Data Reconciliation for Stationary Multi-Phase Multi-Stream Metallurgical Processes

The presence of measurement bias and random noise significantly deteriorates the information quality of plant data. Data reconciliation techniques for steady‐state processes have been widely applied to processing industries to improve the accuracy and precision of the raw measurements. This paper develops an algorithm for simultaneous bias correction and data reconciliation for dynamic processes. The algorithm considers process model error as an important contributing factor in the estimation of the measurement bias and process state variables. It employs black‐box models for the process as would be done when phenomenological models are difficult or impractical to obtain. Simulation results of a distillation column demonstrated that this algorithm effectively compensates constant and non‐constant measurement biases yielding much improved reconciled values of process variables. It has computational advantages over previously proposed algorithms based on non‐linear dynamic data reconciliation because an analytical solution is available when using linear process models to approximate the process.

Suboptimal flowrate estimators and their application to the design of measurement strategies

We present a model based algorithm for simultaneous validation of online analysers and process simulators. Reconciled on-line and historical process data satisfying balance and model equations provides the opportunity to validate and improve process models and soft sensors. Accurate simulation results and laboratory measurements can be used for the validation of online analysers. Validated and reconciled data can be used to the iterative and interactive identification of the unknown parameters of the simulator, e.g. for the determination of kinetic parameters. This method can also be used for monitoring and diagnostics of complex processes because situations when the operating conditions have been significantly changed can be discovered. The approach is illustrated by the analysis of an industrial hydrogenation system. We present the proposed iterative and interactive procedure of model development and analyser validation, the applied data reconciliation method, and the details of the case study. The results show the applicability of the proposed scheme in industrial environment and the benefits of the extracted information in the maintenance and monitoring of advanced model based process engineering tools. The developed tool can increase operating efficiency that is the key of reducing energy consumption and environmental impact. This is especially true in hydrocarbon industry where the operation of the technology is supported by process simulator and on-line analyser based advanced process control systems.

Both measurement and parameters uncertainties affect a parameter estimation process. To date measurement uncertainties have been considered during water quality parameter estimation in the water distribution systems (WDS). User defined parameter uncertainties have been avoided. It is known that some parameters in the distribution systems can be uncertain. For example, demand affects both the WDS hydraulics and water quality and is one of the most uncertain parameters. This paper considered nodal demand and temporal demand factor as the uncertain parameters when estimating wall decay coefficients of the water quality model. To account uncertainties in demand parameters hydraulic and water quality simulation model, EPANET was run in Monte Carlo Simulation (MCS) with predefined coefficients of variation for the uncertain parameters. Measurement uncertainties were added then with the mean concentration to create observed data. A meta-heuristic search technique, the Shuffled Frog Leaping Algorithm (SFLA) coupled with EPANET was used then to calibrate the wall decay parameters. Results show that parameters uncertainties can be as important as measurement uncertainties. Measurement uncertainty does not compensate parameters' uncertainties rather it increases the prediction errors.

The prelims comprise: Introduction Introductory Concepts for Validation of Plant Data Formulation Software Solution Integration in the Process Decision Chain Optimal Design of Measurement System An Example Conclusions

A robust two degree-of-freedom controller for systems with both model and measurement uncertainty

Propagation of uncertainty in nasal spray in vitro performance models using Monte Carlo simulation: Part I. model prediction using Monte Carlo Simulation

Propagation of Uncertainty in Nasal Spray In Vitro Performance Models Using Monte Carlo Simulation: Part II. Error Propagation During Product Performance Modeling

A subset of this work was presented at the 17th IFAC Symposium on Control, Optimization and Automation in Mining, Mineral and Metal Processing in 2016, at the fifth workshop on Mining, Mineral and Metal Processing in 2018, and the Ph.D. thesis of the main author.

Data reconciliation is technique for reducing measurement uncertainty by adjusting measured data to comply with a first-principles process model, most importantly with mass and energy balances. It also provides estimates for modelled unmeasurable process variables and estimates for the uncertainties of the computed values. For computing these estimates the process model has to include estimates of measurement uncertainties defined a priori. A priori consideration of all potential sources of uncertainty is far from trivial. This paper discusses a data-driven approach of uncertainty evaluation, based on identifying and subtracting variability modes affecting multiple measurements. Possible bias in the measurements is not considered. The approach is applied to evaluate the uncertainties of estimates computed with a data reconciliation model of a turbine section of a nuclear power plant.

The increase of measurement data reliability by the data reconciliation method is today an indispensable process in modern control systems of the operation of thermal installations. The results of measurements involve errors which are transmitted to errors of unknown values determined by means of mass and energy balance equations. Experiences from operation of the thermal systems show that assumed for data reconciliation calculations value of measurement uncertainty has significant influence on validation of the results of measurements and calculated values of indicators characterizing the operating conditions. The paper presents an analysis of the influence of the assumed measurement uncertainty on the results of the data reconciliation calculations and value of the specific consumption of chemical energy of fuels in the selected steam combined heat and power unit. Multivariate calculations for different values of assumed uncertainty of measurements and simulated limiting errors of measurements error were carried out. The results of these calculations to determine the sensitivity of the assumed measurements uncertainty on the specific consumption of chemical energy of fuels in the analyzed combined heat and power unit have been used.

Comminution is a frequently-required step in mineral processing and is responsible for almost 90% of all energy consumption in a mineral processing plant. Tumbling mill design has been studied since the middle of the XIX century. There are many comminution models in the literature, with preponderance, however, of Austin’s model (2002) for mineral impact breakage. In this paper, Austin’s model was applied to tubular tumbling mills. Once Austin’s model was proposed for batch processing of narrowly-distributed fraction sizes, an artifice has allowed it to be used in continuous grinding mill processes with widely-distributed fraction sizes. Interesting results were obtained with errors less than 0.005 for mills with sharp residence time distributions.

The Guide to the expression of uncertainty in measurement (GUM) has been the enduring guide on measurement uncertainty for metrologists since its first publication in 1993. According to the GUM, a measurement should always be accompanied by a reasoned and defensible expression of uncertainty, and the primary such expression is the standard uncertainty. In this article, we distinguish between the use of an expression of uncertainty as information for the recipient of a measurement result and its use when propagating uncertainty about inputs to a measurement model in order to derive the uncertainty in a measurand. We propose a new measure of uncertainty, the characteristic uncertainty, and argue that it is more fit for these purposes than standard uncertainty. For the purpose of reporting a measurement result, we demonstrate that standard uncertainty does not have a meaningful interpretation for the recipient of a measurement result and can be infinite. These deficiencies are resolved by the characteristic uncertainty, which we therefore recommend for use in reporting. For similar reasons, we advocate the use of the median estimate as the measured value. For the purpose of propagating uncertainty in a measurement model, we propose simple propagation of the median and characteristic uncertainty and show through some examples that this characteristic uncertainty framework is simpler and at least as reliable and accurate as the propagation of estimate, standard uncertainty and effective degrees of freedom according to the GUM uncertainty framework.

Modeling uncertainty in risk assessment: An integrated approach with fuzzy set theory and Monte Carlo simulation

Effect of redundancy on estimation accuracy in process data reconciliation