Model-based Data Interpretation Research Articles

Mechanistic modeling indicates rapid glyphosate dissipation and sorption-driven persistence of its metabolite AMPA in soil.

Residual concentrations of glyphosate and its main transformation product aminomethylphosphonic acid (AMPA) are often observed in soils. The factors controlling their biodegradation are currently not well understood. We analyzed sorption-limited biodegradation of glyphosate and AMPA in soil with a set of microcosm experiments. A mechanistic model that accounts for equilibrium and kinetic sorption facilitated interpretation of the experimental results. Both compounds showed a biphasic dissipation with an initial fast (up to Days 7-10) and subsequent slower transformation rate, pointing to sorption-limited degradation. Glyphosate transformation was well described by considering only equilibrium sorption. Model simulations suggested that only 0.02-0.13% of total glyphosate was present in the soil solution and thus bioavailable. Glyphosate transformation was rapid in solution (time required for 50 % dissipation of the total initially added chemical [DT50 ]=3.9min), and, despite strong equilibrium sorption, total glyphosate in soil dissipated quickly (DT50 =2.4d). Aminomethylphosphonic acid dissipation kinetics could only be described when considering both equilibrium and kinetic sorption. In comparison to glyphosate, the model simulations showed that a higher proportion of total AMPA was dissolved and directly bioavailable (0.27-3.32%), but biodegradation of dissolved AMPA was slower (DT50 =1.9h). The model-based data interpretation suggests that kinetic sorption strongly reduces AMPA bioavailability, leading to increased AMPA persistence in soil (DT50 =12d). Thus, strong sorption combined with rapid degradation points to low risks of glyphosate leaching by vertical transport through soil in the absence of preferential flow. Ecotoxicological effects on soil microorganisms might be reduced. In contrast, AMPA persists, rendering these risks more likely.

Methodology Maps for Model-Based Sensor-Data Interpretation to Support Civil-Infrastructure Management

With increasing urbanization and depleting reserves of raw materials for construction, sustainable management of existing infrastructure will be an important challenge in this century. Structural sensing has the potential to increase knowledge of infrastructure behavior and improve engineering decision making for asset management. Model-based methodologies such as residual minimization (RM), Bayesian model updating (BMU) and error-domain model falsification (EDMF) have been proposed to interpret monitoring data and support asset management. Application of these methodologies requires approximations and assumptions related to model class, model complexity and uncertainty estimations, which ultimately affect the accuracy of data interpretation and subsequent decision making. This paper introduces methodology maps in order to provide guidance for appropriate use of these methodologies. The development of these maps is supported by in-house evaluations of nineteen full-scale cases since 2016 and a two-decade assessment of applications of model-based methodologies. Nineteen full-scale studies include structural identification, fatigue-life assessment, post-seismic risk assessment and geotechnical-excavation risk quantification. In some cases, much, previously unknown, reserve capacity has been quantified. RM and BMU may be useful for model-based data interpretation when uncertainty assumptions and computational constraints are satisfied. EDMF is a special implementation of BMU. It is more compatible with usual uncertainty characteristics, the nature of typically available engineering knowledge and infrastructure evaluation concepts than other methodologies. EDMF is most applicable to contexts of high magnitudes of uncertainties, including significant levels of model bias and other sources of systematic uncertainty. EDMF also provides additional practical advantages due to its ease of use and flexibility when information changes. In this paper, such observations have been leveraged to develop methodology maps. These maps guide users when selecting appropriate methodologies to interpret monitoring information through reference to uncertainty conditions and computational constraints. This improves asset-management decision making. These maps are thus expected to lead to lower maintenance costs and more sustainable infrastructure compared with current practice.

