Verification And Validation Of Structure Analysis In Engineering

Computational modeling and simulation play important roles in structure analysis of engineering. They contribute to the design, evaluation, and operation of new structural forms and construction technologies. Many decisions are made based on modeling and simulation results in structure engineering. Verification and validation (V&V) are the process to ensure that the computational modeling and simulation are correct and reliable.Verification is the process to determine whether the modeling and simulation are accurate. Validation establishes how accurate a simulation represents the real-world structure by comparison with experimental data. Verification is the prerequisite for validation.Three models are used in verification and validation: conceptual model, mathematical model and computational model. Validation for conceptual model is the prerequisite for verification. The conceptual model is the idealized representation of the structure of engineering, which identifies the key mechanical and physical processes, the fundamental assumptions of the engineering. The mathematical model consists of the mathematical descriptions of the governing equations, initial and boundary conditions and constitutive equations for the mechanics represented in the conceptual model. The computational model is the numerical discretization of the mathematical model.Verification deals with the processes of assessing the fidelity of the mathematical model to the computational model. Verification includes code verification and calculation verification. Code verification focuses on numerical correctness of the algorithms as implemented in code. Calculation verification addresses the accuracy of a particular instance of the numerical solution of the mathematical model.Validation is to quantify confidence in the predictive capability of the model for its intended uses. As the basis for comparing features from experimental data with simulation results, Validation metric should be decided first. Validation experiments are performed to generate data for the purpose of assessing the accuracy of simulation results.6 cases were discussed in this paper: the construction of double track tunnel on Dapu road at Puxi area, 600 MW turbine-generator with three-section core, the large cylinder structure, the apron of Mi-171 helicopter, the foundation of Sluice, the south side of dam foundation for Donghai bridge. In each case, the definition of conceptual model, mathematical model and computational model is discussed. Manufactured solution is developed for code verification. The numerical correctness of the algorithms in code is discussed by comparing with simulate result. Through the particular instance, calculation verification is used to address the numerical error in simulation. Validation experiments or site monitoring are performed to generate data for the purpose of assessing the accuracy of simulation results.It has been proved that the overall process of V&V is feasible in this paper. The guide of V&V needs to be developed in the structure analysis of engineering. V&V will help us to assess whether the computational modeling and simulation is accurate and reliable for its intended use.