Significant Errors In Estimation Research Articles

Accurate and Reliable Food Nutrition Estimation Based on Uncertainty-Driven Deep Learning Model

Mobile Near-Infrared Spectroscopy (NIR) devices are increasingly being used to estimate food nutrients, offering substantial benefits to individuals with diabetes and obesity, who are particularly sensitive to food intake. However, most existing solutions prioritize accuracy, often neglecting to ensure reliability. This oversight can endanger individuals sensitive to specific foods, as it may lead to significant errors in nutrient estimation. To address these issues, we propose an accurate and reliable food nutrient prediction model. Our model introduces a loss function designed to minimize prediction errors by leveraging the relationships among food nutrients. Additionally, we developed a method that enables the model to autonomously estimate its own uncertainty based on the loss, reducing the risk to users. Comparative experiments demonstrate that our model achieves superior performance, with an R2 value of 0.98 and an RMSE of 0.40, reflecting a 5–15% improvement over other models. The autonomous result rejection mechanism showing a 40.6% improvement further enhances robustness, particularly in handling uncertain predictions. These findings highlight the potential of our approach for precise and trustworthy nutritional assessments in real-world applications.

Information Criteria for Signal Extraction Using Singular Spectrum Analysis: White and Red Noise

In singular spectrum analysis, which is applied to signal extraction, it is of critical importance to select the number of components correctly in order to accurately estimate the signal. In the case of a low-rank signal, there is a challenge in estimating the signal rank, which is equivalent to selecting the model order. Information criteria are commonly employed to address these issues. However, singular spectrum analysis is not aimed at the exact low-rank approximation of the signal. This makes it an adaptive, fast, and flexible approach. Conventional information criteria are not directly applicable in this context. The paper examines both subspace-based and information criteria, proposing modifications suited to the Hankel structure of trajectory matrices employed in singular spectrum analysis. These modifications are initially developed for white noise, and a version for red noise is also proposed. In the numerical comparisons, a number of scenarios are considered, including the case of signals that are approximated by low-rank signals. This is the most similar to the case of real-world time series. The criteria are compared with each other and with the optimal rank choice that minimizes the signal estimation error. The results of numerical experiments demonstrate that for low-rank signals and noise levels within a region of stable rank detection, the proposed modifications yield accurate estimates of the optimal rank for both white and red noise cases. The method that considers the Hankel structure of the trajectory matrices appears to be a superior approach in many instances. Reasonable model orders are obtained for real-world time series. It is recommended that a transformation be applied to stabilize the variance before estimating the rank.

Tidal heating as a discriminator for horizons in equatorial eccentric extreme mass ratio inspirals

Tidal heating in a binary black hole system is driven by the absorption of energy and angular momentum by the black hole’s horizon. Previous works have shown that this phenomenon becomes particularly significant during the late stages of an extreme mass ratio inspiral (EMRI) into a rapidly spinning massive black hole, a key focus for future low-frequency gravitational-wave observations by (for instance) the Laser Interferometer Space Antenna mission. Past analyses have largely focused on quasicircular inspiral geometry, with some of the most detailed studies looking at equatorial cases. Though useful for illustrating the physical principles, this limit is not very realistic astrophysically, since the population of EMRI events is expected to arise from compact objects scattered onto relativistic orbits in galactic centers through many-body events. In this work, we extend those results by studying the importance of tidal heating in equatorial EMRIs with generic eccentricities. Our results suggest that accurate modeling of tidal heating is crucial to prevent significant dephasing and systematic errors in EMRI parameter estimation. We examine a phenomenological model for EMRIs around exotic compact objects by parametrizing deviations from the black hole (BH) picture in terms of the fraction of radiation absorbed compared to the BH case. Based on a mismatch calculation, we find that reflectivities as small as |R|2∼O(10−5) are distinguishable from the BH case, irrespective of the value of the eccentricity. We stress, however, that this finding should be corroborated by future parameter estimation studies. Published by the American Physical Society 2024

Two-Point Localization Algorithm of a Magnetic Target Based on Tensor Geometric Invariant.