A framework for occupancy detection and tracking using floor-vibration signals

In sensed buildings, information related to occupant movement helps optimize important functionalities such as caregiving, energy management, and security enhancement. Typical sensing approaches for occupant tracking rely on mobile devices and cameras. These systems compromise the privacy of building occupants and may affect their behavior. Occupant detection and tracking using floor-vibration measurements that are induced by footsteps is a non-intrusive and inexpensive sensing method. Detecting the presence of occupants on a floor is challenging due to ambient noise that may mask footstep-induced floor vibrations. In addition, spurious events such as door closing and falling objects may produce vibrations that are similar to footstep impacts. These events have to be detected and disregarded. Tracking occupants is complicated due to uncertainties associated with walking styles, walking speed, shoe type, health, and mood. Also, spatial variation in structural behavior of floor slabs adds ambiguity to the task of occupant tracking, which cannot be addressed using data-driven strategies alone. In this paper, a framework for occupant detection and tracking is developed. Occupant detection is carried out based on signal information. This method outperforms existing threshold-based methods. Support-vector-machine classifiers, trained with time and frequency-domain features, successfully distinguish footsteps from spurious events and determine the number of occupants walking simultaneously. A model-based data-interpretation approach is used for occupant tracking. Structural-mechanics models are used to identify a population of possible occupant locations and trajectories. Up to two occupants can be tracked by accommodating systematic bias and uncertainties from sources such as modeling assumptions and variability in walking gaits. A hybrid framework for occupant detection and tracking that combines model-free approaches for occupancy detection with structural behavior models for tracking is developed and tested on two full-scale case studies. These studies successfully validate the utility of the framework for buildings having sparse sensor configurations that measure floor vibrations.

Validating model-based data interpretation methods for quantification of reserve capacity

Optimal performance of civil infrastructure is an important aspect of liveable cities. A judicious combination of physics-based models with monitoring data in a validated methodology that accounts for uncertainties is explored in this paper. This methodology must support asset managers when they need to extrapolate current performance to meet future needs. Three model-based data-interpretation methodologies, residual minimization, Bayesian model updating and error-domain model falsification (EDMF), are compared according to their ability to provide accurate interpretations of monitoring data. These comparisons are made using a full-scale case study, a steel-concrete composite bridge in USA. Validation of data interpretation is carried out using cross-validation (leave-one-out and hold-out). A joint-entropy metric is used to evaluate the extent to which the data that is used for validation contains information that is independent of data used for interpreting structural behaviour. Once accurately updated and validated knowledge of structural behaviour is available, it is employed to make predictions of remaining fatigue-life of the bridge. Validated identification of structural behaviour helps ensure accurate predictions of capacity of bridges beyond their design lives. EDMF and a modified form of Bayesian model updating are analytically and numerically equivalent, while EDMF has several practical advantages. Both methods provide accurate identification and safe estimations of the remaining fatigue life of the bridge. Such enhanced understanding of structural behaviour leads to appropriate decisions regarding civil infrastructure assets.

Data-Interpretation Methodologies for Practical Asset-Management

Monitoring and interpreting structural response using structural-identification methodologies improves understanding of civil-infrastructure behavior. New sensing devices and inexpensive computation has made model-based data interpretation feasible in engineering practice. Many data-interpretation methodologies, such as Bayesian model updating and residual minimization, involve strong assumptions regarding uncertainty conditions. While much research has been conducted on the scientific development of these methodologies and some research has evaluated the applicability of underlying assumptions, little research is available on the suitability of these methodologies to satisfy practical engineering challenges. For use in practice, data-interpretation methodologies need to be able, for example, to respond to changes in a transparent manner and provide accurate model updating at minimal additional cost. This facilitates incremental and iterative increases in understanding of structural behavior as more information becomes available. In this paper, three data-interpretation methodologies, Bayesian model updating, residual minimization and error-domain model falsification, are compared based on their ability to provide robust, accurate, engineer-friendly and computationally inexpensive model updating. Comparisons are made using two full-scale case studies for which multiple scenarios are considered, including incremental acquisition of information through measurements. Evaluation of these scenarios suggests that, compared with other data-interpretation methodologies, error-domain model falsification is able to incorporate, iteratively and transparently, incremental information gain to provide accurate model updating at low additional computational cost.

