Interval Width Research Articles

Evaluating the extrapolation potential of random forest digital soil mapping

Spatial soil information is essential for informed decision-making in a wide range of fields. Digital soil mapping (DSM) using machine learning algorithms has become a popular approach for generating soil maps. DSM capitalises on the relation between environmental variables (i.e., features) and a soil property of interest. It typically needs a training dataset that covers the feature space well. Mapping in areas where there are no training data is challenging, because extrapolation in geographic space often induces extrapolation in feature space and can seriously deteriorate prediction accuracy. The objective of this study was to analyse the extrapolation effects of random forest DSM models by predicting topsoil properties (OC, clay, and pH) in four African countries using soil data from the ISRIC Africa Soil Profiles database. The study was conducted in eight experiments whereby soil data from one or three countries were used to predict in the other countries. We calculated similarities between donor and recipient areas using four measures, including soil type similarity, homosoil, dissimilarity index by area of applicability (AOA), and quantile regression forest (QRF) prediction interval width. The aim was to determine the level of agreement between these four measures and identify the method that had the strongest agreement with common validation metrics. The results indicated a positive correlation between soil type similarity, homosoil and dissimilarity index by AOA. Surprisingly, we observed a negative correlation between dissimilarity index by AOA and QRF prediction interval width. Although the cross-validation results for the trained models were acceptable, the extrapolation results were unsatisfactory, highlighting the risk of extrapolation. Using soil data from three countries instead of one increased the similarities for all measures, but it had a limited effect on improving extrapolation. Also, none of the measures had a strong correlation with the validation metrics. This was particularly disappointing for AOA and QRF, which we had expected to be strong indicators of extrapolation prediction performance. Results showed that homosoil and soil type methods had the strongest correlation with validation metrics. The results for this case study revealed limitations of using AOA and QRF as measures of extrapolation effects, highlighting the importance of not relying on these methods blindly. Further research and more case studies are needed to address the effects of extrapolation of DSM models.

An FCM based weather type classification method considering photovoltaic output and meteorological characteristics and its application in power interval forecasting

AbstractWith the increasing installation of photovoltaic (PV) systems, the impact of their randomness and volatility on power system has become a significant concern. To effectively quantify the uncertainty of PV output, it is crucial to develop reliable PV power interval forecasting technology. However, the complex relationship between PV output and meteorological factors makes it challenging for a single forecasting model to meet the forecasting demand. To solve this problem, this paper proposes a weather classification method that takes into account both PV output and meteorological characteristics. Initially, the relationship between PV output and meteorological factors is analyzed, and weather types are classified using fuzzy c‐means algorithm (FCM). Then, an extreme learning machine (ELM) is employed to establish point forecasting model. By combining kernel density estimation, a PV power generation interval forecasting model is derived. The results demonstrate that the FCM‐ELM model achieves higher forecasting accuracy and narrower interval width compared to traditional ELM models, with accuracy improvement of more than 2%. Additionally, the proposed method outperforms seasonal models with an accuracy improvement of more than 1%. The contribution of this paper includes identifying the limitations of traditional weather classification methods, proposing a novel multi‐model approach for PV interval forecasting.

Multiplicity in multiple regression: Defining the issue, evaluating solutions, and integrating perspectives.

When multiple hypothesis tests are conducted, the familywise Type I error probability correspondingly increases. Various multiple test procedures (MTPs) have been developed, which generally aim to control the familywise Type I error rate at the desired level. However, although multiplicity is frequently discussed in the ANOVA literature and MTPs are correspondingly employed, the issue has received considerably little attention in the regression literature and it is rare to see MTPs employed empirically. The present aims are three-fold. First, within the eclectic uses of multiple regression, specific situations are delineated wherein adjusting for multiplicity may be most relevant. Second, the performance of ten MTPs amenable to regression is investigated via familywise Type I error control, statistical power, and, where appropriate, false discovery rate, simultaneous confidence interval coverage and width. Although methodologists may anticipate general patterns, the focus is on the magnitude of error inflation and the size of the differences among methods under plausible scenarios. Third, perspectives from across the scientific literature are discussed, which shed light on contextual factors to consider when evaluating whether multiplicity adjustment is advantageous. Results indicated that multiple testing can be problematic, even in nonextreme situations where multiplicity consequences may not be immediately expected. Results pointed toward several effective, balanced, MTPs, particularly those that accommodate correlated parameters. Importantly, the goal is not to universally recommend MTPs for all regression models, but. rather to identify a set of circumstances wherein multiplicity is most relevant, evaluate MTPs, and integrate diverse perspectives that suggest multiplicity adjustment or alternate solutions. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Analysis of Production Risk in Intensive Chicken Farms – The Case of Kosovo

