Heming Zhang

A model-driven approach to multidisciplinary collaborative simulation for virtual product development

The design and development of complex artifacts and systems is shifting towards a distributed and collaborative paradigm. The simulation environments for such a paradigm, therefore, need to take into account the cooperation between design teams, i.e. supporting multidisciplinary simulation in a distributed environment. However, current simulation tools can not fulfil this requirement as they have been developed to solve the specific problems from different disciplines. Although its already possible to perform multidisciplinary simulations by using several tools together, it is very difficult to implement it when these tools are distributed on the Internet. A solution which can support the integration of distributed simulation models at run-time is presented, involving a computational infrastructure and a high-level modelling approach. Specifically, the infrastructure is constructed by a novel combination of two distributed computing techniques to implement the synchronization of distributed models, as well as to ensure the interoperability at run-time. In addition, a model-driven approach is developed to bridge the high-level model of a simulation system and the infrastructure which implements this model. The solution is evaluated by making a comparison with other approaches, as well as by developing a prototype tool. Its shown in the evaluation that (1) it is viable to develop multidisciplinary simulations in a distributed environment using this solution; (2) the model-driven approach allows designers to focus only on the high-level structure of a design without getting concerned with the details of the infrastructure.

An integrated and collaborative approach for complex product development in distributed heterogeneous environment

Multidisciplinary collaborative modeling and simulation is an effective approach to verify the dynamic behaviour of a complex product such as an aerospace vehicle or a locomotive. The design environment of such products must cater for the requirement for geographical distribution and structural heterogeneity in the modern industrial context. The focus of our research is a solution for such a collaborative modeling and simulation environment. Our approach is to utilize the IEEE standard High Level Architecture for distributed simultaneous simulation execution and management. At the same time, Web services technology is employed to deploy and integrate functional models for the simulation. The method for model description in this framework is proposed. The techniques for simulation execution and service encapsulation are discussed. A prototype system is implemented and a case study of locomotive tilting component design is executed. We find that both High Level Architecture and Web services have their expertise for collaborative simulation and the integration of these two technologies does achieve better interoperability and reusability among heterogeneous simulation components in a distributed environment. The new approach has considerable promise for the efficient and successful complex product design.

Towards a collaborative modeling and simulation platform on the Internet

The design and development of complex engineering systems, e.g. mechatronics, is increasingly characterized as an interdisciplinary process, which entails the co-operative work of distributed teams. However, traditional simulation tools generally emphasize discipline-oriented purposes and centralized running, and cannot adequately support the collaborative simulation in a distributed environment. Although some tools have been developed to enable remote access to simulations, most of them focus on the sharing of information and the provision of simulations via a single and centralized program. The long-term aim of this research is to develop a collaborative modeling and simulation platform, which will provide an integrated environment for multi-disciplinary teams to create, share, and integrate simulation services on the Internet. A step towards such a platform is described which includes the infrastructures for distributed communication, the mechanism for run-time interaction, and the representation of a simulation system at the abstract level. In this paper, we will (1) formulate a collaborative simulation problem; (2) identify the factors influencing the performance of a simulation; (3) propose a method for run-time interaction to achieve optimal simulation accuracy; and (4) illustrate how the components serving different purposes are integrated. A case study is described, based on a prototype implementation of the proposed solution, to explore supporting collaborative simulation in an Internet-distributed environment.

A hybrid genetic algorithm for two-stage multi-item inventory system with stochastic demand

We study a two-stage, multi-item inventory system where stochastic demand occurs at stage 1, and nodes at stage 1 replenish their inventory from stage 2. Due to the complexity of stochastic inventory optimization in multi-echelon system, few analytical models and effective algorithms exist. In this paper, we establish exact stochastic optimization models by proposing a well-defined supply–demand process analysis and provide an efficient hybrid genetic algorithm (HGA) by introducing a heuristic search technique based on the tradeoff between the inventory cost and setup cost and improving the initial solution. Monte Carlo method is also introduced to simulate the actual demand and thus to approximate the long-run average cost. By numerical experiments, we compare the widely used installation policy and echelon policy and show that when variance of stochastic demand increase, echelon policy outperforms installation policy and, furthermore, the proposed heuristic search technique greatly enhances the search capacity of HGA.