Currently, magnetic gradient tensor-based localization methods face challenges such as significant errors in geomagnetic field estimation, susceptibility to local optima in optimization algorithms, and inefficient performance. In addressing these issues, this article propose a two-point localization method under the constraint of overlaying geometric invariants. This method initially establishes the relationship between the target position and the magnetic gradient tensor by substituting an intermediate variable for the magnetic moment. Exploiting the property of the eigenvector corresponding to the minimum absolute eigenvalue being perpendicular to the target position vector, this constraint is superimposed to formulate a nonlinear system of equations of the target's position. In the process of determining the target position, the Nara method is employed for obtaining the initial values, followed by the utilization of the Levenberg-Marquardt algorithm to derive a precise solution. Experimental validation through both simulations and experiments confirms the effectiveness of the proposed method. The results demonstrate its capability to overcome the challenges faced by a single-point localization method in the presence of some errors in geomagnetic field estimation. In comparison to traditional two-point localization methods, the proposed method exhibits the highest precision. The localization outcomes under different noise conditions underscore the robust noise resistance and resilience of the proposed method.

Numerical estimation of the equivalent hydraulic conductivity for canal concrete lining with cracks

The equivalent saturated hydraulic conductivity of concrete lining (Ksce) is a key parameter to estimate seepage loss, while few studies have quantified the effect of cracks on Ksce, potentially leading to significant errors in canal seepage estimation. A three-dimensional numerical model is employed to simulate seepage loss from canals lined with concrete, and then to obtain the value of Ksce by an equivalent method. A total of 58,400 scenarios are simulated for different concrete lining and crack conditions (e.g., crack length, crack width, crack location, hydraulic conductivity of concrete (Ksc), and underlying soil (Ksoil)), and the corresponding Ksce values are estimated. Results show that the Ksce value determined by the equivalent method is reliable in calculating seepage loss from concrete lined canals. Ksce increases linearly with crack length with a slope Ck, and Ck exhibits a power function relationship with crack width, reaching a maximum value when the crack width surpasses the critical threshold, which ranges from 0.424 to 0.435 mm. When the crack width follows a log-normal distribution, the logarithmic value of Ck increases linearly with the mean value of the logarithmic crack width. The influence of crack location and Ksc on Ksce can be negligible. The maximum increment of Ksce of a crack is proportional to Ksoil. Considering the variation of Ksce with crack parameters and Ksoil, equations to determine the value of Ksce are established by adding the increment of Ksce caused by cracks to Ksc. This study establishes a quantitative relationship between Ksce and crack parameters, thereby enhancing the accuracy of seepage loss calculations in lined canal under different damage status.

Measurement of medium-voltage AC air arc temperature and particle number density based on dual-wavelength Moiré deflection technology

AC air arcs are generated in medium-voltage (MV) power systems under the effect of harsh weather conditions, equipment aging, and high penetration of distributed generation, threatening equipment and public safety. The arc current and temperature are low due to the wide application of arc suppression devices. In this scenario, the MV AC air arc does not satisfy the local thermodynamic equilibrium (LTE) condition. In addition, the repeated arcing and extinguishing processes further complicate the arc discharge mechanism, which bring challenges in the modeling and detection of MV AC air arcs. Experimental methods are a direct and efficient approach to determine the properties of arc plasmas. In this study, a dual-wavelength Moiré deflection diagnostic system was established to determine the time evolution of the particle density and radial distribution of the temperature in an MV AC air arc without relying on the LTE assumption. The electron number density and heavy particle number density change transiently during the arc discharge process and change gradient along the radial direction. The heavy particle temperature and electron temperature were then calculated based on the measured particle number density. During the arcing stage, the temperature of the electrons exceeded that of the heavy particles significantly, and the arc deviated from LTE. Finally, the limitations of the traditional single-wavelength Moiré deflection method are analyzed. The classic single-wavelength Moiré deflection method, while capable of estimating heavy particle temperature in plasma, exhibits a significant error in electron density estimation compared to the dual-wavelength Moiré deflection method.

