This is due to the fact that the SSA requires to have detailed electro-thermal characteristics and lifetime models of the converter components. Using stress-strength analysis for reliability modeling for each converter is not possible in practice. Moreover, power networks are complicated systems of systems that include huge number of components. As a result, estimating the converter lifetime for various operating conditions can be time-consuming considering the electro-thermal modeling and Monte Carlo analysis 7, 18, 19, 20, 21, 22. Thereby, following the grid generation and demand, the converter operating condition will be different. In general applications such as battery storage systems, electric vehicle chargers, interlinking converters in micro-grids, and multi-terminal DC grids, such converters are facing different operating conditions depending on power system demand and generation. This concept has been widely applied for photovoltaic and wind power applications 1, 6, 18 in order to design the converter or improve its reliability under a given mission profile. In fact, physics of failure analysis is one step ahead of power system reliability engineering, which just relied on historical data without considering the mission-profile based aging of converters. The validation of the models for converter reliability prediction with power cycling tests has been addressed in the literature e.g., in 16. The models used for strength of components are obtained either empirically or theoretically 6, 16, 17, which are modeling the aging process of power devices 7. According to this method, Stress-Strength Analysis (SSA) is performed to obtain the aging probability of converter components taken into account the various uncertainties. In recent decade, in power electronics engineering model-based reliability analysis on the basis of physics of failures is developed to model the impact of different failure sources and mechanisms on the reliability of converters. Meanwhile, these approaches are still the method of choice used in power system reliability engineering. Due to this fact, physics of failure analysis based approaches have been introduced in recent decade to assess and enhance the reliability of converters 1, 9, 10, 11, 12, 13, 14, 15.
Moreover, they have failed to identify the weakest points of converters, thus not suitable for reliability reinforcement 8.
However, these approaches suffer from poor accuracy. In the conventional power engineering, the reliability is modeled based on historical data provided by the system operator or handbooks such as MIL 217 7. Therefore, reliability modeling in modern power electronic based power systems is of paramount importance. This is due to the fact that the power electronic converters contain failure-prone parts such as power devices which are also prone to wear-out failures according to the field experience 6. However, weak long-term performance of a converter will introduce more operational and maintenance costs compared to the manufacturing costs 4, 5. So far, the short-term characteristics of converters, such as fault ride through capability, and voltage/frequency/power quality support, have been taken into account for design and manufacturing, which are also covered by some standards such as IEC 62109 2 and IEEE 1547 3. Long-term performance and reliability of a power electronic converter have gained increasing interest for optimal design and operation of power electronic systems 1. Numerical case studies evaluate the effectiveness of the proposed reliability modeling approach. Moreover, the proposed lifetime curves can present the long-term performance of converters facilitating optimal system-level design for reliability, reliable operation and maintenance planning in power electronic systems. The proposed approach can quickly predict the converter lifetime under given operating conditions without a further need for extended, time-consuming electro-thermal analysis. Unlike the state-of-the-art theoretical reliability modeling approaches, which employ detailed electro-thermal characteristics and lifetime models of converter components, the proposed method provides a nonparametric surrogate model of the converter based on limited non-linear data from theoretical reliability analysis. The converter reliability is represented by the proposed constant lifetime curves, which have been developed using Artificial Neural Network (ANN) under different operating conditions. This paper proposes a long-term performance indicator for power electronic converters based on their reliability.
0 kommentar(er)
