The analytical capabilities of industrial digital twins rely on the effective usage of intelligent measurement that can be of benefit when it is impossible or cost prohibitive to directly acquire certain data. Using intelligent instrumentation and simulation modelling to estimate complex system characteristics can be an effective way to address the challenge of acquiring imprecise, noisy or uncertain data. The paper presents the development of an intelligent measurement system in the industrial engineering context that is applied to ensuring subsea asset integrity. Its intelligence is achieved through the combined use of machine learning algorithms, the performance of which is evaluated and compared on a publicly available data related to the operation of an oil well - 3W dataset. On the basis of the research and experimental studies completed within the scope of the presented paper, a positive conclusion is drawn regarding the viability and appropriateness of using ML-based models for developing intelligent measurement systems to subsequently integrate them into industrial digital twins. The proposed system has the generality to be applied across a wide range of problem domains requiring processing, analysis and interpretation of data obtained from heterogeneous resources.