Simulation Modelling Practice and Theory最新文献

In the context of distributed driving six-wheel steering (DD-6WS) commercial vehicles, the integration of auxiliary steering systems and direct yaw moment control (DYC) is critical for improving maneuverability and stability while driving. However, the nonlinear dynamic characteristics of vehicles under high-speed conditions make it difficult to fully exploit the benefits of multi-subsystem functionality. To address this issue, a sub-regional linearization (SRL) theory is proposed that uses nonlinear tire dynamic data to accurately capture the dynamics of commercial vehicle models. Additionally, a nonlinear stability criterion (the Lyapunov exponent) and a Mixed-Logic Dynamic (MLD) approach are used to create an intervention mechanism for multiple subsystems. Furthermore, hysteresis control is incorporated to mitigate frequent subsystem interventions caused by minor fluctuations in state variables. The results of simulations across various speed ranges using the HYSDEL toolbox and MATLAB-Trucksim platform demonstrate that using SRL models significantly improves the lateral control stability of commercial vehicles at high speeds while effectively reducing the frequency of triggers for auxiliary systems through the successful implementation of MLD control at high or low speeds. An orderly and precise triggering logic solves challenges caused by coupling and conflicts in vehicle redundant control.

Software-Defined Networking (SDN) technology has emerged as a promising solution to guarantee high reliability in the Industrial Internet of Things (IIoT) ecosystem. Through SDN, both fault tolerance-based Route Protection (RP) and fault tolerance-based Route Restoration (RR) are available to provide traffic rerouting when a network failure occurs in IIoT. RR redefines routes dynamically based on the current network status. However, it increases significantly the recovery time, which is not suitable for the Quality-of-Service (QoS) requirements of IIoT. In contrary, RP ensures fast failover, but it cannot be updated when the network status changes until the timeout interval expires. To deal with these issues, we propose a Dynamic Route Protection (DRP) mechanism that recalculates and reinstalls new optimal link-disjoint routes in accordance with the change rather than awaiting the controller to retransmit new flow instructions. Moreover, DRP responds speedily to forward the data packets from the secondary route to the main route if it is repaired. To recover rapidly when the connection failure affects both a link on the main route and the link on the secondary route simultaneously, the DRP mechanism utilizes the strategy of caching the third route in the controller memory using the Dynamic Hash Table (DHT) structure. DRP considers the heterogeneous traffic flows such as either delay-sensitive or both delay-sensitive and loss-sensitive. Again, this paper introduces a candidate fault tolerance architecture for Software-Defined IIoT (SDIIoT) that decouples IIoT networks into three functional layers. The results from the simulation network and the experimental hardware testbed illustrated that the DRP mechanism outperforms the FT-RP, RR, LFR, Pro-VLAN, and SDNRMbw mechanisms by minimizing the failure recovery time, end-to-end delay, packet violation rate, packet loss rate, and the time required to reuse the main route when it is repaired, while maximizing the packet delivery ratio.

Real-time e-business applications are vital for operational efficiency, but connectivity challenges persist, particularly in remote or crowded areas. Drone Base Station (DBS) architecture, proposed for Beyond fifth Generation (B5G) and Sixth Generation (6G) multi-cell networks, offers on-demand hotspot coverage, addressing connectivity gaps in remote or crowded environments. DBSs provide a promising solution to meet the demanding requirements of high data rates, real-time responsiveness, low latency, and extended network coverage, particularly for real-time e-business applications. A critical challenge in this context involves efficiently allocating the needed number of DBSs to the different hotspot service areas, referred to as cells, to optimize the operator’s total profit under unpredictable user demands, varying area-specific service costs, and price dependence real-time e-service. The objective is to achieve the highest total revenue while minimizing the cost (cost savings) throughout the multi-cell system. This challenge is formulated as a profit-maximization discount return problem, integrating the coverage constraint, the variable cell-dependent operational cost, the e-service-based price and the uncertain demands of users across cells. Traditional optimization methods fail due to environmental uncertainty, which leads to the need to reformulate the problem as a Markov Decision Problem (MDP). We introduce a cloud-based Reinforcement Learning (RL) algorithm for DBS dispatch to address the MDP formulation. This algorithm dynamically adjusts to uncertain per-cell user distributions, considering variable operating costs and service-dependent prices across cells. Through extensive evaluation, the RL-based dispatch approach is compared with reference drone dispatch algorithms, demonstrating superior performance in maximizing operator profit through cost savings by optimizing DBS dispatch decisions based on learned user behaviors, variable operational costs, and e-service types.

