In this paper, the stability analysis for parallel real-time digital simulation models is discussed. The coupling coefficient perturbation method and the simulation stepsize perturbation method are established. For tw...
详细信息
In this paper, the stability analysis for parallel real-time digital simulation models is discussed. The coupling coefficient perturbation method and the simulation stepsize perturbation method are established. For two classes of systems of test equations, we construct the parallel simulation models and prove that they have the stability behaviour which is similar to the original continuous systems.
This paper represents a physical model of two -dimensional transpiration cooling control system with surface heating and ablating. The governing equation is derived and the third boundary conditions are given. They co...
详细信息
This paper represents a physical model of two -dimensional transpiration cooling control system with surface heating and ablating. The governing equation is derived and the third boundary conditions are given. They constitute a two -dimensional variable -domain distributed parameter control system in which the control parameter appears in both the governing equation and the boundary conditions. Applying time semi-step alternating direction method to difference this mathematical model, the tridiagonal difference equations are obtained. For a flying shell of electromagnetic railgun, the mathematical simulation results agree with the realistic physical process, and the control parameter not only can control the temperature of heat shield, but also can control the boundary motion effectively.
In this paper, a mathematical model of real-time simulation is given, and the problem of convergence on real-time Runge-Kutta algorithms is analysed. At last a theorem on the relation between the order of compensation...
详细信息
In this paper, a mathematical model of real-time simulation is given, and the problem of convergence on real-time Runge-Kutta algorithms is analysed. At last a theorem on the relation between the order of compensation and the convergent order of real-time algorithm is proved.
The local analytical methods for digital simulation of the dynamical control system arepresented to overcome the difficulties due to the stiffness, the oscillation with high fre-quency and the discontinuity. The idea ...
详细信息
The local analytical methods for digital simulation of the dynamical control system arepresented to overcome the difficulties due to the stiffness, the oscillation with high fre-quency and the discontinuity. The idea of computation of the analogue computer is intro-duced into the computation of digital simulation, the system is decomposed into somesubsystems, each of which can run independently, and, with the input informations ap-proached to, the local analytical expressions for the solutions of these subsystems are cons-tructed. Finally, a simple analysis about the accuracy estimate and the stability of the algo-rithms are given. The computational results show that the CPU time expended on thedigital simulation of a flying control system is only 1/20 of that needed by the traditionalnumerical methods.
Recent advances in large language models (LLMs) have demonstrated significant potential for automated hardware description language (HDL) code generation from high-level specifications. However, two critical challenge...
详细信息
Recent advances in large language models (LLMs) have demonstrated significant potential for automated hardware description language (HDL) code generation from high-level specifications. However, two critical challenges limit further progress in this domain: the scarcity of quality Verilog training data and the inability of current approaches to generate RTL code optimized for power, performance, and area (PPA) *** paper presents a comprehensive data-centric framework that addresses these limitations through innovations in both pre-fine-tuning data preparation and after-fine-tuning optimization strategies. In the pre-fine-tuning phase, we tackle the data scarcity problem with an automated design-data augmentation framework that generates high-volume, high-quality natural language specifications aligned with corresponding Verilog code and EDA scripts. Our approach creates a complete RTL-level feedback loop by augmenting EDA scripts, RTL code, and EDA tool feedback. In the after-fine-tuning phase, we focus on generating PPA-aware RTL code through a novel search and prompt framework. Our approach implements iterative filtering and selection of LLM-generated Verilog variants while providing high-quality predefined prompts, including composition and interface *** evaluate the effectiveness of our data augmentation method, we fine-tune Llama 2-13B and Llama 2-7B models using the dataset generated by our augmentation framework. The results demonstrate a significant improvement in the Verilog generation tasks with LLMs. Moreover, the accuracy of Verilog generation surpasses that of the current state-of-the-art open-source Verilog generation model, increasing from 58.8% to 70.6% with the same benchmark. Our 13B model has a pass rate improvement compared with GPT-3.5 in Verilog generation and outperforms in EDA script (i.e., SiliconCompiler) generation with only 200 EDA script data. Additionally, to evaluate the effectiveness of the our agent framework,
The rapid advancements in big data and the Internet of Things (IoT) have significantly accelerated the digital transformation of medical institutions, leading to the widespread adoption of Digital Twin Healthcare (DTH...
详细信息
The rapid advancements in big data and the Internet of Things (IoT) have significantly accelerated the digital transformation of medical institutions, leading to the widespread adoption of Digital Twin Healthcare (DTH). The Cloud DTH Platform (CDTH) serves as a cloud-based framework that integrates DTH models, healthcare resources, patient data, and medical services. By leveraging real-time data from medical devices, the CDTH platform enables intelligent healthcare services such as disease prediction and medical resource optimization. However, the platform functions as a system of systems (SoS), comprising interconnected yet independent healthcare services. This complexity is further compounded by the integration of both black-box AI models and domain-specific mechanistic models, which pose challenges in ensuring the interpretability and trustworthiness of DTH models. To address these challenges, we propose a Model-Based Systems Engineering (MBSE)-driven DTH modeling methodology derived from systematic requirement and functional analyses. To implement this methodology effectively, we introduce a DTH model development approach using the X language, along with a comprehensive toolchain designed to streamline the development process. Together, this methodology and toolchain form a robust framework that enables engineers to efficiently develop interpretable and trustworthy DTH models for the CDTH platform. By integrating domain-specific mechanistic models with AI algorithms, the framework enhances model transparency and reliability. Finally, we validate our approach through a case study involving elderly patient care, demonstrating its effectiveness in supporting the development of DTH models that meet healthcare and interpretability requirements.
暂无评论