Error Uncertainty Research Articles

Prediction of compressive strength of high-volume fly ash self-compacting concrete with silica fume using machine learning techniques

The quality and composition of the components in Self-Compacting Concrete (SCC) determine its compressive strength; however, determining these complex relationships through traditional statistical methods is difficult. This study uses five state-of-the-art machine learning techniques, namely, Random Forest (RF), Adaboost, Convolutional Neural Network (CNN), K-Nearest Neighbor (KNN), and Bidirectional Long Short-Term Memory (BI-LSTM), to simulate the strength properties of high-volume fly ash self-compacting concrete (HVFA-SCC) with silica fume. A database with 240 SCC compressive strength tests that included a range of material proportions was put together in order to train these models. The primary constituents that were selected and taken into consideration as input variables were cement, fly ash, super-plasticizer, silica fume, coarse and fine aggregate, and age. Various statistical metrics, regression error characteristics curve, SHAP analysis, and uncertainty analysis were used to validate the models' effectiveness in predicting the compressive strength of HVFA-SCC with varying material compositions. The recommended RF model demonstrated the highest predictive accuracy of all the models that were analysed. While machine learning can predict compressive strengths, its opaque 'black-box' nature limits its use for SCC mix engineers. This study introduces an open-source Random Forest-based GUI to close this gap. This interface helps engineers make mix proportion decisions by accurately estimating SCC compressive strength under various test conditions.

Analysis of Long-Distance Geometric Error Measurement and Uncertainty Based on PSD Laser Collimation Principle

Due to the limitations of traditional geometric error measurement, the measurement accuracy of long-stroke geometric errors is generally not high and the operation is complicated. In response to the above situation, in this study, a geometric error measurement system is built with a laser beam as the reference line and 2D position sensitive detector as the photoelectric conversion device. The single measurement range is 40 m, and the measurement range is further expanded through the principle of segmented splicing. Using an ultra-long guide rail as the measurement object for straightness measurement, the experimental results are similar to those of a laser interferometer. The uncertainty analysis model was obtained through the analysis of quantity characteristics, and based on this, the variance synthesis theorem and probability distribution propagation principle were studied to form two uncertainty synthesis methods. The measurement evaluation results showed that the two methods were basically consistent. The work provided a reference method for the uncertainty evaluation of position-sensitive detector measurement systems in the future.

Small-sample linear profile error uncertainty assessment based on grey system

The uncertainty assessment of the profile error of the cam profile, as defined in the national standard method, is difficult to carry out under conditions of small sample size and absence of probability distribution assumptions. This paper proposes a small-sample assessment model for the uncertainty of the profile error based on grey system. Firstly, the coordinate transformation is conducted using Vector Alignment Method to reduce systematic errors, and the non-uniform rational B-splines curve interpolation is utilized to fit the cam profile curve and perform error assessment. Subsequently, based on the error assessment results, Grey Information Measurement Model (GIMM) for the uncertainty of the profile error in small samples is established. This model employs Grey Relational Analysis to eliminate outliers and evaluates the uncertainty of the profile error by solving grey correlation coefficients. Maximum-Minimum Information Measure Method is used to assess the optimal sample size. Finally, numerical experiments and experimental tests were conducted on the uncertainty of camshaft profile error in automobiles. A total of 15 sets of profile data were compared with Guide to the Representation of Uncertainty in Measurement (GUM) and Monte Carlo Method (MCM) under different sample sizes. The results showed that GIMM achieved evaluation with only 8 sets of data samples under small sample and poor information conditions, with an uncertainty of 0.6338 μm, compared to 0.6346 μm for GUM and 0.6391 μm for MCM. The acceptance rate of GIMM reached 95.2%. This model outperforms other methods, providing a simplified and reliable assessment of cam profile error uncertainty.

