This paper proposes a sparse representation and fully connected neural network (SRFCNN) methodology for residential ac series arc fault identification. The SRFCNN method captures the features of idiographic signal using sparse coding and conducts intelligent feature learning and classification through neural network (NN). The methodology is elaborately structured by a pretreatment layer, a sparse representation layer, and a decision layer. The processing procedure includes the following main steps. A set of customized bases are first designed via dictionary learning to behave as precise feature descriptors rather than choosing a fixed dictionary. The pretreated input samples are accordingly transformed into a series of sparse coefficients and then fed into the NN for recognition. Experimental results indicate that the method works well in distinguishing the signature of the arcing state from that of normal state within a certain single load type, as well as recognizing the two states of various load types (resistive, inductive, capacitive, switching, etc.) all together. The general classification accuracy of every half-cycle signal in the 10 tested classes could reach 94.3% and be probably higher. The methodology's good generality and reliability for arc fault identification show its potential in other unsupervised signal feature recognition and fault detection.

The variety of arc fault induced by different load types makes residential series arc fault detection complicated and challengeable. This paper proposes a hybrid time and frequency analysis and fully-connected neural network (HTFNN) based method to identify series AC arc fault. The HTFNN method recognizes samples state by broadly classifying them into Resistive (Re) category, Capacitive-Inductive (CI) category, and Switching (Sw) category according to the fundamental frequency components of series current. In each category, separate fully-connected neural network (NN) with customized time and frequency indicators being the input is employed for fine class recognition and state identification. This method construction addresses the feature overlap among various cases by separate categories and enables suitable indicator selection within each category. Since the identification complexity in each category is cheaper than identifying all cases together, concise NNs are applied in this paper to reduce computing cost. The general accuracies of normal and arcing state identification in Re, CI and Sw category are 99.64%, 100%, and 98.45% respectively, based on 3950 test samples. Method comparison shows that HTFNN achieves higher detection precision at lower computational complexity for generalized load types. This paper also evaluates the feasibility of implementing the method in hardware for arc fault detection device application.

At metro stations with very large passenger volumes and severe congestion, random disturbances often occur, resulting in the original offline optimal timetable no longer applicable. Deep reinforcement learning has the advantages of self-learning and online learning, making it possible to solve the energy-aimed train timetable rescheduling (ETTR) problem under random disturbances. In this article, a deep reinforcement learning approach (DRLA) has been proposed and applied to the ETTR problem to reschedule the train timetable in order to achieve the online optimal timetable for the minimum energy usage under disturbances. The proposed DRLA has the advantages of a real-time performance, high learning efficiency, and better energy-saving effect compared with the traditional heuristic algorithms such as genetic algorithm (GA), deep learning methods such as the combination of improved GA and long short-term memory (IGSA-LSTM), or other reinforcement learning algorithms such as deep deterministic policy gradient (DDPG) algorithm. Various experiments are conducted to compare the performance of DRLA and other algorithms. The experimental results indicate that in the two-train metro system and the five-train metro system, DRLA can save an average of energy by 5.11% and 7.29% with an average reaction time of only 0.0013 and 0.0019 s against disturbances, respectively.

Metro systems are one of the most important transportation systems in people’s lives. Due to the huge amount of energy it consumes every day, highly-efficient operation of a metro system will lead to significant energy savings. In this paper, a new integrated Energy-efficient Operation Methodology (EOM) for metro systems is proposed and validated. Compared with other energy saving methods, EOM does not incur additional cost. In addition, it provides solutions to the frequent disturbance problems in the metro systems. EOM can be divided into two parts: Timetable Optimization (TO) and Compensational Driving Strategy Algorithm (CDSA). First, to get a basic energy-saving effect, a genetic algorithm is used to modify the dwell time of each stop to obtain the most optimal energy-efficient timetable. Then, in order to save additional energy when disturbances happen, a novel CDSA algorithm is formulated and proposed based on the foregoing method. To validate the correctness and effectiveness of the energy-savings possible with EOM, a real case of Shanghai Metro Line One (SMLO) is studied, where EOM was applied. The result shows that a significant amount of energy can be saved by using EOM.

Reinforcement learning has potential in the area of intelligent transportation due to its generality and real-time feature. The Q-learning algorithm, which is an early proposed algorithm, has its own merits to solve the train timetable rescheduling (TTR) problem. However, it has shortage in two aspects: Dimensional limits of action and a slow convergence rate. In this paper, a deep deterministic policy gradient (DDPG) algorithm is applied to solve the energy-aimed train timetable rescheduling (ETTR) problem. This algorithm belongs to reinforcement learning, which fulfills real-time requirements of the ETTR problem, and has adaptability on random disturbances. Superior to the Q-learning, DDPG has a continuous state space and action space. After enough training, the learning agent based on DDPG takes proper action by adjusting the cruising speed and the dwelling time continuously for each train in a metro network when random disturbances happen. Although training needs an iteration for thousands of episodes, the policy decision during each testing episode takes a very short time. Models for the metro network, based on a real case of the Shanghai Metro Line 1, are established as a training and testing environment. To validate the energy-saving effect and the real-time feature of the proposed algorithm, four experiments are designed and conducted. Compared with the no action strategy, results show that the proposed algorithm has real-time performance, and saves a significant percentage of energy under random disturbances.

Automatic Train Systems (ATSs) have attracted much attention in recent years. A reliable ATS can reschedule timetables adaptively and rapidly whenever a possible disturbance breaks the original timetable. Most research focuses the timetable rescheduling problem on minimizing the overall delay for trains or passengers. Few have been focusing on how to minimize the energy consumption when disturbances happen. In this paper, a real-time timetable rescheduling method (RTTRM) for energy optimization of metro systems has been proposed. The proposed method takes little time to recalculate a new schedule and gives proper solutions for all trains in the network immediately after a random disturbance happens, which avoids possible chain reactions that would attenuate the reuse of regenerative energy. The real-time feature and self-adaptability of the method are attributed to the combinational use of Genetic Algorithm (GA) and Deep Neural Network (DNN). The decision system for proposing solutions, which contains multiple DNN cells with same structures, is trained by GA results. RTTRM is upon the foundation of three models for metro networks: a control model, a timetable model and an energy model. Several numerical examples tested on Shanghai Metro Line 1 (SML1) validate the energy saving effects and real-time features of the proposed method.