Signal detrending in elastographic wave data with large motion artifacts through a 2D Whittaker smoother

A variety of advanced shear wave elastography techniques require the measurement of frequency-resolved phase velocity to properly characterize wave signals where factors such as tissue viscosity or guided wave effects are present. However, when applying these techniques in a dynamic environment where extrinsic motion cannot be limited, nonlinear and nonstationary trends can be expected in the motion signal used for phase velocity estimation. Because such extrinsic motion signals can contaminate a broad range of frequencies, they can produce significant errors in the phase velocity estimates. In this study, we propose a spatiotemporal detrending approach for elastography wave motion signals built from a 2D Whittaker smoother. This smoother has its origins in nonparametric regression and is well suited for fitting nonlinear curves of arbitrary shape. The resulting detrending filter was applied to Lamb wave signals from clinical ultrasound bladder vibrometry measurements. We found that the proposed detrending filter allowed for phase velocity estimates to be made in acquisitions that are otherwise be too corrupted by large motion. This works was supported by a grant from the NIH (DK131685)

Test mass charge estimation for the space inertial sensor with extended Kalman filter

Charge Management System (CMS), aimed at mitigating charge-induced noise on an isolated free-falling test mass (TM), is a crucial component of space inertial sensors in various spaceborne gravitational missions. The estimation of TM charge is one of the tasks of CMS, directly impacting CMS performance. However, current methods for TM charge estimation suffer from slow response and significant estimation errors. This paper presents a new charge estimation method that combines the force modulation principle with an extended Kalman filter (EKF). We analyzed the relationship between the TM surface potential and TM motion, then established a mathematical model. The estimation of TM charge is accomplished through the EKF algorithm. A charge estimation simulation model was developed in Simulink, and experiments were conducted to evaluate the performance of the charge estimation methods under varying charge conditions. The experimental results demonstrate that the EKF method obtains more accurate estimation results and faster convergence rates, compared with Kalman Filter (KF) and Quadrature demodulation (QD) methods. The KF and QD methods show larger errors when the estimated charge is greater than 106e order, meanwhile the QD method exhibits slower convergence rates. The presented method is expected to be beneficial for the CMS of on-orbit inertial sensors.

A Novel Transfer Learning-Based Cell SOC Online Estimation Method for a Battery Pack in Complex Application Conditions

The crucial function of a battery management system is to estimate the state of charge (SOC). However, complex application conditions like different temperatures, aging, and inconsistency between cells would cause a distribution difference between data domains and lead to a significant estimation error in the established SOC model. Transfer learning's domain adaptation offers a way to lessen the disparity in distribution across the data domains, and yet it also requires that the target space be the same across domains, which is challenging to do with the SOC online estimate. This article proposes a SOC differential processing method as well as designs a combined transfer learning method. As experiments of actual battery pack provided by China First Automobile Work-based confirmed, the proposed method can obtain precise and robust SOC estimation results with a mean absolute percentage error of less than 1.5%.

Dual Closed-Loop Position Observer With Nonlinearity-Decoupled Model for Feature- Flux-Based Sensorless SRM Drives

The feature-flux-based sensolress drives is attractive for switched reluctance motor (SRM) because its simple implementation and high reliability. However, its low model-utilization rate inevitably leads to the discrete position estimation manner. As a result, high frequency noise may cause significant estimation errors. For this problem, a position-adaptive dual closed-loop observer is proposed based on the nonlinearity decoupled model. First, a nonlinearity-decoupled flux model is studied to remodel the continuous linearized position-relevant function. Then, the position-adaptation law is designed to force the remodeled function adapt to the actual position. With the real-time regulation in the dual closed-loop structure, the continuous position estimation can be realized with high-frequency noise suppression. Finally, the studied observer is experimentally verified on an 8kW SRM platform.

Parallel Complementary Virtual Arrays Algorithm for Direction of Arrival (DOA) Estimation

This Paper discusses the challenges faced by previous method 2D Direction of Arrival (DOA) systems, such as low degrees of freedom, poor resolution, and significant estimation errors in scenarios with small snapshots. In response to these issues, the present method proposes a low-complexity 2D Direction of Arrival (DOA) estimation algorithm based on a parallel complementary virtual array.
 The algorithm utilizes two mutually parallel complementary linear arrays to generate a virtual array, addressing the limitations of traditional parallel arrays. It constructs an extended matrix with enhanced 2D angular degrees of freedom using covariance and cross-covariance matrices. The final step involves obtaining automatic matching 2D angle estimates through Singular Value Decomposition (SVD) and Estimation of Signal Parameters via Rotational Invariance Techniques (ESPRIT).
 In comparison to traditional 2D DOA estimation methods, the proposed algorithm better exploits the information from the array's received data. It can identify more incoming signals, offering high resolution without the need for 2D linear search or angle parameter matching. Importantly, it demonstrates effective estimation even in scenarios with low Signal-to-Noise Ratio (SNR) and small snapshots. Experimental simulation results validate the effectiveness and reliability of the proposed algorithm.