A computational efficient approach for distributionally robust unit commitment with enhanced disjointed layered ambiguity set

AbstractTo achieve the sustainable development of the society, renewable energy dominated power systems are gradually formed. However, the uncertainty of renewable power poses challenges for power system operations, such as balancing the load and generation in day‐ahead unit commitment problems. To address this issue, an enhanced disjointed layered ambiguity set is proposed in this paper to effectively capture the uncertainty of renewable generation forecast errors. On this basis, a two‐stage distributionally robust unit commitment optimization problem is formulated, which is computational challenge. To this end, a novel reformation approach and a worst‐case support set identification method are proposed to derive the optimal solution in a computationally efficient way. Simulation results demonstrate that the proposed ambiguity set brings lower operation costs of unit commitment while maintaining similar reliability compared to existing methods. In detail, the proposed ambiguity set achieves a reduction in operating costs of approximately 10.01%. The proposed novel reformation approach and solution method are also shown to have higher computation efficiency than the existing Benders decomposition method, resulting in a reduction in computation time to less than 3.78% of the original duration.

A complexity efficient penta-diagonal quantum smoothing filter for bio-medical signal denoising: a study on ECG

Extracting information-bearing signal from a noisy environment has been a practical challenge in both classical and quantum computing formalism, especially in critical signal processing applications. To filter out the effect of noise, we propose a quantum smoothing filter built upon quantum formalism-based circuits applied for electrocardiogram signal denoising. The proposed quantum filter is a conceptually novel framework with an advantage in computational complexity as compared to the existing classical filters, such as discrete wavelet transform and empirical mode decomposition, whereas it achieves similar performance metrics for the accuracy of the filter. Further, we exploit the penta-diagonal Toeplitz structure of the smoothing filter, which gives approximately 48%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$48\\%$$\\end{document} gate cost reduction for 10 qubit circuit compared to the standard Hamiltonian simulation without structure. The run-time complexity using the quantum matrix inversion technique for the structured matrix is given by O~κ2poly(logN)εP\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ilde{{\\mathscr {O}}}\\left( \\frac{\\kappa ^2 \ ext {poly}(\\log {N})}{\\varepsilon _P}\\right)$$\\end{document} for condition number κ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\kappa$$\\end{document} of the N×N\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$N\ imes N$$\\end{document} filter matrix within precision εP\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\varepsilon _P$$\\end{document}. Embedding fixed sparsity of the banded matrix, the quantum filter shows potentially better run-time complexity than classical filtering techniques. For the quantifiable research results of our work, we have shown several performance metrics, such as mean-square error and peak signal-to-noise ratio analysis, with a bound of error due to observation noise, simulation error and quantum measurement uncertainty.

Multi time scale optimal dispatch of distribution network considering load demand response

Considering the uncertainty of wind power output and load as well as the differences in the operating characteristics of various adjustable resources over different time scales, a multitime scale optimal dispatching model based on stochastic optimization and distributed robust optimization is designed for the distribution network. Firstly, the demand response of the user load side is divided into price-based load and incentive-based load, and price elasticity coefficients are introduced to describe it. Then, with the minimum operation cost of the distribution network, the loss cost, and the penalty cost of abandoned wind as the objective function, various active and reactive regulation elements as well as the system power flow formulas are taken as constraints. In the day-ahead and real-time stages, a stochastic optimization method based on scenarios and opportunity constraints is introduced. In the intraday stage, a self-regressive sliding window prediction model is used to predict wind power and load processing, and a Wasserstein distance-based prediction error uncertainty set is established to establish an intraday distributed robust rolling optimization model. The improved IEEE 33-node case is used for simulation verification, and the results show that the model can effectively reduce system costs and improve system operation stability.

