Energy-efficient Train Research Articles

Energy-Efficient Train Control: A Comparative Study Based on Permanent Magnet Synchronous Motor and Induction Motor

Energy-Efficient Train Control: A Comparative Study Based on Permanent Magnet Synchronous Motor and Induction Motor

Exploring deep echo state networks for image classification: a multi-reservoir approach

Echo state networks (ESNs) belong to the class of recurrent neural networks and have demonstrated robust performance in time series prediction tasks. In this study, we investigate the capability of different ESN architectures to capture spatial relationships in images without transforming them into temporal sequences. We begin with three pre-existing ESN-based architectures and enhance their design by incorporating multiple output layers, customising them for a classification task. Our investigation involves an examination of the behaviour of these modified networks, coupled with a comprehensive performance comparison against the baseline vanilla ESN architecture. Our experiments on the MNIST data set reveal that a network with multiple independent reservoirs working in parallel outperforms other ESN-based architectures for this task, achieving a classification accuracy of 98.43%. This improvement on the classical ESN architecture is accompanied by reduced training times. While the accuracy of ESN-based architectures lags behind that of convolutional neural network-based architectures, the significantly lower training times of ESNs with multiple reservoirs operating in parallel make them a compelling choice for learning spatial relationships in scenarios prioritising energy efficiency and rapid training. This multi-reservoir ESN architecture overcomes standard ESN limitations regarding memory requirements and training times for large networks, providing more accurate predictions than other ESN-based models. These findings contribute to a deeper understanding of the potential of ESNs as a tool for image classification.

Energy-saving operation in urban rail transit: A deep reinforcement learning approach with speed optimization

The energy consumption of urban rail transit plays a significant role in the operating costs of trains. It is particularly crucial to decrease the energy consumption of the traction power supply in subway systems, as it accounts for approximately half of the total energy consumption of the subway operating organization. To overcome the limitations of traditional real-time speed profile generation methods and the limited exploration capabilities of popular reinforcement learning algorithms in the speed domain, this paper presents the Energy-Saving Maximum Entropy Deep Reinforcement Learning (ES-MEDRL) algorithm. The ES-MEDRL algorithm incorporates Lagrange multipliers and maximum policy entropy as penalties to formulate a novel objective function. This function aims to intensify exploration in the speed domain, minimize train traction energy consumption, and ensure a balance between ride comfort, punctuality, and safety within the subway system. This leads to the optimization of speed profile strategies. To further reduce energy consumption, this paper proposes a secondary optimization strategy for the energy-saving speed profile. This approach involves trading acceptable travel time for improved energy efficiency. To validate the performance of the proposed model and algorithm, numerical experiments are conducted using the Yizhuang Line of the Beijing Metro. The findings demonstrate a minimum 20 % increase in energy efficiency with the ES-MEDRL algorithm compared to manual driving. This algorithm can guide energy-efficient train operations at the planning level.

Physics-informed deep learning to quantify anomalies for real-time fault mitigation in 3D printing

In 3D printing processes, there are many thermal stress related defects that can have a significant negative impact on the shape and size of the structure. Such anomalies in the heat transfer of the printing process need to be detected at an early stage. Understanding heat transfer is crucial, and simulation models can offer insights while reducing the need for costly experiments. Traditional numerical solvers for heat transfer can be complex to adapt to diverse printed part geometries, and their reliance on predefined mathematical models limits their flexibility. Our physics-informed deep learning (PIDL) approach eliminates the need for discretization, simplifying the analysis of complex geometries and enabling automation. The drawback of parametric PIDL is their scalability for high-dimensional problems. Computational time, energy and cost of training prevent real-time analysis. It often takes only a few seconds to print a single layer. We can show an energy efficient transfer and training strategy to reduce the computational effort of PIDL significantly. The approach is able to quantify relevant effects of thermal stresses and mitigate errors during selective laser melting (SLM). To this end, heat transfer is modelled, simulated and analysed using high-dimensional data obtained from printing experiments with different geometries of metal components. The proposed method is applied to the solving forward problem of heat transfer prediction. The governing results are based on the heat equation, which is integrated into a deep neural network (DNN).