A solution of multidisciplinary collaborative simulation for complex engineering systems in a distributed heterogeneous environment

This paper presents an integrated approach to multidisciplinary collaborative simulation for complex engineering systems. The formulized paradigm of multidisciplinary collaborative simulation for complex engineering systems is principally analyzed. An IEEE HLA and web services based framework is proposed to provide a heterogeneous, distributed and collaborative running environment where multidisciplinary modeling, running management and post-processing of collaborative simulation are undertaken. The mechanism of multidisciplinary collaborative modeling, disciplinary model transformation, and time-synchronized simulation advancement are studied in detail. A prototype with the functions of multidisciplinary modeling, running management and post-processing for collaborative simulations is developed, and a typical complex engineering system is chosen as a case study to demonstrate the effectiveness of this new approach towards collaborative simulation.

Using collaborative computing technologies to enable the sharing and integration of simulation services for product design

Nowadays simulation is commonly used in engineering design for verifying design concepts before physical prototypes are produced. The simulation of complex products such as mechatronics in general involves a synergy of multiple traditional disciplinary areas and entails the collaborative work of a multidisciplinary team. A need thus arises for supporting the effective and efficient integration of subsystem models at simulation runtime and in a distributed environment. These models are generally created using different simulation tools and depend on the inputs from each other to perform numerical integration. As such, many issues need to be addressed, e.g. system modeling, the use of computing technologies, and the runtime interaction between models. In this paper, a service-oriented paradigm is presented which is underpinned by collaborative computing technologies to enable the provision of simulation models as services as well as the integration of these services for performing simulation tasks in product design. As well as the implementation of such a paradigm, a method for the interaction between models is in particular developed to achieve high accuracy for the simulation of design problems involving the solving of system equations. Preliminary evaluation work shows that the proposed paradigm underpinned by collaborative computing technologies is viable and have great potential in supporting collaborative simulation development in industry and the method for interaction control successfully achieves better accuracy compared with traditional methods.

Combinative algorithms for the multidisciplinary collaborative simulation of complex mechatronic products based on major step and convergent integration step

Multidisciplinary collaborative simulation (MCS) is an important area of research in the domain of multidisciplinary design optimization (MDO). Although previous research for MCS has to some extent addressed some issues like using of multiple tools, integration stability, control of step size, data synchronization, etc, further work is still necessary to study how to achieve improved precision. A theoretical model is formulated to describe and analyze the integration process of MCS. A basic algorithm with equal major steps is proposed based on the model, along with two methods of implementation for the model, namely the serial method and the parallel method. A further algorithm based on convergent integration step is proposed, which has a more flexible strategy for run-time integration. The influence of interpolation techniques on simulation performance is studied as well. Simulations of the performance of various algorithms with different interpolation techniques are performed for both a simple numerical example and a complex mechatronic product. The novel algorithm based on convergent integration step, when used with a high-order interpolation technique, has better performance in terms of precision and efficiency. The innovation of this paper is mainly on the validation of high precision of the proposed convergent integration step algorithm.

Truncation error calculation based on Richardson extrapolation for variable-step collaborative simulation

Collaborative simulation is an effective approach to performing simulation analysis for complex systems by integrating models developed for different engineering disciplines. Collaborative simulation issues include the modeling of coupled multidisciplinary systems, and the simulation running time integration of these models that are solved parallelly. Estimation of the local truncation error of coupling models is the key to solve multidisciplinary collaborative simulation problem, which is actually used to restrict the simulation step. This paper presents a variable-step method based on Richardson extrapolation for calculating the local truncation error to solve collaborative simulation problem of multidisciplinary coupling models. Formulas for estimating the local truncation error are derived through theoretical analysis by using integration methods and interpolation technologies, respectively. The simulation experiments are illustrated to validate the accuracy and efficiency of proposed collaborative simulation algorithm in comparison with the usual combinative algorithm.

A Service-Oriented Approach for the Collaborative Simulation of Complex Engineering Systems

The design and development of complex engineering systems generally encompasses multi-disciplinary knowledge and entails systematic analysis at an early design stage. A paradigm known as collaborative simulation is introduced to address this characteristic: that is, decomposing the system into manageable subsystems and incorporating them at simulation run-time. However, many key limitations still remain in current collaborative simulation solutions, especially in the modern context of distributive and collaborative design. Our motivation in this research is to develop such a tool that can support the effective deployment and dynamic integration of simulation models distributed on the Internet. As a major implementation technology in industry, Web Services have advantages such as modularized system structure and universally available computing capability. We propose a service-oriented approach that treats a collaborative simulation system as loosely-coupled components, e.g. simulation models and simulation management, all of which are wrapped as independent Web Services. A prototype system is developed, based on which a simple case study is looked at. It is evidenced in this work that Web Services and service-oriented architecture have great potential in developing engineering software tools.