Coordinated optimal operation of integrated electrical and transportation network considering source-load uncertainties in severe weather scenarios

The large-scale penetration of electric vehicles and the emerging dynamic wireless charging technology strengthen the coupling between distribution networks and transportation networks. However, the randomness of travels in the transportation network, the fluctuation of the output of clean energy such as wind power and photovoltaic, and the sudden change of weather will have a significant impact on the safe and stable operation of the two networks. In this context, based on the background of wireless charging technology, a cooperative optimal operation scheduling model of electric transportation integrated network is proposed, which considers the uncertainty of traffic demand and photovoltaic prediction error, and regards it as the uncertainty factor of load side and source side of power grid. The Information gap decision theory (IGDT) is introduced to model the bilateral uncertainty of the network, and the corresponding risk avoidance model and risk speculation model are obtained. Furthermore, considering the effect of inclement weather scenarios on the uncertain factors of road traffic and charging facilities, a modified integrated network model of electric traffic under inclement weather scenarios is proposed. Finally, a case study is carried out on a test system consisting of a 12-node transportation network and a 20-node power distribution network. The results show that the information gap decision theory can effectively deal with the changes of dual networks under multiple uncertain scenarios, form a risk decision-making model, and provide theoretical support for different decision-making methods.

XWaveNet: Enabling uncertainty quantification in short-term ocean wave height forecasts and extreme event prediction

To aid accurate short-term forecast of wave heights and extreme events prediction from buoy data, a deep learning approach is proposed. Wave heights are forecasted up to 72 h in advance from multivariate meteorological and oceanographic data, followed by quantification of uncertainty in the forecasted values, and estimation of exceedance probabilities for extreme event conditions. Using Monte Carlo dropout technique, network regularization and forecast uncertainty assessment were performed. To forecast wave heights, a stacked long short-term memory (LSTM) model was trained on historical buoy data augmented with positionally encoded features. Per comparative evaluation of different models, across the 72 h forecast horizon, positionally encoded temporal attributes resulted in a mean prediction error and forecast uncertainty of only 3 % and 13.5 %, respectively. Contrarily, without temporal feature augmentation mean error was 5 % and uncertainty, 15 %. Therefore, the former model was ensembled with a Random Forest model for exceedance probability estimation. The hybrid model yielded a 21.4 % better separability between the extreme and normal observations in the test data from a nearby buoy in the northern Gulf of Mexico than individual models. All the four extreme events in the test set were correctly predicted with 97 % mean accuracy. Further evaluation of the deep ensemble model on 5 other spatially diverse buoy datasets from the Gulf of Mexico, Atlantic and Pacific coasts, suggested excellent generalizability for short-term wave forecasts and extreme event prediction, with an average maximum error of just 5.2 % across all the 6 buoy test datasets and the entire forecast horizon.

Multi-camera joint calibration algorithm for guiding machining robot positioning considering ambient light and error uncertainty

Multi-camera guided robotic machining is widely used in the removal of flash and burrs on the complex automotive castings and forgings. The multi-camera calibration process, however, is easily affected by the ambient light and uncertain error, thereby resulting in the undesired robot positioning accuracy. To overcome the challenging problem, this paper proposes a multi-camera joint calibration (MCJC) algorithm by considering both the ambient light and error uncertainty to optimize and compensate the positioning error of the image-guided machining robot. The high dynamic range imaging is used at first to obtain stable calibration image data for countering the effects of ambient light changes. Then the uncertainty error in camera self-calibration process is removed by the confidence threshold filtering. The hand-eye calibration matrix is corrected finally using the deviation feedback optimization for positioning accuracy enhancement of the image-guided machining robot. Taking the automotive flywheel shell as the experimental object, the results demonstrate that the average absolute positioning error is controlled within 1.3 mm, the optimum can reach 0.43 mm, and the average relative positioning error is smaller than 0.5 mm. Compared with the existing state-of-the-art algorithms, the proposed algorithm shows high positioning accuracy and strong algorithm robustness, and can meet the requirements for the subsequent image-guided robotic machining.