Dual-environment feature fusion-based method for estimating building-scale population distributions

ABSTRACTInformation on the population distribution at the building scale can help governments make supplemental decisions to address complex urban management issues. However, the discontinuity and strong spatial heterogeneity of research units at the building scale make it challenging to fuse multi-source geographic data, which causes significant errors in population estimation. To address this problem, this study proposes a method for population estimation at the building scale based on Dual-Environment Feature Fusion (DEFF). The dual environments of buildings were constructed by splitting the physical boundaries and extracting features suitable for the dual-environment scale from multi-source geographic data to describe the complex environmental features of buildings. Meanwhile, Data Quality Weighting based Technique for Order of Preference by Similarity to Ideal Solution (DQW-TOPSIS) method was proposed to assign appropriate weights to the features of the external environment for better feature fusion. Finally, a regression model was established using dual-environment features for building-scale population estimation. The experimental areas chosen for this study were Jianghan and Wuchang Districts, both located in Wuhan City, China. The estimated results of the DEFF were compared with those of the ablation experiments, as well as three publicly accessible population datasets, specifically LandScan, WorldPop, and GHS-POP, at the community scale. The evaluation results showed that DEFF had an R2 of approximately 0.8, Mean Absolute Error (MAE) of approximately 1200, Root Mean Square Error (RMSE) of approximately 1700, and both Mean Absolute Percentage Error (MAPE) and Symmetric Mean Absolute Percentage Error (SMAPE) of approximately 26%, indicating an improved performance and verifying the validity of the proposed method for fine-scale population estimation.

Experimental and numerical analysis of different vertical connections of precast shear walls with special regard towards deformability

AbstractNowadays precast concrete multi‐story dual system structures require Finite Element Method structural analysis. The stiffness of the vertical connections between wall panels influences the internal forces distribution. European standard EN 1992‐1‐1 recommends verification of strength and deformability of precast connections to be assisted by testing. Six different connection layouts were tested under pure shear with a special attention towards deformability. All connections showed close to/equivalent monolithic initial stiffness. Depending on the connection layout, the post‐cracking stiffness is reduced. The equation for shear resistance of interfaces between concrete cast at different times according to EN 1992‐1‐1 can lead to significant estimation errors due to its weak physical meaning (in disagreement with experimental observed failure mechanisms). A non‐linear finite element analysis solution strategy is validated by experimental results that shows how stochastic parameters can inflict large variation in terms of strength and stiffness.

Inferring nonlinear fractional diffusion processes from single trajectories

We present a method to infer the arbitrary space-dependent drift and diffusion of a nonlinear stochastic model driven by multiplicative fractional Gaussian noise from a single trajectory. Our method, fractional Onsager-Machlup optimisation (fOMo), introduces a maximum likelihood estimator by minimising a field-theoretic action which we construct from the observed time series. We successfully test fOMo for a wide range of Hurst exponents using artificial data with strong nonlinearities, and apply it to a data set of daily mean temperatures. We further highlight the significant systematic estimation errors when ignoring non-Markovianity, underlining the need for nonlinear fractional inference methods when studying real-world long-range (anti-)correlated systems.

Excluding quartz content from the estimation of saturated soil thermal conductivity: Combined use of differential effective medium theory and geometric mean method

