ASME V&V 10-2019 pdf download.Standard for Verification and Validation in Computational Solid Mechanics.
2.2 General Concepts of V&V 2.2.1 Definitions. Some basic terms that provide the basis for the rest of this Standard include (a) Code. A code is the computer implementation of algorithms developed to facilitate the formulation and approx- imate solution of a class of problems. (b) Model. A model is the representation ofa system, phenomena, or process under specific physical conditions. The representation includes conceptual, mathematical, and computational models. (c) Simulation Results. Simulation results are raw or processed calculations obtained by running the computational model. (d) Verification and Validation. The terms “verification” and “validation” have been used interchangeably in casual conversation as synonyms for the collection ofcorroborative evidence. The definitions used in this Standard are largely consistent with those published by the DoD [ref (1)] and the AIAA [ref. (2)]. (1) Verification is the process of determining that a computational model accurately represents the underlying mathematical model and its solution. (2) Validation is the process ofdetermining the degree to which the model is an accurate representation ofcorre- sponding physical experiments from the perspective of the intended uses of the model. Additional terms that form part of the shared language for V&V as used herein are found in Mandatory Appendix I. In essence, verification entails gathering evidence to establish that the computational implementation of the math- ematical model and its associated solution are correct. Validation, on the other hand, entails comparing simulation outputs with experimental outputs to establish evidence that the appropriate and adequate models were used to answer the questions of interest and to quantify the uncertainties within the process. Validation is attained through meeting the criteria established specifically for determination of validation, i.e., acceptable agreement is obtained.
2.2.2 Objectives. The general objectives ofV&V are to assess the reliability ofthe computer software and numerical methods used in the simulation and assesses the accuracy of the simulation with respect to available experimental observations. The model builderconsiders the model validated forthose response quantities atthe experimentlocations within the parameter space once predetermined requirements for demonstration ofaccuracy are met. For the decision maker or other stakeholder, the intended use also defines limitations on the applicability of the model. An example of an intended use is to predict the response of a particular make and model of automobile in frontal impacts against a wall atspeeds up to 30 mph. Validation might consistofnumerically simulating the compaction ofthe front end and the acceleration ofthe occupant compartment to within 20% for tests at 10, 20, and 30 mph. The model couldthenbeusedto predictresponse quantities atotherscenarios across theparameterspace. Apredictedresponsefor other makes or models of automobiles, for higher speeds, or for rear-end or side collisions would be away from the locations atwhich the model was validated. Forsome ofthose cases, the predicted response would be veryfarawayfrom thevalidationpoints,whicharedescribedingreaterdetailinpara.2.3.Scenariosfartherawayfromthevalidationlocation within the parameter space typically have a higher level ofuncertainty and a corresponding lower level ofconfidence in their predicted response. Requirements for accuracy and predictive capability would have to reflect the separation between the validation points and the intended use point.ASME V&V 10 pdf download.