Quantification of uncertainty in robot pose errors and calibration of reliable compensation values

Due to their inherent characteristics, robots inevitably suffer from pose errors, and accurate prediction is the key to error compensation, which facilitates the application of robots in high-precision scenarios. Existing studies almost follow the points-view, and the compensation effect depends entirely on the accuracy of the point prediction, which leads to overconfident prediction results. In order to quantify the pose errors uncertainty and achieve more accurate prediction, a method to quantify of uncertainty in robot pose errors and calibration of reliable compensation values is proposed in this paper. In the proposed method, a distribution-free joint prediction model is designed to realize the simultaneous prediction of points and uncertainty intervals. Based on this, the reliable compensation value calibration strategy is innovatively proposed. The proposed method is verified on five tasks including spatial motions, constant load and milling processing, showing accurate joint prediction capability and reliable accuracy improvement. In addition, through online compensation experiments, the pose errors are reduced by 90 %, which promotes the application of robots in higher-precision scenarios.

A home-lab to study uncertainties using smartphone sensors and determine the optimal number of measurements

We present a home-lab experimental activity, successfully proposed to our students during covid19 pandemic, based on state-of-the-art technologies to teach error analysis and uncertainties to science and engineering students. In the last decade the appearance of smartphones considerably affected our daily life. Thanks to their built-in sensors, this revolution has impacted in many areas and, in particular, the educational field. Here we show how to use smartphone sensors to teach fundamental concepts for science students such as any measurement is useless unless a confidence interval is specified or how to determine if a result agrees with a model, or to discern a new phenomenon from others already known. We explain how to obtain and analyse experimental fluctuations and discuss in relation with the Gaussian distribution. In another application we show how to determine the optimal number of measurements as a function of the standard error and the digital resolution of a given sensor.

Analytical performance specifications based on biological variation data- considerations, strengths and limitations.

Analytical performance specifications (APS) are typically established through one of three models: (i)outcome studies, (ii) biological variation (BV), or (iii) state-of-the-art. Presently, The APS can, for most measurands that have a stable concentration, be based on BV. BV based APS, defined for imprecision, bias, total allowable error and allowable measurement uncertainty, are applied to many different processes in the laboratory. When calculating APS, it is important to consider the different APS formulae, for what setting they are to be applied and if they are suitable for the intended purpose. In this opinion paper, we elucidate the background, limitations, strengths, and potential intended applications of the different BV based APS formulas. When using BV data to set APS, it is important to consider that all formulae are contingent on accurate and relevant BV estimates. During the last decade, efficient procedures have been established to obtain reliable BV estimates that are presented in the EFLM biological variation database. The database publishes detailed BV data for numerous measurands, global BV estimates derived from meta-analysis of quality-assured studies of similar study design and automatic calculation of BV based APS.

Dynamic calibration method for track geometry measurement system-a case study in China

With the rapid development of railway construction, the mileage of railway detection has increased dramatically, and railway companies have higher requirements for the repeatability and accuracy of track geometric dynamic detection data. Therefore, the track geometry measurement system needs to be calibrated to improve the measurement accuracy to ensure the safety of railway operation. However, the existing calibration methods of the track geometry measurement system can only realize the static calibration and lack the dynamic calibration method, which cannot calibrate the track geometry measurement system under the real operation state. Therefore, a dynamic calibration method for the track geometry measurement system is proposed in this paper. A real and reliable test environment is provided for the track geometry measurement system by presetting the track geometry irregularity and laying the rail dynamic displacement monitoring system on the calibration test line, and the dynamic and static differences of the track are analyzed based on the test environment. Based on the bisection method, this paper reduces the measurement error of the track geometry measurement system and analyzes the measurement repeatability of the track geometry measurement system. Finally, the error correction model of the track geometry measurement system is established, and the error analysis and uncertainty evaluation of the track geometry measurement system are carried out. The results show that the track geometry measurement system calibrated by the dynamic calibration method proposed in this paper meets the requirements of the measurement error for the gauge and cross level. This study can establish the value traceability relationship of the track geometry measurement system, realize the calibration and evaluation of the accuracy of the track geometry measurement system under the real operation state, and improve the reliability and accuracy of the track geometry dynamic detection results.