A model-based data-interpretation framework for post-earthquake building assessment with scarce measurement data

Recent earthquake events throughout the world have once again exposed the vulnerability of buildings with respect to earthquakes. It is unlikely and unsustainable to design and - especially in regions with low-to-moderate seismic hazard – to retrofit all buildings to remain within elastic displacement ranges during earthquakes with high return periods. Therefore, post-earthquake assessment plays a fundamental role in the resilience of cities, given the potential to reduce time between an earthquake event and the clearance for (renewed) occupancy of a building. In this paper, a framework for model-based data interpretation of measurements of earthquake-damaged structures is presented. The framework allows engineers to combine ambient-vibration measurements and visual inspection to reduce parametric uncertainty of a high-fidelity model using the error-domain model-falsification methodology. For building types that have limited stiffness contributions from non-structural elements (i.e. shear-wall buildings) and for which non-ductile failure modes (such as out-of-plane failure) can be excluded, reduction in natural frequency and damage grades derived from visual inspection provide global measurement sources for structural identification. The application of the proposed methodology to a shear-resisting building tested on a shake table illustrates that vulnerability-curve predictions provide accurate damage estimates for subsequent earthquakes with probabilities between 50% and 100% for five measured scenarios. In complete absence of baseline information regarding the initial building state and the earthquake signal, parametric uncertainty is reduced by up to 76%. This study thus demonstrates usefulness for certain building types to enhance post-seismic vulnerability predictions.

Rapid S-Curve Update Using Ensemble Variance Analysis With Model Validation

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 185630, “Rapid S-Curve Update Using Ensemble Variance Analysis With Model Validation,” by Jincong He, Shusei Tanaka, Xian-Huan Wen, and Jairam Kamath, Chevron, prepared for the 2017 SPE Western Regional Meeting, Bakersfield, California, USA, 23–27 April. The paper has not been peer reviewed. In the complete paper, the authors propose a novel method to rapidly update the prediction S-curves given early production data without performing additional simulations or model updates after the data come in. The approach has been successfully applied in a Brugge waterflood benchmark study, in which the first 2 years of production data [rate and bottomhole pressure (BHP)] were used to update the S-curve of the estimated ultimate recovery. To the authors’ knowledge, the proposed work flow, including the model validation and the denoising techniques, is novel. The proposed work flow is also general enough to be used in other model-based data-interpretation applications. Introduction As surveillance data are obtained from the field, the S-curves of the key metrics need to be updated accordingly. This is normally accomplished by a two-step approach. First, the data are assimilated through history matching to calibrate the model parameter uncertainties to obtain their posterior distributions. Then, a probabilistic forecast is performed on the basis of the posterior distributions of the parameters to update the S-curve of the key metrics. However, obtaining an S-curve update with the traditional approach can take weeks or months after the data come in. There is a need for rapid interpretation of the incoming data and update of the S-curve without going through a full-blown history-matching and probabilistic-forecast process. Recently, the approach called direct forecast (also called data-space inversion) has been a focus of attention. In direct forecast, the statistical relationship between the measurement data and the business objective is established on the basis of simulation-model responses before the data acquisition. This direct relationship can then be used to rapidly update the prediction of the objective once the data become available. A process called canonical functional component analysis is proposed to map the data and forecast variables into a low-dimensional space; multilinear regression was implemented in the reduced space to establish the data/objective relationship. The authors explore the use of a simpler and more-intuitive method called ensemble variance analysis (EVA) for rapid update of the S-curve. The idea of EVA is to explore covariance between data and objectives, then use an analytical formula to calculate the posterior S-curve. In the complete paper, the authors adapt the formulation for rapid S-curve update after data come in. While the direct forecast has attracted attention, less attention has been paid to validating the consistency between observed data and the simulation model after data come in. Blindly applying the direct-forecasting formula without identifying unmodeled features can lead to an incorrect posterior S-curve and misinformed decisions. The authors propose a procedure that detects and removes features in the measurement data that are inconsistent with the simulation responses. The complete paper describes formulation of the problem of updating S-curves with measurement data.

