•An adaptive soft sensor methodology is proposed for quality prediction of nonlinear and time-varying chemical processes.•A localization scheme is presented to construct concise local model set.•A ...novel selective ensemble learning strategy is used to enhance prediction accuracy.•Thorough performance investigation is presented through two chemical processes.
This paper proposes an adaptive soft sensing method based on selective ensemble of local partial least squares models, referring to as the SELPLS, for quality prediction of nonlinear and time-varying chemical processes. To deal with the process nonlinearity, we partition the process state into local model regions upon which PLS models are constructed, through a statistical hypothesis testing based adaptive localization procedure. Two main delightful advantages of this localization strategy are that, redundant local models can be effectively detected and deleted and the local model set can be easily augmented online without retraining from scratch. In addition, a local model weighting mechanism is proposed to adaptively differentiate the contributions of local models by explicitly quantifying their generalization abilities for the current process dynamics. Finally, the selective ensemble learning strategy combines partial local models instead of all available models through Bayesian inference, which is able to reach a good equilibrium between the prediction bias and variance. The proposed SELPLS based soft sensor is applied to a simulated continuous stirred tank reactor and a real-life industrial sulfur recovery unit. Extensive simulation results demonstrate the effectiveness of the proposed scheme in contrast with several state-of-the-art adaptive soft sensing approaches.
Practical data assimilation algorithms often contain hyper-parameters, which may arise due to, for instance, the use of certain auxiliary techniques like covariance inflation and localization in an ...ensemble Kalman filter, the re-parameterization of certain quantities such as model and/or observation error covariance matrices, and so on. Given the richness of the established assimilation algorithms, and the abundance of the approaches through which hyper-parameters are introduced to the assimilation algorithms, one may ask whether it is possible to develop a sound and generic method to efficiently choose various types of (sometimes high-dimensional) hyper-parameters. This work aims to explore a feasible, although likely partial, answer to this question. Our main idea is built upon the notion that a data assimilation algorithm with hyper-parameters can be considered as a parametric mapping that links a set of quantities of interest (e.g., model state variables and/or parameters) to a corresponding set of predicted observations in the observation space. As such, the choice of hyper-parameters can be recast as a parameter estimation problem, in which our objective is to tune the hyper-parameters in such a way that the resulted predicted observations can match the real observations to a good extent. From this perspective, we propose a hyper-parameter estimation workflow and investigate the performance of this workflow in an ensemble Kalman filter. In a series of experiments, we observe that the proposed workflow works efficiently even in the presence of a relatively large amount (up to 10
3
) of hyper-parameters, and exhibits reasonably good and consistent performance under various conditions.
Localization plays an important role in manufacturing the curved thin-walled parts, which can determine the distribution of machining allowance and has great influence on the manufacturing accuracy. ...The registration algorithm is the most efficient way to locate the billet in the machining coordinate system of the machine tool by computing a transformation matrix. However, the localization of the curved thin-walled parts is complicated and challenging since the billets are individual, the shape and location of which are unknown. This paper attempts to develop an adaptive localization approach with the constraints of the profile and thickness based on on-machine measured data. The framework for constrained adaptive localization approach is illustrated, in which on-machine measurement, registration, isometric mapping and allowance optimization are involved. The details of the on-machine measurement for both the profile inspection and the thickness measurement are presented. An isometric mapping method is employed to build the separate point pairs between the measured points and the nominal shape of the part. A constrained optimization algorithm for the machining allowance is performed iteratively until meeting the constraints of the profile and thickness tolerance ranges. Finally, a case study of constrained adaptive localization was carried out, the results of which confirm the validity of the proposed approach.
Reservoir models are often subject to uncertainties, which, if not properly taken into account, may introduce biases to the subsequent reservoir management process. To improve reliability and reduce ...uncertainties, it is crucial to condition reservoir models on available field datasets through history matching. There are different types of field data. Among others, production data are the most common choice, but they are subject to a major limitation of carrying relatively low value of information. On the other hand, inter-well tracer data have been shown to provide additional information about well-to-well connectivity and reservoir dynamics. However, jointly history matching production and inter-well tracer data still remains challenging due to the lack of a coherent quantitative workflow to fully integrate them. This work can be considered a step towards tackling this noticed problem. To this end, we propose a non-intrusive and derivative-free ensemble history matching workflow, in which reservoir models are more coherently conditioned on both production and inter-well tracer data with the help of a recently developed technique (correlation-based adaptive localization). The workflow is successfully implemented in the Brugge benchmark case. Our study indicates that the history matching algorithm matches the production data well, regardless of the presence or absence of the tracer data. Nevertheless, by including tracer data as an additional source of information, we are able to improve the quality of the estimated reservoir models, in terms of both improved data match and reduced model discrepancies. As such, the finding of this study can help to achieve a better understanding of the impacts of tracer data on history matching performance, and the proposed workflow could serve as a useful too for more proper uncertainty quantification, and more coherent utilization of different types of field data in real case studies in general.