Uncertainty quantification in multiaxial fatigue life prediction using Bayesian neural networks

In practical engineering scenarios, unforeseen failures may result in significant costs and risks. Assessing the uncertainty related to multiaxial fatigue life prediction is crucial for engineering applications. However, establishing a reliable multiaxial fatigue life prediction model remains challenging due to uncertainties in physical models, material properties, and measurement data. This paper proposes an effective Bayesian Neural Network (BNN) model for quantifying the uncertainty in multiaxial fatigue life prediction. Fatigue parameters in the multiaxial fatigue life prediction model are used as inputs for the BNN. The BNN method utilizes No-U-turn Sampler (NUTS) and Automatic Differentiation Variational Inference (ADVI) to address the inference problem in the model and perform life predictions for unknown non-proportional loading paths and different materials. Validation using experimental data from six different materials confirms the effectiveness of the BNN model. Experiments show that the performance of BNN models is beneficial in most cases compared to classical ML models, assessed through different error standards, statistical tests, Taylor diagrams, uncertainty analysis, and scatter index analysis. The proposed method provides a clear understanding of the uncertainty in multiaxial fatigue life prediction, offering a promising approach for simulating fatigue life under multiaxial loading in engineering applications.

25-hydroxyvitamin-D3 analysis with high-performance liquid chromatography in terms of total analytical error and measurement uncertainty

25-hydroxyvitamin-D3 analysis with high-performance liquid chromatography in terms of total analytical error and measurement uncertainty

SatERR: A Community Error Inventory for Satellite Microwave Observation Error Representation and Uncertainty Quantification

Abstract Satellite observations are indispensable for weather forecasting, climate change monitoring, and environmental studies. Understanding and quantifying errors and uncertainties associated with satellite observations are essential for hardware calibration, data assimilation, and developing environmental and climate data records. Satellite observation errors can be classified into four categories: measurement, observation operator, representativeness, and preprocessing errors. Current methods for diagnosing observation errors still yield large uncertainties due to these complex errors. When simulating satellite errors, empirical errors are usually used, which do not always accurately represent the truth. We address these challenges by developing an error inventory simulator, the Satellite Error Representation and Realization (SatERR). SatERR can simulate a wide range of observation errors, from instrument measurement errors to model assimilation errors. Most of these errors are based on physical models, including existing and newly developed algorithms. SatERR takes a bottom-up approach: errors are generated from root sources and forward propagate through radiance and science products. This is different from, but complementary to, the top-down approach of current diagnostics, which inversely solves unknown errors. The impact of different errors can be quantified and partitioned, and a ground-truth testbed can be produced to test and refine diagnostic methods. SatERR is a community error inventory, open-source on GitHub, which can be expanded and refined with input from engineers, scientists, and modelers. This debut version of SatERR is centered on microwave sensors, covering traditional large satellites and small satellites operated by NOAA, NASA, and EUMETSAT.

Probing O2-dependence of cyclopentyl reactions via isomer-resolved speciation

Modeling chemical kinetics relevant to low-temperature combustion requires complete description of reactions involving critical species such as hydroperoxyalkyl radicals, Q˙OOH, which undergo competing unimolecular reactions and bimolecular reactions with O2. The balance of flux across the two pathways affects rates of chain-branching and depends on temperature, pressure, and oxygen concentration. Accordingly, the influence of [O2] on product formation from alkyl + O2 reactions and the subsequent fate of Q˙OOH and related products is central to the development of an accurate chemical kinetics mechanism. However, chemical reactions consuming Q˙OOH-mediated species are often simplified to such a degree that mechanism truncation error (uncertainty derived from incomplete reaction networks) becomes significant.For the specific purpose of determining the extent to which expanded sub-mechanisms ameliorate inaccuracies in model predictions resulting from mechanism truncation error, the present work integrates speciation measurements from jet-stirred reactor experiments on cyclopentane oxidation with modeling using an expanded detailed chemical kinetics mechanism. The experiments utilize vacuum ultraviolet-absorption spectroscopy and mass spectrometry for isomer-resolved speciation of intermediates produced from cyclopentane oxidation at 835 Torr from 500 – 1000 K. [O2]-dependent experiments were also conducted from 0.12 – 3.66 · 1018 molecules cm–3 at 825 K to examine the influence on species profiles. The expanded mechanism is produced by merging detailed sub-mechanisms for Q˙OOH-mediated species (cyclopentene and 1,2-epoxycyclopentane) produced using Reaction Mechanism Generator (RMG) with an existing detailed mechanism for cyclopentane.Model predictions using the expanded mechanism yielded improvements in species mole fractions for both temperature- and O2-dependence. Ignition delay time simulations were also conducted and show appreciable sensitivity in the negative-temperature coefficient region due to incorporation of the detailed sub-mechanisms due to influence on O˙H radical populations. Sensitivity analysis revealed reactions that merit theoretical rate calculations for additional improvements and include species such as H2O2 and species related to R˙ + O2 reactions of cyclopentene and 1,2-epoxycyclopentane.

