Dynamic Robustness of Open-Source Project Knowledge Collaborative Network Based on Opinion Leader Identification

A large amount of semantic content is generated during designer collaboration in open-source projects (OSPs). Based on the characteristics of knowledge collaboration behavior in OSPs, we constructed a directed, weighted, semantic-based knowledge collaborative network. Four social network analysis indexes were created to identify the key opinion leader nodes in the network using the entropy weight and TOPSIS method. Further, three degradation modes were designed for (1) the collaborative behavior of opinion leaders, (2) main knowledge dissemination behavior, and (3) main knowledge contribution behavior. Regarding the degradation model of the collaborative behavior of opinion leaders, we considered the propagation characteristics of opinion leaders to other nodes, and we created a susceptible–infected–removed (SIR) propagation model of the influence of opinion leaders’ behaviors. Finally, based on empirical data from the Local Motors open-source vehicle design community, a dynamic robustness analysis experiment was carried out. The results showed that the robustness of our constructed network varied for different degradation modes: the degradation of the opinion leaders’ collaborative behavior had the lowest robustness; this was followed by the main knowledge dissemination behavior and the main knowledge contribution behavior; the degradation of random behavior had the highest robustness. Our method revealed the influence of the degradation of collaborative behavior of different types of nodes on the robustness of the network. This could be used to formulate the management strategy of the open-source design community, thus promoting the stable development of OSPs.

Physics-Informed Machine Learning for Degradation Diagnostics of Lithium-Ion Batteries

State of health (SOH) estimation of lithium-ion batteries has typically been focused on estimating present cell capacity relative to initial cell capacity. While many successes have been achieved in this area, it is generally more advantageous to not only estimate cell capacity, but also the underlying degradation modes which cause capacity fade because these modes give further insight into maximizing cell usage. There have been some successes in estimating cell degradation modes, however, these methods either require long-term degradation data, are demonstrated solely on artificially constructed cells, or exhibit high error in estimating late-life degradation. To address these shortfalls and alleviate the need for long-term cycling data, we propose a method for estimating the capacity of a battery cell and diagnosing its primary degradation mechanisms using limited early-life degradation data. The proposed method uses simulation data from a physics-based half-cell model and early-life degradation data from 16 cells cycled under two temperatures and C rates to train a machine learning model. Results obtained from a four-fold cross validation study indicate that the proposed physics-informed machine learning method trained with only 60 early life data (five data from each of the 12 training cells) and 30 high-degradation simulated data can decrease estimation error by up to a total of 9.77 root mean square error % when compared to models which were trained only on the early-life experimental data.