This thesis proposes a method for accurate temperature estimation of thermally-aware power electronics systems. The duality between electrical systems and thermal systems was considered for thermal modeling. High dimensional thermal models present a challenge for online estimation. RC (resistor-capacitor) circuits that create a tradeoff between accuracy and complexity were used to simulate the dynamic thermal behavior of power electronics. The complexity of the thermal network was further reduced by applying a structure-preserving model order reduction technique. The reduced order thermal model was an RC circuit with fewer capacitors. Preserving the physical correspondence between the reduced order model and the physical system allows the user to use the reduced order thermal model in the sensor placement optimization process. The accuracy of the thermal estimates can be easily increased by increasing the number of sensors in the system. However, a large number of sensors increases the cost and complexity of the system. It might also interfere with the circuit design and create packaging problems. An optimal number and optimal placement of temperature sensors was found. The optimal sensor placement problem was solved by maximizing the trace of observability Gramian. The optimal number of temperature sensors was based on the state estimation error obtained from a Kalman filter. The dynamic thermal behavior of the power electronics systems was represented by a linear state space model by applying the conservation of energy principle. Therefore, assuming Gaussian noise, it is well-known that a Kalman filter is an optimal estimator for such systems. A continuous-discrete Kalman filter was used to estimate the dynamic thermal behavior of power electronics systems using an optimal number of temperature sensors placed at optimal locations. The proposed method was applied on 2-D and 3-D power electronics systems. Theoretical results were validated experimentally using IR thermal imaging and thermocouples. It was shown that the proposed method can accurately iii reconstruct the dynamic temperature profile of power electronics systems using a small number of temperature sensors. for Power Optimization of Electro-Thermal Systems (POETS). I feel so lucky to be part of this multidisciplinary research center that gave me the opportunity to work with extraordinary people from different disciplines and different institutions. Tables Table 2.1 Thermal-electric analogy used to model resistor-capacitor thermal models. ........... 6 Table 5.1 Dimesions for the inverter regions defined in Figure 5
Increasing the efficiency and density of power electronic systems (PESs) is an important objective for many high-impact applications, such as electric vehicle charging and aircraft electrification. Due to compactness and high heat dissipation, careful thermal monitoring of such PESs is required. Strategic placement of temperature sensors can improve the accuracy of real-time temperature distribution estimates. Enhanced temperature estimation supports increased power throughput and density because PESs can be operated in a less conservative manner while still preventing thermal failure. This article presents new methods for temperature sensor placement for 2- and 3-dimensional PESs that (1) improve computational efficiency (by orders of magnitude in at least one case), (2) support the use of more accurate evaluation metrics, and (3) are scalable to high-dimension sensor placement problems. These methods are tested via sensor placement studies based on a single-phase flying capacitor multi-level (FCML) prototype inverter. Information-based metrics are derived from a resistance-capacitance (RC) lumped parameter thermal model. Other more general metrics and system models are possible through the application of a new continuous relaxation strategy introduced here for placement representation. A new linear programming (LP) formulation is presented that is compatible with a particular type of information-based metric. This LP strategy is demonstrated to support an efficient solution of finely discretized large-scale placement problems. The optimal sensor locations obtained from these methods were tested via physical experiments. The new methods and results presented here may aid the development of thermally aware PESs with significantly enhanced capabilities.
Accurate temperature estimation of high density active power electronic systems is vital for dynamic thermal management. Accurate and reliable estimation is especially important in regions that are close to failure, either due to high temperature or significant materials or component sensitivity. Improved estimation can support lower safety factors and enhanced system performance. An investigation of optimal temperature sensor placement methods is presented here, focusing primarily on methods utilizing information-based metrics. In addition, physics-based metrics are explored in an initial study that may have the potential to be more closely aligned with overall system utility. Studies are based on a 2 kW, single-phase, seven-level, GaN-based inverter. A lumped-parameter reduced-order thermal model, developed in previous work, is used for real-time temperature estimation. A continuous relaxation of a 2D placement domain led to a novel linear programming formulation that supports solution of finely-discretized sensor placement problems with minimal computational expense. Improved sensor placement performance metrics account for multiple loading conditions and estimation accuracy with respect to failure prevention.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.