BAYESIAN CALIBRATION WITH ADAPTIVE MODEL DISCREPANCY

We investigate a computer model calibration technique inspired by the well-known Bayesian framework of Kennedy and O'Hagan (KOH). We tackle the full Bayesian formulation where model parameter and model discrepancy hyperparameters are estimated jointly and reduce the problem dimensionality by introducing a functional relationship that we call the full maximum a posteriori (FMP) method. This method also eliminates the need for a true value of model parameters that caused identifiability issues in the KOH formulation. When the joint posterior is approximated as a mixture of Gaussians, the FMP calibration is proven to avoid some pitfalls of the KOH calibration, namely missing some probability regions and underestimating the posterior variance. We then illustrate two numerical examples where both model error and measurement uncertainty are estimated together. Using the solution to the full Bayesian problem as a reference, we show that the FMP results are accurate and robust, and avoid the need for high-dimensional Markov chains for sampling.

Optimization and control strategy for wind turbine aerodynamic performance under uncertainties

Aerodynamic performance of wind turbine governs the overall energy efficiency, which has been an ever-lasting research focus in the field of wind power technology. Due to the coupling effect among the highly complex environmental and structural uncertainties, the practical aerodynamic performance may not be reliably predicted. To aggravate, this performance declines with time in service. It is of great significance to efficiently and reliably assess the impact of uncertain factors and reduce these influences on wind turbine aerodynamic performance. This paper establishes an uncertainty analysis and robustness optimization model of wind turbine aerodynamic performance considering wind speed and pitch angle error uncertainties. An approach combined the no-instrusive probabilistic collocation method is used, and the blade element momentum theory is applied to quantify influences of variable uncertainties on NREL 5 MW wind turbine aerodynamic performance. The optimization target is to reduce the sensitivity of wind turbine aerodynamic performance to uncertainties, as well as maintain capture power. The results show that the wind turbine aerodynamic and mechanical performance will be greatly affected with uncertain factors. By optimizing and adjusting wind turbine rotor speed and blade pitch angle, the wind turbine rotor power and thrust load variation can be reduced to 9.14% and 9.36%, respectively, which indeed reduces the uncertainty effects.

MODEL ERROR ESTIMATION USING PEARSON SYSTEM WITH APPLICATION TO NONLINEAR WAVES IN COMPRESSIBLE FLOWS

In data assimilation, the description of the model error uncertainty is of the utmost importance because, when incorrectly defined, it may lead to information loss about the real state of the system. In this work, we proposed a novel approach that finds the optimal distribution for describing the model error uncertainty within a particle filtering framework. The method was applied to nonlinear waves in compressible flows. We investigated the influence of observation noise statistics, resolution of the numerical model, smoothness of the solutions, and sensor location. The results showed that in almost all situations the Pearson Type I is preferred, but with different curve-shape characteristics, namely, skewed, nearly symmetric, ∩-, ∪-, and J-shaped. The distributions became, in most cases, ∪-shaped when the sensors were located near the discontinuities.