In the realm of cloud computing, effective resource allocation can significantly enhance the energy efficiency of datacenters. Task scheduling and Virtual Machine Placement (VMP) are two pivotal aspects of resource allocation. However, in current research, they are often treated separately, overlooking the potential for integrated optimization. In this paper, we propose an integrated solution for task scheduling and VMP in energy-efficient datacenters, based on queueing theory and Deep Reinforcement Learning (DRL) methods. This novel and comprehensive approach provides an alternative perspective for resource scheduling strategies in datacenters. We construct a queueing theory model for task scheduling, aiming to minimize the number of VMs that need to be instantiated, while ensuring that Service Level Agreement (SLA) violation remains at a low level. Furthermore, we design a VMP algorithm based on DRL for real-time selection of Physical Hosts (PHs) for deploying VMs. Finally, we conduct a simulation evaluation using a small-scale datacenter. The experimental results demonstrate that our method consistently ensures a lower rate of SLA violation. Compared to existing algorithms, the DRL-based VMP algorithm enables a more balanced utilization of the various resources in the PHs and reduces the total power consumption of the datacenter by more than 10% on average.

In this paper, the fretting wear behaviour of steel wires working in coal mining technology is studied numerically. In past studies, the fretting process of steel wires was carried out numerically considering that the coefficient of friction (COF) and wear coefficient (WC) are constant parameters. However, it has been noticed experimentally that COF increases up to a certain number of fretting cycles and then becomes constant, i.e. a steady-state stage, depending on the loading conditions. This increase in COF during the fretting process is also known as the running-up stage. The fretting wear model is modified to evaluate the influence of the varying coefficient of friction (VCOF), which is associated with the variable wear coefficient (VWC), so the influence of VWC is also considered. The subroutine UMESHMOTION used to implement the wear law is also modified to study the effect of VCOF and VWC. Therefore, in this study, the numerical results of a three-dimensional finite element (FE) model are compared, with analytical results of contact area and contact stresses, and with experimental results of peak wear depth. After validating the FE model, the wear scar, the increasing wear depth, wear volume, and the decreasing contact stress with increasing fretting cycles are determined numerically considering VCOF and VWC using cycle jump approach. The energy dissipation effect of frictional force and fretting amplitude is also studied for varying interaction properties of fretting wear models. The numerical simulations are performed by considering both elastic and plastic material properties to analyse the influence of varying interaction properties on fretting wear models at the running-up stage. The results indicate that the VWC model exhibits comparable impacts on both the elastic and plastic models. The results also show that the VWC fretting wear model leads to higher wear scar, wear volume, and wear depth values at the running-up stage as well as at the steady state stage, which are close to the experimental data.

Metal additive manufacturing (AM) technology has been utilized in many industries including automotive, aerospace, and medical. AM Ti6Al4V (Ti64) alloy is highly noticed for production of medical instruments such as dental implants and the machining process is mostly needed during the production or post-processing of these components. Numerical model, as a powerful tool, can be efficiently used for analyzing the machining process. A customized model was employed using a user-written subroutine in this work to evaluate machinability and microstructural changes in cryogenic machining of AM Ti64 alloy. For this purpose, the microstructural changes were simulated as the new numerical outputs. The numerical results of cutting forces, temperature, nano-hardness, and alpha lamellae thickness (grain size) were successfully verified by corresponding experiments from literature. Then, the impact of tool geometry (including rake and clearance angles, cutting edge radius, and nose radius) on the machinability performance was examined. It was found that, the variation of clearance and rake angles were more effective on depth of the hardened layer compared to the other parameters. Thickness of alpha lamellae phase near the machined surface and depth of the affected layer by nano-hardness changes were changed from 0.9 to 1.58 µm, and from 18 to 40 µm, respectively. Overall, it was concluded that the variation of insert positioning made by tool holder (change in rake and clearance angles) was an effective parameter on the process outputs when machining AM Ti64 alloy.