Driving Strategy Using an Improved Ant Colony System for Energy-Efficient Train

Optimal energy-efficient train operation optimization is one of the widely studied areas in transportation science, which can significantly reduce energy consumption that accounts for a large proportion of operating costs. In order to adapt to the complex and changeable railway line conditions such as gradient, slope length, and speed limit and avoid the error in tracking speed curve, an optimal driving strategy decision-making (ODSD) model is proposed in this paper. The model considers the non-fixed sequence of driving regimes, and the regimes are directly selected in the discrete micro-subsegments of an equal time-division pattern. To solve this model efficiently, an improved ant colony system algorithm with the difference edges (ACSd) is proposed, which takes the heuristic effect of the difference between the best solutions of two adjacent iterations, i.e., “the difference edges,” into account. Additionally, energy-efficient heuristic factor and speed heuristic factor are presented to balance energy saving and speed. The results demonstrate that ACSd performs better than the basic ant colony system algorithm in solving traveling salesman problem (TSP) and provides more flexible driving strategies for the ODSD model.

Increasing realism in modelling energy losses in railway vehicles and their impact to energy-efficient train control

AbstractThe reduction of energy consumption is an increasingly important topic of the railway system. Energy-efficient train control (EETC) is one solution, which refers to mathematically computing when to accelerate, which cruising speed to hold, how long one should coast over a suitable space, and when to brake. Most approaches in literature and industry greatly simplify a lot of nonlinear effects, such that they ignore mostly the losses due to energy conversion in traction components and auxiliaries. To fill this research gap, a series of increasingly detailed nonlinear losses is described and modelled. We categorize an increasing detail in this representation as four levels. We study the impact of those levels of detail on the energy optimal speed trajectory. To do this, a standard approach based on dynamic programming is used, given constraints on total travel time. This evaluation of multiple test cases highlights the influence of the dynamic losses and the power consumption of auxiliary components on railway trajectories, also compared to multiple benchmarks. The results show how the losses can make up 50% of the total energy consumption for an exemplary trip. Ignoring them would though result in consistent but limited errors in the optimal trajectory. Overall, more complex trajectories can result in less energy consumption when including the complexity of nonlinear losses than when a simpler model is considered. Those effects are stronger when the trajectory includes many acceleration and braking phases.

Time-efficient Design of Energy-efficient Train Operation Strategy with Multi-fidelity Train Simulators

Time-efficient Design of Energy-efficient Train Operation Strategy with Multi-fidelity Train Simulators

Energy-Efficient Train Control with Onboard Energy Storage Systems considering Stochastic Regenerative Braking Energy

Energy-Efficient Train Control with Onboard Energy Storage Systems considering Stochastic Regenerative Braking Energy

Blockchain for energy efficiency training in the construction industry

The construction sector faces the urgent need to prioritize energy efficiency due to an increasingly stringent regulatory landscape in response to the climate change agenda. Central to this transition is the pivotal role of education in equipping professionals with the necessary knowledge and skills. Educational solutions have emerged as powerful tools for promoting awareness and interventions to mitigate climate change. This article provides a case study that highlights the successful utilisation of computer technology in delivering digital solutions to advance energy education and promote more informed energy practices in the construction industry. The utilisation of digital technologies can enhance collaborative efforts in energy efficiency training, which is of critical significance in ensuring the security, sovereignty, transparency, immutability, and decentralisation of interventions related to energy education. This paper presents a framework that utilises Blockchain technology to facilitate training labelling and authenticity based on smart contracts and mobile passports to provide a secure and efficient solution for the delivery of training and education in the energy domain. Our research examines the challenges and opportunities related to energy efficiency training within the construction industry. By integrating industry-specific insights, exemplifications, and case studies, we provide an in-depth understanding of the interconnection between energy efficiency education and digital solutions with the unique context of the construction industry. We underscore the importance of leveraging digital platforms as educational tools to foster a deeper understanding and adoption of energy-efficient practices. We demonstrate that educational solutions play a pivotal role in driving awareness and interventions for mitigating climate change, greatly empowering individuals and organizations to adopt energy-efficient practices and to address sustainability objectives.