Abstract Research background The egg production sector in the poultry industry in Kosovo is well developed, with intensive chicken farms that meet 99% of domestic demand (MAFRD, 2021). However, this industry is not without risks. As such, this research investigates the production risks faced by intensive chicken farms. Purpose This study aims to investigate the severity of production risks and familiarize farmers with their levels. Research methodology Both qualitative and quantitative statistical methods were utilized in this study. Indicators are defined to measure the perception of farmers according to the Likert scale ranging from 1 (very low) to 5 (very high). A quantitative analysis is then performed to calculate the interval width, loss distribution, damage standard deviation, and coefficient of variation. Results The qualitative analysis reveals that farmers’ perceptions of production risks are not necessarily related to financial damages. The risk matrix is used to convey the levels and severity of production risks. The quantitative analysis shows that production risks have a high degree of dispersion (82.35%), a high standard deviation (€13,526), and a very high coefficient of variation (204%). Novelty This study is novel in that it is the first to explore production risks in the intensive chicken production industry in Kosovo. Previous studies have only focused on financial risk, market risk, legal risk and human resource risk.

Study on the Influence of Structure and Material Parameters on Dynamic Sealing Characteristics of Composite Seal Ring

Aiming at the failure problem of composite sealing ring, the influence of elastic modulus of carbon fiber content, groove width and number of grooves on dynamic sealing characteristics of sealing ring was analyzed by finite element method. The results showed that the maximum stress after precompression was located in the middle of the contact between slip ring and O ring, and the maximum contact pressure was located to the left of the inner circle of slip ring. After applying the fluid pressure, the maximum stress is located at the right side of the contact between the slip ring and O ring and the middle part of the contact between the slip ring and O ring, and the maximum contact pressure is located at the middle part of the inner circle of the slip ring. In dynamic sealing state, the magnitude and position of the maximum contact pressure and the maximum stress of left and right stroke are basically the same; The width and number of grooves increase the contact pressure and stress, but the width of groove interval is opposite.

A measurement of Hubble constant using cosmographic approach combining fast radio bursts and supernovae

ABSTRACT The Hubble constant (H0) is a crucial parameter in cosmology. Different cosmic observations have resulted in varying estimates results for H0, that is the well-known H0 tension. To address this issue, it is beneficial to use other data set to constrain H0. In this paper, via the cosmographic approach based on the Friedman–Lemaitre–Robertson–Walker (FLRW) metric to the dispersion measure (DM) of the intergalactic medium DMIGM(z) of fast radio bursts (FRBs), we obtain the Taylor expansion of 〈DMIGM(z)〉 in terms of redshift z. Utilizing the uncalibrated type Ia supernaove (SNe Ia) and 18 localized FRBs, we give the cosmological constraints, $H_0=65.5^{+6.4}_{-5.4}$ km s−1 Mpc−1 (68 per cent CL), cosmological deceleration parameter q0 = −0.50 ± 0.20 and the jerk parameter $j_0=-0.1^{+2.0}_{-2.5}$. To demonstrate the impact of parameter degeneracies on our analysis methods, we compare the results using three different forms of fIGM(z) and two different prior distributions of Ωb, 0. After that, we find that the uncertainty in H0 is not significantly affected by the width of the prior interval of fIGM(z) and Ωb, 0, but the mean value is influenced by the priors for fIGM(z) and Ωb, 0 due to parameter degeneracies with H0. Employing fIGM(z) that evolves with redshift, we obtain $H_0=69.0^{+6.7}_{-5.7}$ km s−1 Mpc−1. In addition, the mock analyses provide a posterior estimation of H0 with an accuracy of 4.6 per cent and higher precision for q0 and j0 in the near future.

