Research On The Establishment Of Unified Measurement Computing Framework

Measurement is a basic human activity.It is defined as a process of experimentally obtaining one or more quantity values that can reasonably be attributed to a quantity.Measurement computing means the computing problem involved in the process of performing the measurement operation,including the solution acquisition of the measurement result,the validity analysis of the measurement signal and the function expression acquisition of the measurement signal.In order to facilitate the learning and application of measurement knowledge,it is necessary to establish a measurement computing framework to effectively organize the existing measurement computing technique,and systematically study the relevant concepts,models and methods.However,in measurment practice,the specific situation is often ever-changing and the actual problems encountered are endless,so it is impossible to investigate the measurement computing for all scenarios and problems.It is necessary to establish a unified measurement computing framework that can perform computational tasks according to the standardized process,and can be applied to the general measurement scenario.This leads to the basic question of this thesis-how to build a measurement computing framework with standard organizational structure that is suitable for the general measurement scenario.To solve this issue,this thesis focuses on the following key problems:1.Computing problem for measurement operator.The purpose of measurement is to obtain the value of the quantity to be measured,and in practice the first to be observed is the measurement signal,such as the data collected via analog-to-digital converter(ADC).The task of the measurement operator is to extract the measurand from the collected data,but the potential measurement operator is massiveis,and its computing method and performance are also different.In addition,time and resource constraints for measurement require that the measurement operator can be implemented quickly and cheaply.Therefore,the computing problem for measurement operator is a key issue in the establishment of measurement computing framework.2.Computing problem for signal validity analysis.The basic principle of the measurement operator to obtain the measurand is to compare the samples with the signal model.However,not all signal models are feasible and effective.There are multiple measurement signals for the same quantity based on different physical laws,and there are pros and cons between them.The validity analysis is an important means to judge the validity of the signal model and to identify the optimal signal model,which needs to quantify some validity indicators: signal discrimination and signal sensitivity,etc.Therefore,the computing problem for signal validity analysis is a key issue in the establishment of measurement computing framework.3.Computing problem for signal model identificaiton.Measurement signal,as the carrier of the measurand,is defined as the quantity that is a function of the measurand.However,the function expression of the measurement signal is initially unknown and needs to be obtained by signal model identification.The detailed operation includes the selection of the function type,the estimation of the model parameters and the determination of the model order,etc.Therefore,the computing problem for signal model identification is a key issue in the establishment of measurement computing framework.To solve the above key technical problems,the thesis proposed a computing framework for measurement operator based on a universal operator,presented a computing framework for signal validity analysis focus on obtaining the values of signal discrimination and signal sensitivity,put forward a computing framework for signal model identification based on algebraic polynomial.These technical methods as solutions to the three key problems have been verified through rigorous theoretical analysis,which constitute a unified measurement computing framework with strong technical feasibility.It is conducive to improving the standardization and the specialization of measurement computing,and guiding the measurement practice.