4 results about "Software verification and validation" patented technology

The deployment of large language model (LLM) and retrieval-augmented generation (RAG) systems in industrial automation environments present unique technical challenges that are not adequately addressed by conventional software validation approaches. A computing system can define a comprehensive architecture with cross-validation methodologies that can perform systematic detection of complex failure modes that might be missed by conventional testing approaches, while providing the reliability, safety, and real-time performance characteristics essential for successful deployment of LLM and RAG systems in industrial automation environments.

The invention discloses a fluid simulation platform for aerodynamic characteristic analysis of civil aircrafts, and belongs to the technical field of aerodynamic numerical simulation. The fluid simulation platform adopts a layered architecture and comprises a platform layer, a server side, a client side and a user layer, and the platform layer comprises software and hardware infrastructures; the server comprises a basic service layer and a data layer; the client comprises an application layer and an access layer, and the application layer comprises a pre-processing module, a solver module, a post-processing module and a verification and confirmation module; the access layer comprises an aviation special fluid simulation software interface; and the user layer comprises technicians and maintainers. According to the method, a software heterogeneous component packaging and embedding technology is selected, data formats, interface styles and other elements are uniformly defined, different types of fluid simulation components are integrated, and a civil aircraft aerodynamic characteristic analysis scheme which is coherent in use experience and complete in simulation process is provided for researchers or engineers under different working conditions.

This application relates to a method, apparatus, computer device, storage medium, and computer program product for verifying fuel rod software. The method involves acquiring a verification start command for the fuel rod software to be verified, obtaining software calculation data output by the software based on the start command, acquiring standard result data, determining verification parameters of the software based on the standard result data, and then sequentially extracting the software calculation data corresponding to each verification parameter and comparing it with the standard result data to obtain at least one comparison result. Based on the comparison results of each verification parameter, the verification result of the fuel rod software to be verified is obtained through classification and statistics. This method allows for automated software verification and validation, reducing human error and improving the efficiency of software verification and validation analysis.

The deployment of large language model (LLM) and retrieval-augmented generation (RAG) systems in industrial automation environments present unique technical challenges that are not adequately addressed by conventional software validation approaches. A computing system can define a comprehensive architecture with cross-validation methodologies that can perform systematic detection of complex failure modes that might be missed by conventional testing approaches, while providing the reliability, safety, and real-time performance characteristics essential for successful deployment of LLM and RAG systems in industrial automation environments.