Search

The concept of integrating physics-based and data-driven approaches has become popular for modeling sustainable energy systems. However, the existing literature mainly focuses on the data-driven surrogates generated to replace physics-based models. These models often trade accuracy for speed but lack the generalisability, adaptability, and interpretability inherent in physics-based models, which are often indispensable in the modeling of real-world dynamic systems for optimization and control purposes. In this work, we propose a novel architecture for generating model-integrated neural networks (MINN) to allow integration on the level of learning physics-based dynamics of the system. The obtained hybrid model solves an unsettled research problem in control-oriented modeling, i.e., how to obtain an optimally simplified model that is physically insightful, numerically accurate, and computationally tractable simultaneously. We apply the proposed neural network architecture to model the electrochemical dynamics of lithium-ion batteries and show that MINN is extremely data-efficient to train while being sufficiently generalizable to previously unseen input data, owing to its underlying physical invariants. The MINN battery model has an accuracy comparable to the first principle-based model in predicting both the system outputs and any locally distributed electrochemical behaviors but achieves two orders of magnitude reduction in the solution time.

Capacity knees have been observed in experimental tests of commercial lithium-ion cells of various chemistry types under different operating conditions. Their occurrence can have a significant impact on safety and profitability in battery applications. To address concerns arising from possible knee occurrence in battery applications, this work proposes an algorithm to identify capacity knees as well as their onset from capacity fade curves. The proposed capacity knee identification algorithm is validated on both synthetic degradation data and experimental degradation data of two different battery chemistries, and is also benchmarked to the state-of-the-art knee identification algorithm in the literature. The results demonstrate that our proposed capacity knee identification algorithm could successfully identify capacity knees when the state-of-the-art knee identification algorithm failed. The results can contribute to a better understanding of capacity knees and the proposed capacity knee identification algorithm can be used to, for example, systematically evaluate the knee prediction performance of both model-based methods, and data-driven methods and facilitate better classification of retired automotive batteries from safety and profitability perspectives.

Growing demands for electrification result in increasingly larger battery packs. Due to factors such as cell position in the pack and variations in the manufacturing process, the packs exhibit variations in the performance of their constituent cells. Moreover, due to the fixed cell configuration, the weakest cell renders the pack highly susceptible to these variations. Reconfigurable battery pack systems, which have increased control flexibility due to additional power electronics, present a promising solution for these issues. Nevertheless, to what extent they can prolong the battery lifetime has not been investigated. This simulation study analyzes the potential of dynamic reconfiguration for extending battery lifetime w.r.t. several parameters. Results indicate that the lifetime extension is larger for series than for parallel configurations. For the latter, the dominant factor is equivalent full cycles spread at the end of life, but resistance increase with age and the number of cells in parallel are also influential. Finally, for the former, the number of series-connected elements amplifies these effects., Comment: Accepted to the 25th European Conference on Power Electronics and Applications (EPE 2023 ECCE Europe)

Battery cycle life prediction using early degradation data has many potential applications throughout the battery product life cycle. For that reason, various data-driven methods have been proposed for point prediction of battery cycle life with minimum knowledge of the battery degradation mechanisms. However, managing the rapidly increasing amounts of batteries at end-of-life with lower economic and technical risk requires prediction of cycle life with quantified uncertainty, which is still lacking. The interpretability (i.e., the reason for high prediction accuracy) of these advanced data-driven methods is also worthy of investigation. Here, a Quantile Regression Forest (QRF) model, having the advantage of not assuming any specific distribution of cycle life, is introduced to make cycle life range prediction with uncertainty quantified as the width of the prediction interval, in addition to point predictions with high accuracy. The hyperparameters of the QRF model are optimized with a proposed alpha-logistic-weighted criterion so that the coverage probabilities associated with the prediction intervals are calibrated. The interpretability of the final QRF model is explored with two global model-agnostic methods, namely permutation importance and partial dependence plot.

A key problem in process control is to decide which inputs should control which outputs. There are multiple ways to solve this problem, among them using gramian based measures, which include the Hankel interaction index array, the participation matrix and the $\Sigma_{2}$ method. The gramian based measures however have issues with input and output scaling. Generally, this is resolved by scaling all inputs and outputs to have equal range. However, we demonstrate how this can result in an incorrect pairing and examine alternative methods of scaling the gramian based measures, using either row or column sums, or by utilizing the Sinkhorn-Knopp algorithm. The benefits of these scaling strategies are first illustrated by applying them to the control structure selection for a heat exchanger network. Then, to more systematically analyze the benefits of the scaling schemes, a multiple input multiple output model generator is used to test the different schemes on a large number of systems. This, along with implementation of automatic controller tuning, allows for a statistical comparison of the scaling methods. This assessment shows considerable benefits to be gained from the alternative scaling of the gramian based measures, especially when using the Sinkhorn-Knopp algorithm. The use of this method also has the advantage that the results are completely independent of the original scaling of the inputs and outputs., Comment: 20 pages, 2 figures