Saturated soil thermal conductivity (λsat) is the maximum soil thermal conductivity value of a given soil. Although it can be determined accurately with a heat pulse sensor, there are challenges to prepare fully saturated soil samples. Numerous models have been developed to estimate λsat, and among these, the geometric mean method (GMM) generally performs well. The GMM requires soil mineral composition or quartz content information, which is unavailable for most soils. Earlier studies commonly used assumed that quartz content (fquartz) was equal to sand content (fsand) or to 0.5 × fsand, which led to significant λsat estimation errors especially on coarse-textured soils. We derived a novel method to estimate λsat from soil porosity (ϕ) based on a combination of the GMM and differential effective medium theory (DEM). The new DEM-GMM approach has a single parameter, cementation exponent (m). Using a calibration dataset of 43 soils, we determined best fit m values for soils in three groups: 1.66 for Group I (fsand < 0.4), 1.62 for Group II (0.4 <= fsand < 1) and m = -1.34ϕ+1.70 for Group III (fsand = 1). Using best fit m values for different groups, the new model can estimate λsat values from ϕ. Independent validation results on another 46 soils showed that the new model outperformed the GMM method with the assumption that fquartz = fsand or fquartz = 0.5 × fsand. The mean RMSE, Bias and R2 values of the DEM-GMM approach were 0.202 W m−1 K−1, 0.013 W m−1 K−1 and 0.89, respectively, and corresponding values of the GMM with the two assumptions were 0.295 and 0.476 W m−1 K−1, 0.056 and -0.28 W m−1 K−1, 0.80 and 0.82, respectively. The robust performance of the DEM-GMM approach suggests that it can be incorporated into thermal conductivity models to accurately estimate the thermal conductivity of unsaturated soils.

A novel method for global reliability sensitivity analysis by adaptive bivariate multiplicative dimensional reduction integral

The existing fractional moment-constrained maximum entropy method can efficiently analyse global reliability sensitivity (GRS). However, this method estimates the fractional moments by the univariate multiplicative dimensional reduction integral (UM-DR-I) that generates a significant error in GRS estimation, which may result from ignoring the interactions between input variables. Moreover, due to the direct bivariate decomposition of the fractional power performance function, the recently developed generalised bivariate additive dimensional reduction integral (GB-DR-I) may also result in higher computational cost and lower accuracy than that obtained with UM-DR-I regarding fractional moment and GRS estimations. To improve the efficiency of solving for GRS with an acceptable accuracy, a bivariate multiplicative dimensional reduction integral (BM-DR-I) is proposed to estimate the unconditional and conditional fractional moments required when estimating the GRS. The main innovation is that the proposed method analytically derives the BM-DR-I formulae for estimating the unconditional and conditional fractional moments; moreover, it discovers for the first time that the unconditional and conditional fractional moments can be estimated by sharing the same integral grid. Accordingly, the proposed BM-DR-I eliminates the dependence of the computational cost on the input dimensionality when estimating the GRS. In addition, to further improve the efficiency of estimating fractional moments and GRS sequentially with sufficient accuracy, an adaptive strategy for selecting effective bivariate terms was employed to construct an adaptive BM-DR-I. Four examples verify that adaptive BM-DR-I is superior to similar existing methods for estimating the GRS.

Careful feedback active noise and vibration control algorithm robust to large secondary path changes

Feedback active noise and vibration control (ANVC) reduces narrowband noise and vibration when there is no reference signal. However, most ANVC algorithms require a secondary path (plant) model and can become unstable when this model is inaccurate. This work proposes an algorithm less sensitive to plant modeling errors and can cope with large and fast plant changes. This is achieved by modeling the plant and the noise using an autoregressive exogenous input (ARX) model. This model allows obtaining the expected values of the future residual noise and determining the optimal control signal. Since the model coefficient uncertainty (variance) is considered when calculating the expected value, the resulting controller is a careful controller. This allows solving problems due to significant model estimation errors when information is insufficient to estimate the full ARX model.

State-of-Health Estimation and Anomaly Detection in Li-Ion Batteries Based on a Novel Architecture with Machine Learning

Variations across cells, modules, packs, and vehicles can cause significant errors in the state estimation of LIBs using machine learning algorithms, especially when trained with small datasets. Training with large datasets that account for all variations is often impractical due to resource and time constraints at initial product release. To address this issue, we proposed a novel architecture that leverages electronic control units, edge computers, and the cloud to detect unrevealed variations and abnormal degradations in LIBs. The architecture comprised a generalized deep neural network (DNN) for generalizability, a personalized DNN for accuracy within a vehicle, and a detector. We emphasized that a generalized DNN trained with small datasets must show reasonable estimation accuracy during cross validation, which is critical for real applications before online training. We demonstrated the feasibility of the architecture by conducting experiments on 65 DNN models, where we found distinct hyperparameter configurations. The results showed that the personalized DNN achieves a root mean square error (RMSE) of 0.33%, while the generalized DNN achieves an RMSE of 4.6%. Finally, the Mahalanobis distance was used to consider the SOH differences between the generalized DNN and personalized DNN to detect abnormal degradations.