The optimal solution to the energy-efficient train control in a multi-trains system–Part 2: the optimality and the uniqueness

When a train travels in a multi-trains system, the power flow of other trains and the track grades make up the spatial–temporal area (STA) for the train. Finding the optimal solution for the energy-efficient train control (EETC) problem in STA can help reduce the net energy consumption. This paper studies the analytic method to obtain the optimal solution. In Part 1, the algorithm for the problem was designed. The underlying structure of the algorithm is the connection between three optimal states through optimal feasible strategy. In Part 2, the optimality of the optimal feasible strategy is verified through a generalised local energy functional, and its uniqueness is proved based on the variational method. Additionally, we discuss the influence of external power on the optimal solution of the classical EETC problem. Case studies using data for a real freight railway line are given to illustrate our results.

The optimal solution to the energy-efficient train control in a multi-trains system-part 1: the algorithm design

When a train travels in a multi-trains system, the power flow of other trains and the track grades make up the spatial–temporal area (STA) for the train. Finding the optimal solution for the energy-efficient train control problem in STA can help reduce the net energy consumption. This paper studies the analytic method to obtain the optimal solution. In Part 1, we propose an algorithm specifically designed for this problem. The underlying structure of the algorithm is the connection between three optimal states through the optimal feasible strategy. We propose an algebraic method to calculate the optimal feasible strategy and discuss how it intersects with the speed limit. In Part 2, we will discuss the optimality and uniqueness of the optimal feasible strategy. Case studies using data from a real freight railway line are given to demonstrate the effectiveness of the proposed algorithm.

The Impact of the Level of Training of Airport Security Control Operators on the Energy Consumption of the Baggage Control Process

The article focuses on the study of the impact of the level of training of security control operators (SCOs) at airports on the energy consumption of the passenger baggage control process. With the constant growing emphasis on security at airports, the intensification of training processes for security personnel, especially those dealing with baggage control, has become very dynamic. An essential aspect in times of sustainable development is optimizing all kinds of processes (including training processes) to reduce energy consumption. The analysis of the demand for energy used to conduct this type of training and the impact of the operator’s training level on the energy consumption of the control process are entirely ignored and have not been the subject of research by scientists so far. Therefore, this is a research gap that the authors are trying to fill in this article. The impact of safety system operator training levels on ensuring optimal energy efficiency was critically analyzed. The added value of the article is the authors’ model assessing the influence of the level of training of the SCO on the energy consumption of the control process. The effects of the frequency, duration and level of operator training on energy consumption rates were investigated. The authors’ activities aimed to identify the most energy-efficient approaches to training without compromising its quality and, thus, the safety of passengers. The article discusses potential strategies for minimizing energy use and draws conclusions that can help airport administrations and training providers adopt sustainable and energy-efficient training practices.

SmartDeal: Remodeling Deep Network Weights for Efficient Inference and Training.

The record-breaking performance of deep neural networks (DNNs) comes with heavy parameter budgets, which leads to external dynamic random access memory (DRAM) for storage. The prohibitive energy of DRAM accesses makes it nontrivial for DNN deployment on resource-constrained devices, calling for minimizing the movements of weights and data in order to improve the energy efficiency. Driven by this critical bottleneck, we present SmartDeal, a hardware-friendly algorithm framework to trade higher-cost memory storage/access for lower-cost computation, in order to aggressively boost the storage and energy efficiency, for both DNN inference and training. The core technique of SmartDeal is a novel DNN weight matrix decomposition framework with respective structural constraints on each matrix factor, carefully crafted to unleash the hardware-aware efficiency potential. Specifically, we decompose each weight tensor as the product of a small basis matrix and a large structurally sparse coefficient matrix whose nonzero elements are readily quantized to the power-of-2. The resulting sparse and readily quantized DNNs enjoy greatly reduced energy consumption in data movement as well as weight storage, while incurring minimal overhead to recover the original weights thanks to the required sparse bit-operations and cost-favorable computations. Beyond inference, we take another leap to embrace energy-efficient training, by introducing several customized techniques to address the unique roadblocks arising in training while preserving the SmartDeal structures. We also design a dedicated hardware accelerator to fully utilize the new weight structure to improve the real energy efficiency and latency performance. We conduct experiments on both vision and language tasks, with nine models, four datasets, and three settings (inference-only, adaptation, and fine-tuning). Our extensive results show that 1) being applied to inference, SmartDeal achieves up to 2.44x improvement in energy efficiency as evaluated using real hardware implementations and 2) being applied to training, SmartDeal can lead to 10.56x and 4.48x reduction in the storage and the training energy cost, respectively, with usually negligible accuracy loss, compared to state-of-the-art training baselines. Our source codes are available at: https://github.com/VITA-Group/SmartDeal.