2081. COVID-19 mRNA Vaccine Effectiveness against Hospitalizations among U.S. Children and Adolescents During Omicron Variant Predominance

Abstract Background SARS-CoV-2 Omicron variants have increased vaccine-induced immune evasion and their emergence coincided with waning COVID-19 vaccine effectiveness (VE). The duration of immunity from the primary series and VE of booster doses against pediatric COVID-19 (Omicron)-associated hospitalizations has not been characterized in the United States. Methods Using a case-control design, we examined VE against laboratory-confirmed COVID-19-associated hospitalizations, enrolling case-patients hospitalized for COVID-19 and SARS-CoV-2 test-negative controls with COVID-like illness from 31 hospitals in 23 states. VE was estimated through multivariable logistic regression by comparing the odds of antecedent primary series or booster COVID-19 mRNA vaccination within and beyond 60 days prior to hospitalization. Results were analyzed by age group (5–11 and 12–18 years) among patients admitted December 19, 2021–March 30, 2023. Results We enrolled 1,242 case-patients (972 [78%] of whom were unvaccinated, 164 [13%] of whom received life support, and 11 of whom died) and 1,309 controls. Among children aged 5-18 years, VE of 2 mRNA doses (complete primary series) against pediatric COVID-19 hospitalization was 64% (95% confidence interval [CI]: 44–76%) at 14-60 days after dose 2 (median time since dose 2, 38 days); VE waned to 38% (95% CI: 23–50%) at ≥60 days (median time since vaccination, 213 days). Within 60 days of a monovalent booster dose, VE was 52% (95% CI: -3–77%); among children aged 5-11 years, VE of a bivalent booster dose was 79% (95% CI: 25-94%). Among case-patients who required life support or died, 122/164 (74%) were unvaccinated and only 3 (2%) had received a bivalent booster dose.Figure 1.Vaccine effectiveness against COVID-19-related hospitalizations, by age group, vaccine dose, and time since last dose. VE estimates were only plotted if confidence interval width was <200. Adjusted models included ≥1 underlying medical condition (yes/no), age in years, census region, biweekly date of hospital admission, and social vulnerability index (SVI) score. Sex and race/ethnicity were tested as confounders in the full model but were dropped for absence of evidence of confounding.Table 1.Critical outcomes among enrolled COVID-19 case-patients ages 5-18 years, by vaccine dose and time since last vaccine dose. Abbreviations: ICU, intensive care unit; ECMO, extracorporeal membrane oxygenation Conclusion COVID-19 mRNA vaccination reduced the likelihood of pediatric COVID-19 hospitalization by >60%; VE waned but remained protective, even at a median of 213 days after dose 2. Although not reaching statistical significance due to limited power, monovalent boosters appeared to temporarily restore protection, and bivalent booster doses among 5–11-year-olds were nearly 80% protective against hospitalization. Nearly three quarters of children requiring life support or who died were unvaccinated. Disclosures Regina Simeone, PhD, Pfizer: Stocks/Bonds Natasha B. Halasa, MD, MPH, Merck: Grant/Research Support|Quidell: Grant/Research Support|Quidell: donation of kits|Sanofi: Grant/Research Support|Sanofi: vaccine support

Estimating extinction time using radiocarbon dates

The extinction of a species is a key demographic event often signalling major climatic, ecological and/or evolutionary shifts that can be investigated using the fossil record. In that context, radiocarbon (14C) dating has become a popular tool to time and test for scenarios of extinction that can inform on how species respond to past and future ecosystem changes. We develop CRIWM, a method for estimating extinction (and arrival) time from time series of 14C dates while accounting for probability density functions (PDF) deriving from 14C calibration. The sister method GRIWM assumes normal chronometric errors and is inappropriate for 14C chronologies as PDFs are often non-Gaussian and multi-modal. Compared to GRIWM, CRIWM reduces by 4-fold the gap between true and estimated extinction times and the width of the confidence intervals, and is less sensitive to sample size, dating errors and the temporal distribution of specimens in a species’ fossil record. We build the R package Rextinct with three user-friendly functions for computing CRIWM and GRIWM, and the PDF moments, modes and intercepts of 14C calibrations. CRIWM and GRIWM accept time series comprising only 14C dates or observations (fossils, sightings) dated by multiple chronometric methods, and calibration curves for fossils sampled in the Northern and Southern Hemispheres and marine environments. For both methods, we implement two different estimators of extinction time whether they are reliant on a p-value (original conceptualization) or not (novel). CRIWM and Rextinct are robust tools to time extinction and arrival events using 14C chronologies spanning the entire Holocene to 50,000 calendar years Before Present, and can be used to investigate demographic phenomena queried through the fossil record such as migrations, domestications and extirpations.