Set superset query is widely used in e-commerce processing and many other domains, particularly in cloud computing environments. Indexing is an efficient way to model e-commerce data. Many existing indexes, however, primarily focus on enhancing either query performance or space efficiency, often neglecting the need to strike a balance between these two factors. We have observed that upper nodes closer to the root of a tree are frequently accessed, while lower nodes near the leaves tend to entail expensive storage costs. To address this issue, we introduce TLI model, a trie and level-ordered unary degree sequence (LOUDS) hybrid model. The upper part of TLI is a trie, which is optimized for superior query performance. The lower part of TLI uses the LOUDS structure. TLI strikes a good balance between query performance and space utilization. To seamlessly integrate these two parts, we have developed efficient connecting strategies. Our simulation results on localhost demonstrate that TLI outperforms its competitors in terms of both space and time efficiency. Remarkably, it enhances query performance by up to 1.89 times, with a modest 6.72% increase in space overhead compared to LOUDS-based alternatives.

Pedestrian modeling and simulation has become an interesting topic that has gained scientists’ interest over the past few decades. While this field has yielded significant achievements in various applications, questions have been raised regarding the applicability of simulations in high-density scenarios. To this purpose, this paper comprehensively reviews pedestrian simulation models specifically focused on high-density situations. The review examines a total of 116 articles and categorizes their approaches for modeling pedestrian behaviors to different decisional levels. The strengths and limitations of these modeling approaches are compared and evaluated using different criteria for dense crowd simulations, such as their ability to simulate common emergent behaviors in crowded situations, performance, validation, and capacity to integrate into high-level modeling. Finally, the review provides potential directions for future research and development of dense crowd simulations.

As global shipping undergoes rapid expansion, pivotal waterway transport systems—including significant nodes like the Panama Canal, the Suez Canal, and the Three Gorges-Gezhouba dams—are increasingly emerging as system-wide bottlenecks that limit transportation capabilities. Recognizing the pressing need for efficient traffic organization at these critical junctures, we designed a hybrid simulation model, which integrates Cellular Automaton and Multi-Agent methods, to analyse traffic efficiency and evaluate different ship organization schemes at these key waterway nodes. The Three Gorges-Gezhouba dams serve as a case study, where we crafted and executed four simulation scenarios that accommodate a range of variables such as different traffic organization schemes, traffic flow volumes, and anchorage capacities. Key operational indicators such as the maximum average waiting time of ships at the anchorage, and the period when the anchorage along the waterway reaches saturation, provide insights into the system's operational condition. The simulation outcomes highlight the proposed model's capability to accurately quantify the impact of implementing a linkage-control scheme and underscore the utility of dynamic adjustment of water area ranges under linkage-control for managing various traffic scenarios. Consequently, our research not only enriches high-precision simulation methodologies but also bolsters decision-making processes concerning ship traffic organization at Waterway Transport Key Nodes.

Simulation plays a crucial role in transportation studies. However, most simulation tools are individually developed to tackle specific transportation problems, making it challenging to incorporate multiple simulation tools into a unified setting and generate collaborative output. In this study, we develop a co-simulation system that integrates MATSim with an external fleet-based simulator to extend MATSim's functionalities. The overall structure enables the integration of MATSim simulation and multiple external simulations, which results in a cohesive simulation output. Though only one external simulator engages in the current development, the framework can be easily adapted to involve more fleet-based simulators that meet the system requirements. As a result, more complex transportation systems can be simulated using the framework without the need to develop these dedicated MATSim extensions, e.g. any new fleet algorithm from emergent R&D. The developed co-simulation system is named the Fleet Demand (FD) Simulator. We demonstrate the functionality of the FD Simulator by showcasing a simulation scenario involving MATSim and a ride-pooling simulator, which integrates novel ride-pooling services into the MATSim environment. First, we show the co-simulation system's capability to generate reliable results consistent with those produced by using the "DRT" extension-enabled MATSim. Less than 10 % discrepancies between the two results are observed. We then use the FD Simulator to evaluate ride-pooling services under various scenarios, where we assign different service parameters to two service fleets. Operations of the two fleets are simulated in two separate external simulation environments, showcasing the FD simulator's ability of engaging multiple simultaneous simulations. The affected service parameters are not adjustable in the "DRT" extension, showing the advantage of the co-simulation system. By running these scenarios using the FD Simulator, travel decisions made by agents in MATSim are observed when facing heterogeneous ride-pooling services. The results highlight the relevance of the co-simulation system in evaluating complex transportation systems.