-drive. Doubly-fed induction generator (DFIG) systems occupy a dominant position in wind power generation systems due to their maturity, stability, and economy. However, in actual operation, alternating thermal stress caused by changes in random wind speed and operating conditions gradually reduces the reliability of wind power converters, which significantly impacts the stable operation of wind power systems.
Therefore, developing converter condition monitoring and reliability assessment technologies is of great significance for formulating and optimizing converter health management, reducing the failure rate of wind turbine converters, and improving the operational reliability of wind turbines.
Recently, researchers Huang Tao and Chen Minyou, among others, from the State Key Laboratory of Safety and New Technology for Power Transmission and Distribution Equipment and Systems (Chongqing University), proposed a reliability assessment model for wind power converters that considers fatigue accumulation and health status. Experimental results show that switching frequency and ambient temperature have a significant impact on converter reliability, and the failure rate of wind power converters increases significantly with the aging of power devices. The relevant research results were published in the 20th issue of the *Journal of Electrical Engineering* in 2018, with the title "Reliability Assessment Model for Wind Power Converters Considering Fatigue Accumulation and Health Status".
In converter fault statistics, failures caused by power devices account for over 31%. In recent years, researchers have conducted experimental analyses of two main failure modes of power modules, finding that the solder layer fails before the aluminum leads, and solder layer failure is the primary failure mode for IGBT modules. From the perspective of solder layer failure, analytical models have been used to predict the lifespan of IGBT modules, with different models exhibiting varying fitting accuracy and complexity. In actual load analysis, a linear fatigue accumulation model is used to evaluate device reliability; however, this model does not consider the influence of load sequence and cannot reflect the accelerated aging process of the device. Even when the accelerated aging process is considered, the effect of low thermal load on device aging cannot be demonstrated.
Existing reliability assessment models for wind power converters mostly only consider large loads or use the Miner linear fatigue accumulation model, while ignoring the impact of small loads and health status on the remaining lifespan of the module during actual operation.
Due to the randomness and fluctuation of wind speed, wind power converters are often subjected to low thermal loads. Researchers at the State Key Laboratory of Safety and New Technology for Transmission and Distribution Equipment and Systems (Chongqing University) used the SKM50GB12T4 module as the core of their research. Starting from the perspective of solder layer fatigue, they established a power device lifetime prediction model considering low load effects and designed a power cycle test to verify the model's accuracy. Subsequently, based on simulations of the converter junction temperature fluctuation curve at actual wind speeds, they used the lifetime model to calculate the failure rate of a 1.5MW doubly-fed converter.
Meanwhile, considering other subsystems of the converter such as drive, control system, and passive components, a reliability assessment model for wind power converters that takes into account fatigue accumulation and health status is proposed, combining online monitoring of power devices with converter reliability assessment. The impact of operating parameters on the reliability of doubly-fed wind power converters under different aging conditions is comprehensively analyzed, and corresponding reliability measures are proposed.
1) Traditional fatigue accumulation models differ from actual module lifespan because they do not consider the effects of low thermal loads during actual operation. In contrast, segmented fatigue accumulation models better reflect the module failure process and provide more accurate predictions.
2) The operating frequency of the machine-side converter is the output frequency of the doubly fed wind turbine. Its junction temperature fluctuation is more severe than that of the grid-side converter. The larger junction temperature fluctuation results in a higher failure rate for the machine-side converter, while the failure rate of the grid-side converter remains at a low level.
3) The higher the switching frequency and the higher the ambient temperature, the lower the reliability of the converter. When the converter is in the linear region, i.e., the early stage of aging, its failure rate is low and rises slowly; when the converter is close to the nonlinear region, the failure rate of the converter increases exponentially.
4) In wind farms with high regional wind speeds, the converter will frequently operate under heavy loads. In such cases, the switching frequency of the modules can be appropriately reduced, while simultaneously improving heat dissipation. While ensuring power quality meets standards, the switching frequency of components should be minimized to improve converter reliability.
Disclaimer: This article is a reprint. If there are any copyright issues, please contact us promptly for deletion (QQ: 2737591964). We apologize for any inconvenience.