Abstract 13558: Age-Dependent Change of Electrocardiographic Characteristics and Cardiac Events in the Long QT Syndrome Type-3 With E1784K Mutation of SCN5A

Introduction: The pathogenic variant of E1784K in SCN5A is known as an overlapped phenotype; long-QT syndrome type-3 (LQT3) and Brugada syndrome (BrS), however it is still unclear age-dependent changes of the ECG findings and which ECG parameters are associated with cardiac events (CEs) in LQT3 patients with SCN5A -E1784K. Methods: Total 79 LQT3 patients with SCN5A -E1784K (45 probands, 42 males, age 23±17 years, QTc interval: 500±33 ms) were enrolled from the NCVC LQTS registry (2003~2023). We retrospectively evaluated age-dependent change of the clinical characteristics, ECG parameters and those associated with CEs. Results: Overall, 14 (18%) patients had CEs (10 syncope and 4 ventricular fibrillation), in which 8 occurred under 25 years old, but 6 occurred more than 35 years old. Thus, we divided all into two groups: younger (n=51) or older (n=28) than 30 years old. The younger group shows higher prevalence of male and proband compared with older group, and half of the syncope occurred during exercise. In the ECG parameters, P-R interval and QRS width were significantly longer in older group, in contrast, presence of T-wave alternance, notched, biphasic T-wave and bradycardia were more prevalent in younger group although the QTc interval and QTc peak-end were not different between the two group. Univariate analysis for CEs shows that longer QTc peak-end interval and absence of family history of LQTS are associated with CEs in younger group, whereas in older group, not only the longer QTc peak-end interval and absence of family history but also the Brugada ECG were associated with CEs. Finally, multivariate logistic regression analysis for all patients revealed that longer P-R interval, QTc peak-end and absence of family history of LQTS were independently associated with CEs (p=0.049, p<0.001, p=0.001, respectively). ROC analysis revealed that P-R interval ≥180 ms and QTc peak-end interval ≥86 ms were significantly related to previous CEs. Conclusion: In LQT3 with SCN5A -E1784K variant, ECG parameters were age-dependently altered and longer P-R and QTc peak-end interval as well as absence of family history of LQTS can be associated with CEs. Furthermore, in adult patients, more careful follow-up should be required for those with LQT3 overlapped with BrS ECG.

Abstract 16984: A Wearable Device Reproducibly Identifies Premature Ventricular Contractions and Tachyarrhythmias in Heart Failure Patients

Introduction: occurrence of ventricular arrhythmias (VAs) has a dramatic impact on prognosis of patients with heart failure (HF). Hypothesis: wearable devices, including smartwatch (SW), might provide real-time informations in patients at risk of VAs. Aims: we therefore evaluated if SW detection of QRS interval width and voltage could reproducibly detect VAs and/or premature ventricular contractions (PVCs) in HF patients. Methods: HF patients (N=100) underwent continuous electrocardiographic (ECG) recordings through telemetry and SW during hospital admission. Left ventricular ejection fraction (EF) and laboratory pro-BNP levels assessments were obtained as well. Correlation of the ECG results was addressed using Pearson’s analysis for normally distributed data and Spearman’s analysis for non-parametric data. Bland-Altman was used for reliability between telemetry and SW ECG measurements. Results: mean EF was 45.92±12.12%; mean pro-BNP was 2826±5823 pg/mL. Using the SW and the ECG data, QRS width and voltage correlation coefficients were 0.990 (p<0.001) and 0.974 (p<0.001), respectively. Concordance was found between the two assessments for QRS width (bias, 0.20; SD 3.0; lower limit, –5.68; upper limit, 6.08), and voltage (bias, -0.029; SD, 0.611; lower limit, –1.227; upper limit, 1.169). Moreover, we found correlation coefficients of 0.964 (p<0.001) and 0.970 (p<0.001), respectively, for PVCs width and PVCs voltage using SW and telemetry or other ECG report. Similarly, both techniques showed positive concordance for PVCs width (bias, 0.368; SD, 2.65; lower limit, -4.83; upper limit, 5.56) and PVCs voltage (bias, 0.54; SD, 1.25; lower limit, -1.91; and upper limit, 2.996). Remarkably, VAs identification by the SW was reproducible and standardized. Conclusions: continuous monitoring through SW might be considered in HF patients upon discharge.

Transmission accuracy–axial backlash–fatigue life-driven tolerance optimization of planetary roller screw mechanism

Abstract The planetary roller screw mechanism (PRSM) is an advanced linear transmission device. The relationship between tolerance allocation and performance risk still remains elusive, which is a challenge for its future applications. This work proposes a novel transmission accuracy–axial backlash–fatigue life-driven tolerance optimization method for the screw, roller, and nut of PRSM. A computational framework for PRSM transmission accuracy, axial backlash, and fatigue life calculation is developed to work on the parametric variation of design parameters including the eccentric, pitch, nominal diameter, and flank angle. Combinations of parametric variation are obtained by the Latin hypercube sampling-based tolerance statistical model to rapidly evaluate PRSM performance risk under operation conditions and tolerance parameters. The optimal tolerance parameters with the expanded width of tolerance interval and the minimum PRSM performance risk probability are obtained using the non-dominated sorting genetic algorithm. Results reveal that PRSM performance risk probabilities change from 89.25 to 68.72% and 58.1 to 56.86%, with 29.94 and 17.38% tolerance interval width increase under the high-precision and heavy-loading operation cases studied, respectively.