Interstitial IgG antibody pharmacokinetics assessed by combined in vivo- and physiologically-based pharmacokinetic modelling approaches.

For therapeutic antibodies, total tissue concentrations are frequently reported as a lump sum measure of the antibody in residual plasma, interstitial fluid and cells. In terms of correlating antibody exposure to a therapeutic effect, however, interstitial pharmacokinetics might be more relevant. In the present study, we collected total tissue and interstitial antibody biodistribution data in mice and assessed the composition of tissue samples aiming to correct total tissue measurements for plasma and cellular content. All data and parameters were integrated into a refined physiologically-based pharmacokinetic model for monoclonal antibodies to enable the tissue-specific description of antibody pharmacokinetics in the interstitial space. We found that antibody interstitial concentrations are highly tissue-specific and dependent on the underlying capillary structure but, in several tissues, they reach relatively high interstitial concentrations, contradicting the still-prevailing view that both the distribution to tissues and the interstitial concentrations for antibodies are generally low. For most therapeutic antibodies, the interstitium is the target space. Although experimental methods for measuring antibody pharmacokinetics (PK) in this space are not well established, thus making quantitative assessment difficult, the interstitial antibody concentration is assumed to be low. In the present study, we combined direct quantification of antibodies in the interstitial fluid with a physiologically-based PK (PBPK) modelling approach, with the aim of better describing the PK of monoclonal antibodies in the interstitial space of different tissues. We isolated interstitial fluid by tissue centrifugation and conducted an antibody biodistribution study in mice, measuring total tissue and interstitial concentrations in selected tissues. Residual plasma, interstitial volumes and lymph flows, which are important PBPK model parameters, were assessed in vivo. We could thereby refine the PBPK modelling of monoclonal antibodies, better interpret antibody biodistribution data and more accurately predict their PK in the different tissue spaces. Our results indicate that, in tissues with discontinuous capillaries (liver and spleen), interstitial concentrations are reflected by the plasma concentration. In tissues with continuous capillaries (e.g. skin and muscle), ∼50-60% of the plasma concentration is found in the interstitial space. In the brain and kidney, on the other hand, antibodies are restricted to the vascular space. Our data may significantly impact the interpretation of biodistribution data of monoclonal antibodies and might be important when relating measured concentrations to a therapeutic effect. By contrast to the view that the antibody distribution to the interstitial space is limited, using direct measurements and model-based data interpretation, we show that high antibody interstitial concentrations are reached in most tissues.

Measurement, Data Interpretation, and Uncertainty Propagation for Fatigue Assessments of Structures

The real behavior of existing structures is usually associated with large uncertainty that is often covered by the use of conservative models and code practices for the evaluation of remaining fatigue lives. To make better decisions related to retrofit and replacement of existing bridges, new techniques that can quantify fatigue reserve capacity are required. This paper presents a population-based prognosis methodology that takes advantage of in-service behavior measurements using model-based data interpretation. This approach is combined with advanced traffic and fatigue models to refine remaining fatigue-life predictions. Study of a full-scale bridge revealed that this methodology provides less conservative estimations of remaining fatigue lives. In addition, this approach propagates uncertainties associated with finite-element, traffic, and fatigue-damage models to quantify their effects on fatigue-damage assessments and shows that traffic models and structural model parameters are the most influential sources of uncertainty.

Improving simulation predictions of wind around buildings using measurements through system identification techniques

Wind behavior in urban areas is receiving increasing interest from city planners and architects. Computational fluid dynamics (CFD) simulations are often employed to assess wind behavior around buildings. However, the accuracy of CFD simulations is often unknown. Measurements can be used to help understand wind behavior around buildings more accurately.In this paper, a model-based data interpretation framework is presented to integrate information obtained from measurements with simulation results. Multiple model instances are generated from a model class through assigning values to parameters that are not known precisely, including those for inlet wind conditions. The information provided by measurements is used to falsify model instances whose predictions do not match measurements and to estimate the parameter values of the simulation. The information content of measurement data depends on levels of measurement and modeling uncertainties at sensor locations. Modeling uncertainties are those associated with the model class such as effects associated with turbulent fluctuations or thermal processes.The model-based data interpretation framework is applied to the study of the wind behavior around the buildings of the Treelodge@Punggol estate, located in Singapore. The framework incorporates modeling and measurement uncertainties and provides probability-based predictions at unmeasured locations. This paper illustrates the possibility to improve approximations of modeling uncertainties through avoiding falsification of the entire set of model instances. It is concluded that the framework has the potential to infer time-dependent sets of parameter values and to predict time-dependent responses at unmeasured locations.