Research on the cooperative train control method in the metro system for energy saving

PurposeTo address the problem that the current train operation mode that train selects one of several offline pre-generated control schemes before the departure and operates following the scheme after the departure, energy-saving performance of the whole metro system cannot be guaranteed.Design/methodology/approachA cooperative train control framework is formulated to regulate a novel train operation mode. The classic train four-phase control strategy is improved for generating specific energy-efficient control schemes for each train. An improved brute force (BF) algorithm with a two-layer searching idea is designed to solve the optimisation model of energy-efficient train control schemes.FindingsCase studies on the actual metro line in Guangzhou, China verify the effectiveness of the proposed train control methods compared with four-phase control strategy under different kinds of train operation scenarios and calculation parameters. The verification on the computation efficiency as well as accuracy of the proposed algorithm indicates that it meets the requirement of online optimisation.Originality/valueMost existing studies optimised energy-efficient train timetable or train control strategies through an offline process, which has a defect in coping with the disturbance or delays effectively and promptly during real-time train operation. This paper studies an online optimisation of cooperative train control based on the rolling optimisation idea, where energy-efficient train operation can be realised once train running time is determined, thus mitigating the impact of unpredictable operation situations on the energy-saving performance of trains.

Energy efficiency training’s effect on fatigue scores of children with a cancer diagnosis†

Abstract Objective Fatigue is the most common complication of cancer treatment procedure during chemotherapy, leading to limited daily activities, mood swings, and ultimately impaired functional status. This study aimed to determine the effect of energy efficiency training on the fatigue scores of children with a cancer diagnosis. Methods In the present clinical trial study, 30 children with a cancer diagnosis were randomly assigned variously to the control and testing groups. The cancer-related fatigue questionnaire for children with cancer was employed as the data collection tool. The intervention of the testing group was conducted in 4 sessions of 45–60 min of face-to-face interaction, group sessions, and question-and-answer sessions. Meanwhile, routine interventions were employed in the control group. Quantitative variables were reported as mean and standard deviation, and qualitative variables as frequency and percentage. Stata 11 statistical software and a change score approach were used to analyze the data. Results The mean fatigue score of the testing group increased from 72.12 to 77.53. There was a statistically significant difference between the mean fatigue scores of the testing group before and after the intervention (P = 0.013), and the children’s fatigue score decreased. There was a statistically significant difference between the mean fatigue scores of the testing and control groups before and after the intervention (P = 0.001), and the fatigue score of the children in the testing group had decreased compared to the control group. Conclusions Energy efficiency techniques can be employed to increase energy and reduce fatigue during pediatric chemotherapy.

A 28-nm 8-bit Floating-Point Tensor Core-Based Programmable CNN Training Processor With Dynamic Structured Sparsity

Training deep/convolutional neural networks (DNNs/CNNs) requires a large amount of memory and iterative computation, which necessitates speedup and energy reduction, especially for edge devices with resource/energy constraints. In this work, we present an 8-bit floating-point (FP8) training the processor which implements: 1) highly parallel tensor cores (fused multiply–add trees) that maintain high utilization throughout forward propagation (FP), backward propagation (BP), and weight update (WU) phases of the training process; 2) hardware-efficient channel gating for dynamic output activation sparsity; 3) dynamic weight sparsity (WS) based on group Lasso; and 4) gradient skipping based on the FP prediction error. We develop a custom instruction set architecture (ISA) to flexibly support different CNN topologies and training parameters. The 28-nm prototype chip demonstrates large improvements in floating-point operations (FLOPs) reduction (7.3 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times$</tex-math> </inline-formula> ), energy efficiency (16.4 TFLOPS/W), and overall training latency speedup (4.7 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times$</tex-math> </inline-formula> ), for both supervised and self-supervised training tasks.