Stage-by-Stage Hydraulic Fracture and Reservoir Characterization through Integration of Post-Fracture Pressure Decay Analysis and the Flowback Diagnostic Fracture Injection Test Method

Summary The post-fracture pressure decay (PFPD) technique is a low-cost method allowing for stage-by-stage hydraulic fracture characterization. The analysis of the PFPD data is complex, with data affected by both hydraulic fracture and reservoir properties. Available analysis methods in the literature are oversimplified; reservoir or fracture properties are often assumed to be constant along the horizontal well, and therefore changes in the trend of pressure decay data are attributed to hydraulic fracture or reservoir properties only. Moreover, methods analogous to those applied to the analysis of conventional diagnostic fracture injection tests (DFITs) are often used and ignore critical mechanisms involved in main-stage hydraulic fracture stimulation. A conceptual numerical simulation study was first conducted herein to understand the key mechanisms involved in main-stage hydraulic fracturing. An analytical model was then developed to account for the dynamic behavior of the hydraulic fracture, leakoff, proppant distribution, multiple fractures, and propped- and unpropped-closure events. The analytical model is cast in the form of a new straightline analysis (SLA) method that provides stage-by-stage estimates of the ratio of unpropped fracture surface area to total fracture surface area. The SLA method was validated against numerical simulation results. Moreover, to account for the variation of reservoir properties along the horizontal well, the PFPD model is integrated with DFIT-flowback (DFIT-FBA) tests, performed at some points along the lateral, to obtain a reliable stage-by-stage hydraulic fracture and reservoir characterization approach. The practical application of the proposed integrated approach was demonstrated using PFPD and DFIT-FBA data from a horizontal well completed in 22 stages in the Montney Formation. The numerical simulation study demonstrated that the use of proppant and injection into multiple clusters (creating multiple fractures) results in multiple closure events. The closure process may start early after the pump-in period at a pressure significantly higher than the minimum in-situ stress. Using DFIT-based analytical models, which ignore the presence of proppant, causes significant errors in hydraulic fracture and reservoir property estimation. The PFPD field data examined herein exhibited a similar pressure trend to the numerical simulation cases. The ratio of unpropped fracture surface area to total fracture surface area was determined stage by stage using the PFPD SLA method, constrained by DFIT-FBA data. Engineers can use this information to optimize the hydraulic fracture stimulation design in real time, optimize the well spacing, and forecast the production. The cost and time advantages of this diagnostic method make this approach very attractive.

Evaluation of GC/MS-Based 13C-Positional Approaches for TMS Derivatives of Organic and Amino Acids and Application to Plant 13C-Labeled Experiments.

Analysis of plant metabolite 13C-enrichments with gas-chromatography mass spectrometry (GC/MS) has gained interest recently. By combining multiple fragments of a trimethylsilyl (TMS) derivative, 13C-positional enrichments can be calculated. However, this new approach may suffer from analytical biases depending on the fragments selected for calculation leading to significant errors in the final results. The goal of this study was to provide a framework for the validation of 13C-positional approaches and their application to plants based on some key metabolites (glycine, serine, glutamate, proline, α-alanine and malate). For this purpose, we used tailor-made 13C-PT standards, harboring known carbon isotopologue distributions and 13C-positional enrichments, to evaluate the reliability of GC-MS measurements and positional calculations. Overall, we showed that some mass fragments of proline_2TMS, glutamate_3TMS, malate_3TMS and α-alanine_2TMS had important biases for 13C measurements resulting in significant errors in the computational estimation of 13C-positional enrichments. Nevertheless, we validated a GC/MS-based 13C-positional approach for the following atomic positions: (i) C1 and C2 of glycine_3TMS, (ii) C1, C2 and C3 of serine_3TMS, and (iii) C1 of malate_3TMS and glutamate_3TMS. We successfully applied this approach to plant 13C-labeled experiments for investigating key metabolic fluxes of plant primary metabolism (photorespiration, tricarboxylic acid cycle and phosphoenolpyruvate carboxylase activity).