•Integrated ensemble-based workflow for jointly history matching production and tracer data.•Uncertainty quantification aided by correlation-based adaptive localization.•Performance validation in a field-scale case study.•Demonstrated benefits of including tracer data into the joint history matching workflow.
This work aims to help improve the performance of an iterative ensemble smoother (IES) in reservoir data assimilation problems, by introducing a data-driven procedure to optimize the choice of ...certain algorithmic hyper-parameters in the IES. Generally speaking, algorithmic hyper-parameters exist in various data assimilation algorithms. Taking IES as an example, localization is often useful for improving its performance, yet applying localization to an IES also introduces a certain number of algorithmic hyper-parameters, such as localization length scales, in the course of data assimilation. While different methods have been developed in the literature to address the problem of properly choosing localization length scales in various circumstances, many of them are tailored to specific problems under consideration, and may be difficult to directly extend to other problems. In addition, conventional hyper-parameter tuning methods determine the values of localization length scales based on either empirical (e.g., using experience, domain knowledge, or simply the practice of trial and error) or analytic (e.g., through statistical analyses) rules, but few of them use the information of observations to optimize the choice of hyper-parameters. The current work proposes a generic, data-driven hyper-parameter tuning strategy that has the potential to overcome the aforementioned issues. With this proposed strategy, hyper-parameter optimization is converted into a conventional parameter estimation problem, in such a way that observations are utilized to guide the choice of hyper-parameters. One noticeable feature of the proposed hyper-parameter tuning strategy is that it iteratively estimates an ensemble of hyper-parameters. In doing so, the resulting hyper-parameter tuning procedure receives some practical benefits inherent to conventional ensemble data assimilation algorithms, including the nature of being derivative-free, the ability to provide uncertainty quantification to some extent, and the capacity to handle a large number of hyper-parameters. Through 2D and 3D case studies, it is shown that when the proposed hyper-parameter tuning strategy is applied to tune a set of localization length scales (up to the order of 103) in a parameterized localization scheme, superior data assimilation performance is obtained in comparison to an alternative hyper-parameter tuning strategy without utilizing the information of observations.
•A correlation-based automatic and adaptive localization (AutoAdaLoc) scheme re-formulated as a parameterized AutoAdaLoc (P-AutoAdaLoc) scheme.•Localization length scales in the P-AutoAdaLoc scheme treated as algorithmic hyper-parameters inherent to an iterative ensemble smoother (IES).•A Continuous Hyper-parameter OPtimization (CHOP) procedure designed to iteratively update an ensemble of algorithmic hyper-parameters (including localization length scale as a special case).•In comparison to the original AutoAdaLoc scheme, superior data assimilation performance achieved by the P-AutoAdaLoc scheme in numerical case studies.
The application of object detection technology has a positive auxiliary role in advancing the intelligence of bird recognition and enhancing the convenience of bird field surveys. However, challenges ...arise due to the absence of dedicated bird datasets and evaluation benchmarks. To address this, we have not only constructed the largest known bird object detection dataset, but also compared the performances of eight mainstream detection models on bird object detection tasks and proposed feasible approaches for model lightweighting in bird object detection. Our constructed bird detection dataset of GBDD1433-2023, includes 1433 globally common bird species and 148,000 manually annotated bird images. Based on this dataset, two-stage detection models like Faster R-CNN and Cascade R-CNN demonstrated superior performances, achieving a Mean Average Precision (mAP) of 73.7% compared to one-stage models. In addition, compared to one-stage object detection models, two-stage object detection models have a stronger robustness to variations in foreground image scaling and background interference in bird images. On bird counting tasks, the accuracy ranged between 60.8% to 77.2% for up to five birds in an image, but this decreased sharply beyond that count, suggesting limitations of object detection models in multi-bird counting tasks. Finally, we proposed an adaptive localization distillation method for one-stage lightweight object detection models that are suitable for offline deployment, which improved the performance of the relevant models. Overall, our work furnishes an enriched dataset and practice guidelines for selecting suitable bird detection models.
Localization is essential to effectively assimilate satellite radiances in ensemble Kalman filters. However, the vertical location and separation from a model grid point variable for a radiance ...observation are not well defined, which results in complexities when localizing the impact of radiance observations. An adaptive method is proposed to estimate an effective vertical localization independently for each assimilated channel of every satellite platform. It uses sample correlations between ensemble priors of observations and state variables from a cycling data assimilation to estimate the localization function that minimizes the sampling error. The estimated localization functions are approximated by three localization parameters: the localization width, maximum value, and vertical location of the radiance observations. Adaptively estimated localization parameters are used in assimilation experiments with the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) model and the National Oceanic and Atmospheric Administration (NOAA) operational ensemble Kalman filter (EnKF). Results show that using the adaptive localization width and vertical location for radiance observations is more beneficial than also including the maximum localization value. The experiment using the adaptively estimated localization width and vertical location performs better than the default Gaspari and Cohn (GC) experiment, and produces similar errors to the optimal GC experiment. The adaptive localization parameters can be computed during the assimilation procedure, so the computational cost needed to tune the optimal GC localization width is saved.
