Accurate prediction of Remaining Useful Life (RUL) of lithium-ion battery plays an increasingly crucial role in the intelligent battery health management systems. The advances in Deep Learning introduce new data-driven approaches to this problem. This paper proposes an integrated deep learning approach for RUL prediction of lithium-ion battery by integrating autoencoder with Deep Neural Network (DNN). Firstly, we present a multi-dimensional feature extraction method with autoencoder model to represent battery health degradation. Then, the RUL prediction model based DNN is trained for multi-battery remaining cycle life estimation. The proposed approach is applied to the real dataset of lithium-ion battery cycle life from NASA, and the experiment results show that the proposed approach can improve the accuracy of RUL prediction.

In modern society, many infrastructures are interdependent owing to functional and logical relations among components in different systems. These networked infrastructures can be modeled as interdependent networks. In the real world, different networks carry different traffic loads whose values are dynamic and stem from the load redistribution in the same network and disturbance from the interdependent network. Interdependency makes interdependent networks so fragile that even a slight initial disturbance may lead to a cascading failure of the entire systems. In this paper, interdependencies among networks are modeled and a failure cascade process is studied considering their effects on failure propagation. Meanwhile, an in-process restoration strategy after the initial failure is investigated. The restoration effects depend strongly on the trigger timing, restoration probability and priority of the restoration actions along with the additional disturbances. Our findings highlight the necessity to decrease the large-scale cascading failure by structuring and managing an interdependent network reasonably.

Many real networks, such as infrastructure networks, are interdependent and their structure is influenced by spatial constraints. However, the existence of spatial constraints and dependency links makes the system more vulnerable to an initial random failure. In this paper, we model the interdependent network with the strength of embedding length ζ . We propose an effective recovery strategy which recovers the boundary of the failed nodes during the cascading of failures with probability γ . We find that without changing the transition type, both the range and the duration of the cascading failure can be decreased by increasing γ . Our model could be used to improve the robustness of real-world network systems.

Due to the rapid development of information technology, a huge and complex traffic network has been established across various sectors including aviation, aerospace, vehicles, ships, electric power, and industry. However, because of the complexity and diversity of its structure, the complex traffic network is vulnerable to be attacked and faces serious security challenges. Therefore, this paper innovatively proposes a traffic network resilience recovery method based on resilience trend forecasting. In this paper, the risk value is introduced into the analysis of network fault propagation process, and the Susceptible, Infectious, Recovered, Dead‐Risk (SIRD‐R) fault propagation model is established. The resilience model of traffic network, which encompasses real‐time resilience and overall resilience, is constructed through the integration of network resilience bearing capacity and resilience recovery capacity. Then, the resilience of complex traffic network is forecasted by using long short‐term memory network, and the resilience recovery strategy of complex traffic network based on forecasting is proposed. Finally, the effectiveness and scalability of the proposed method are demonstrated through experimental analysis conducted on a diverse range of complex traffic networks, affirming its applicability in real‐world scenarios.

Complex networks have been widely studied recent years, but most researches focus on the single, non-interacting networks. With the development of modern systems, many infrastructure networks are coupled together and therefore should be modeled as interdependent networks. For interdependent networks, failure of nodes in one network may lead to failure of dependent nodes in the other networks. This may happen recursively and lead to a failure cascade. In the real world, different networks carry different traffic loads. Overload and load redistribution may lead to more nodes’ failure. Considering the dependency between the interdependent networks and the traffic load, a small fraction of fault nodes may lead to complete fragmentation of a system. Based on the robust analysis of interdependent networks, we propose a costless defense strategy to suppress the failure cascade. Our findings highlight the need to consider the load and coupling preference when designing robust interdependent networks. And it is necessary to take actions in the early stage of the failure cascade to decrease the losses caused by the large-scale breakdown of infrastructure networks.