A global thermodynamic measurement data reconciliation model considering boundary conditions and parameter correlations and its applications to natural gas compressors

Abstract

With the improvement of digitalization and automation level of the thermodynamic systems, the stable and accurate data resources are indispensable for improving efficiency and reducing energy consumption to characterize the operation state of energy systems. Data reconciliation technology helps to improve the raw data quality by establishing the constraint equations and objective functions, which can decrease measurement uncertainty and identify gross errors to accurately estimate performance and economic indicators of thermodynamic systems. In this paper, a novel data reconciliation model considering the relative residual and variable correlation is proposed to weaken the influence of operating conditions and measurement uncertainty in the thermodynamic system on the performance evaluation parameters. Moreover, different objective functions are designed and compared to quantify the evaluation results. The effectiveness of the model is verified by natural gas compressor, which is a typical thermodynamic system. In the simulation experiments, the measurement data under different operating conditions are constructed by adding the Gaussian white noise to simulation data, and the reconciled results of three objective functions and three correlation functions are further analyzed and compared. The results indicate that the proposed model has higher accuracy and stability. In the application for field data, the accuracy of performance evaluation based on the reconciled parameters is higher than that of directly calculation by the raw data or other reconciliation models.