Plain Language Summary
Assimilation of satellite radiance observations is essential for numerical weather predictions, especially where conventional observations are limited. Satellite radiances effectively measure integrated quantities over an atmospheric column, but it is not straightforward to define the vertical location of such a nonlocal observation. This results in complexities for localizing the impact of radiance observations; however, localization is crucial to effectively assimilate satellite radiances. This study investigated an adaptive localization approach to estimate an effective vertical localization independently for each assimilated channel of every satellite platform. Three localization parameters, including the localization width, maximum value, and vertical location of the radiance observations, were examined. It is demonstrated that using the adaptive localization width and vertical location for radiance observations is more beneficial than also including the maximum localization value. The adaptive localization method performs similarly to the optimal Gaspri and Cohn (GC) function, but the adaptive localization has significant computational cost advantages because it does not require intensive tuning of the localization width like the GC localization function.
Key Points
An adaptive method is proposed to estimate effective vertical localizations independently for each channel of every satellite platform
Using adaptive localization width and vertical location for radiances is more beneficial than including the maximum localization value
The adaptive localization parameters outperform the default GC and produce similar results to the optimal GC but with less computations
Indoor localization is becoming critical to empower Internet of Things for various applications, such as asset tracking, geolocation, and smart cities. Wi-Fi-based indoor localization using received ...signal strength (RSS) has drawn much attention over the past decade because it does not require extra infrastructure and specialized hardware. It is well known that the localization accuracy using RSS is rather susceptible to the changing environment. Localization by fusing multiple fingerprint functions of RSS is a promising strategy to overcome the above drawback. However, the existing fusion techniques cannot make full use of the intrinsic complementarity among multiple fingerprint functions. It also fails to exploit the knowledge obtained in the offline phase and thus shows low accuracy in the complex environment. This paper proposes a knowledge aided adaptive localization (KAAL) approach by using a global fusion profile (GFP) to mitigate the above shortcomings. First, we propose a GFP construction algorithm by minimizing position errors over all fingerprint functions with weight constraints in the offline phase. Based on the knowledge from GFP and the trained multiple fingerprint models, we then derive two KAAL algorithms, namely, multiple function averaging and optimal function selection, to achieve highly accurate localization results. Experimental results demonstrate that our proposed localization approach is superior to the existing methods both in simulated and real environments.
One important aspect of successfully implementing an ensemble Kalman filter (EnKF) in a high dimensional geophysical application is covariance localization. But for satellite radiances whose vertical ...locations are not well defined, covariance localization is not straightforward. The global group filter (GGF) is an adaptive localization algorithm, which can provide adaptively estimated localization parameters including the localization width and vertical location of observations for each channel and every satellite platform of radiance data, and for different regions and times. This adaptive method is based on sample correlations between ensemble priors of observations and state variables, aiming to minimize sampling errors of estimated sample correlations. The adaptively estimated localization parameters are examined here for typhoon Yutu (2018), using the regional model WRF and a cycling EnKF system. The benefits of differentiating the localization parameters for TC and non-TC regions and varying the localization parameters with time are investigated. Results from the 6-h priors verified relative to the conventional and radiance observations show that the adaptively estimated localization parameters generally produce smaller errors than the default Gaspari and Cohn (GC) localization. The adaptively estimated localization parameters better capture the onset of RI and yield improved intensity and structure forecasts for typhoon Yutu (2018) compared to the default GC localization. The time-varying localization parameters have slightly advantages over the time-constant localization parameters. Further improvements are achieved by differentiating the localization parameters for TC and non-TC regions.
This paper presents MFAM (Multiple Frequency Adaptive Model-based localization method), a novel model-based indoor localization method that is capable of using multiple wireless signal frequencies ...simultaneously. It utilizes indoor architectural model and physical properties of wireless signal propagation through objects and space. The motivation for developing multiple frequency localization method lies in the future Wi-Fi standards (e.g., 802.11ah) and the growing number of various wireless signals present in the buildings (e.g., Wi-Fi, Bluetooth, ZigBee, etc.). Current indoor localization methods mostly rely on a single wireless signal type and often require many devices to achieve the necessary accuracy. MFAM utilizes multiple wireless signal types and improves the localization accuracy over the usage of a single frequency. It continuously monitors signal propagation through space and adapts the model according to the changes indoors. Using multiple signal sources lowers the required number of access points for a specific signal type while utilizing signals, already present in the indoors. Due to the unavailability of the 802.11ah hardware, we have evaluated proposed method with similar signals; we have used 2.4 GHz Wi-Fi and 868 MHz HomeMatic home automation signals. We have performed the evaluation in a modern two-bedroom apartment and measured mean localization error 2.0 to 2.3 m and median error of 2.0 to 2.2 m. Based on our evaluation results, using two different signals improves the localization accuracy by 18% in comparison to 2.4 GHz Wi-Fi-only approach. Additional signals would improve the accuracy even further. We have shown that MFAM provides better accuracy than competing methods, while having several advantages for real-world usage.
PDF