A Data-Driven Iterative Learning Approach for Optimizing the Train Control Strategy

The energy-efficient train control (EETC) problem is investigated in this paper. And a soft actor-critic-based method is proposed to optimize the train driving strategy. Firstly, EETC problem is converted to the inverse problem, i.e., minimizing the trip time of the journey with constant energy consumption. Based on the conversion, the EETC problem is reformulated as a finite Markov decision process which can be solved by deep reinforcement learning algorithms. Secondly, an optimization method based on the soft actor-critic (SAC) method is designed to calculate the optimal driving strategy of the train with introducing the reservoir sampling method. Finally, some case studies are conducted to verify the effectiveness and performance of the proposed method. Simulation results demonstrate that a good energy-saving performance can be achieved. In single interval, the SAC-based method can reduce about 1.65&#x0025; of the energy consumption compared with numerical method. And the energy consumption reduction can be extended to be 6.49&#x0025; when the proposed approach is applied in multiple intervals.

Direct multiple shooting for computationally efficient train trajectory optimization

Energy efficient train control has been an active field of research for several decades, with Pontryagin’s maximum principle and dynamic programming being the two most common approaches for the computation of optimal trajectories. In this paper, we detail how direct multiple shooting can be used in the same context and highlight its two main advantages: the flexibility in the problem formulation and the availability of computationally efficient open-source software. Using the proposed framework, which is made publicly available, we are able to solve particularly challenging train trajectory optimization problems within only a few seconds.

A data-driven bi-objective matheuristic for energy-optimising timetables in a passenger railway network

Energy-efficient train timetabling (EETT) is essential to achieve the full potential of energy-efficient train control, which can reduce operating costs and contribute to a reduction in CO2 emissions. This article proposes a bi-objective matheuristic to address the EETT problem for a railway network. To our knowledge, this article is the first to suggest using historical data from train operation to model the actual energy consumption, reflecting the different driving behaviours. The matheuristic employs a genetic algorithm (GA) based on NSGA-II. The GA uses a warm-start method to generate the initial population based on a mixed-integer program. A greedy first-come-first-served fail-fast repair heuristic is used to ensure feasibility throughout the evolution of the population. The objectives taken into account are energy consumption and passenger travel time. The matheuristic was applied to a real-world case from a large North European train operating company. The considered network consists of 107 stations and junctions, and 18 periodic timetables for 9 train lines. Our results show that for an entire network, a reduction up to 3.3% in energy consumption and 4.64% in passenger travel time can be achieved. The results are computed in less than a minute, making the approach suitable for integration with a decision support tool.

Hierarchical optimal control framework to automatic train regulation combined with energy-efficient speed trajectory calculation in metro lines

In high-density metro lines, frequent disturbances could lead to a domino effect of train delays if no adjustment strategy is imposed timely. In this paper, to enhance train timetable adherence and reduce energy consumption, we provide a hierarchical optimal control framework for the automatic train regulation problem with energy-efficient speed trajectory based on the model predictive control method. Specifically, by considering the dynamic passenger flows, the non-linear train regulation model is proposed at the higher level to enhance timetable adherence, while at the lower level the energy-efficient speed trajectories of the trains are generated on-line in a distributed manner. The interaction of real-time information exists between the higher level and the lower level, i.e., the upper model provides the train running time and the numbers of onboard passengers to the lower model, while the lower model feedbacks the updated train information (e.g., the actual arrival time) to the upper model, which provides the potential for the proposed framework to respond to disturbances at the operational stage. Moreover, a customized model predictive control method combined with Radau pseudospectral method (RPM) is designed to generate the energy-efficient train speed trajectory at the lower level in response to uncertain operational conditions. Numerical cases based on the Beijing Metro Yizhuang line demonstrate the effectiveness and robustness of our proposed hierarchical optimal control framework to deal with uncertain disturbances.