A model-based data-interpretation framework for improving wind predictions around buildings

Although Computational Fluid Dynamics (CFD) simulations are often used to assess wind conditions around buildings, the accuracy of such simulations is often unknown. This paper proposes a data-interpretation framework that uses multiple simulations in combination with measurement data to improve the accuracy of wind predictions. Multiple simulations are generated through varying sets of parameter values. Sets of parameter values are falsified and thus not used for predictions if differences between measurement data and simulation predictions, for any measurement location, are larger than an estimate of uncertainty bounds. The bounds are defined by combining measurement and modeling uncertainties at sensor locations. The framework accounts for time-dependent and spatially-distributed modeling uncertainties that are present in CFD simulations of wind. The framework is applied to the case study of the CREATE Tower located at the National University of Singapore. Values for time-dependent inlet conditions, as well as values for the roughness of surrounding buildings, are identified with measurements carried out around the CREATE Tower. Results show that, on average, ranges of horizontal wind-speed predictions at an unmeasured location have been decreased by 65% when measurement data are used.

Augmenting simulations of airflow around buildings using field measurements

Computational fluid-dynamics (CFD) simulations have become an important tool for the assessment of airflow in urban areas. However, large discrepancies may appear when simulated predictions are compared with field measurements because of the complexity of airflow behaviour around buildings and difficulties in defining correct sets of parameter values, including those for inlet conditions. Inlet conditions of the CFD model are difficult to estimate and often the values employed do not represent real conditions. In this paper, a model-based data-interpretation framework is proposed in order to integrate knowledge obtained through CFD simulations with those obtained from field measurements carried out in the urban canopy layer (UCL). In this framework, probability-based inlet conditions of the CFD simulation are identified with measurements taken in the UCL. The framework is built on the error-domain model falsification approach that has been developed for the identification of other complex systems. System identification of physics-based models is a challenging task because of the presence of errors in models as well as measurements. This paper presents a methodology to estimate modelling errors. Furthermore, error-domain model falsification has been adapted for the application of airflow modelling around buildings in order to accommodate the time variability of atmospheric conditions. As a case study, the framework is tested and validated for the predictions of airflow around an experimental facility of the Future Cities Laboratory, called “BubbleZERO”. Results show that the framework is capable of narrowing down parameter-value sets from over five hundred to a few having possible inlet conditions for the selected case-study. Thus the case-study illustrates an approach to identifying time-varying inlet conditions and predicting wind characteristics at locations where there are no sensors.

Quantifying the Effects of Modeling Simplifications for Structural Identification of Bridges

Several long-span, prestressed, segmental box girder bridges were built in the early 1980s and many of them are affected by long-term residual deformations. Although some models have been proposed to describe their structural behavior, several uncertainties remain. This paper examines the effects of errors introduced by model simplifications on predicted values. The results are used to improve the estimation of parameter values using model-based data-interpretation strategies. The procedure is illustrated for the Grand-Mere Bridge located in Canada. This bridge is affected by excessive long-term vertical displacements. Model simplifications such as in its degree of complexity are found to have an important influence on prediction errors. Representing these errors by zero-mean independent Gaussian noise does not adequately describe the relationships among the errors observed in this case study. Estimated errors are used in the interpretation of the ambient vibration acceleration data recorded on the structure. The interpretation approach employed is based on error-domain model falsification. This study provides ranges of parameter values that can be used subsequently to characterize more accurately aspects such as long-term creep and shrinkage behavior.