Confidence interval comparison: Precision of maximum likelihood estimates in LLOQ affected data.

When data is derived under a single or multiple lower limits of quantification (LLOQ), estimation of distribution parameters as well as precision of these estimates appear to be challenging, as the way to account for unquantifiable observations due to LLOQs needs particular attention. The aim of this investigation is to characterize the precision of censored sample maximum likelihood estimates of the mean for normal, exponential and Poisson distribution affected by one or two LLOQs using confidence intervals (CI). In a simulation study, asymptotic and bias-corrected accelerated bootstrap CIs for the location parameter mean are compared with respect to coverage proportion and interval width. To enable this examination, we derived analytical expressions of the maximum likelihood location parameter estimate for the assumption of exponentially and Poisson distributed data, where the censored sample method and simple imputation method are used to account for LLOQs. Additionally, we vary the proportion of observations below the LLOQs. When based on the censored sample estimate, the bootstrap CI led to higher coverage proportions and narrower interval width than the asymptotic CI. The results differed by underlying distribution. Under the assumption of normality, the CI's coverage proportion and width suffered most from high proportions of unquantifiable observations. For exponentially and Poisson distributed data, both CI approaches delivered similar results. To derive the CIs, the point estimates from the censored sample method are preferable, because the point estimate of the simple imputation method leads to higher bias for all investigated distributions. This biased simple imputation estimate impairs the coverage proportion of the respective CI. The bootstrap CI surpassed the asymptotic CIs with respect to coverage proportion for the investigated choice of distributional assumptions. The variety of distributions for which the methods are suitable gives the applicant a widely usable tool to handle LLOQ affected data with appropriate approaches.

Particle Swarm Training of a Neural Network for the Lower Upper Bound Estimation of the Prediction Intervals of Time Series

Many time series forecasting applications use ranges rather than point forecasts. Producing forecasts in the form of Prediction Intervals (PIs) is natural, since intervals are an important component of many mathematical models. The LUBE (Lower Upper Bound Estimation) method is aimed at finding ranges based on solving optimization problems taking into account interval width and coverage. Using the Particle Swarm Training of simple neural network, we look for a solution to the optimization problem of the Coverage Width-Based Criterion (CWC), which is the exponential convolution of conflicting criteria PICP (Prediction Interval Coverage Probability) and PINRW (Prediction Interval Normalized Root-mean-square Width). Based on the concept of the Pareto compromise, it is introduced as a Pareto front in the space of specified criteria. The Pareto compromise is constructed as a relationship between conflicting criteria based on the found solution to the optimization problem. The data under consideration are the financial time series of the MOEX closing prices. Our findings reveal that a relatively simple neural network, comprising eight neurons and their corresponding 26 parameters (weights of neuron connections and neuron signal biases), is sufficient to yield reliable PIs for the investigated financial time series. The novelty of our approach lies in the use of a simple network structure (containing fewer than 100 parameters) to construct PIs for a financial time series. Additionally, we offer an experimental construction of the Pareto frontier, formed by the PICP and PINRW criteria.

A Multi-Agent Prediction Method for Data Sampling and Transmission Reduction in Internet of Things Sensor Networks.

Sensor networks can provide valuable real-time data for various IoT applications. However, the amount of sensed and transmitted data should be kept at a low level due to the limitations imposed by network bandwidth, data storage, processing capabilities, and finite energy resources. In this paper, a new method is introduced that uses the predicted intervals of possible sensor readings to efficiently suppress unnecessary transmissions and decrease the amount of data samples collected by a sensor node. In the proposed method, the intervals of possible sensor readings are determined with a multi-agent system, where each agent independently explores a historical dataset and evaluates the similarity between past and current sensor readings to make predictions. Based on the predicted intervals, it is determined whether the real sensed data can be useful for a given IoT application and when the next data sample should be transmitted. The prediction algorithm is executed by the IoT gateway or in the cloud. The presented method is applicable to IoT sensor networks that utilize low-end devices with limited processing power, memory, and energy resources. During the experiments, the advantages of the introduced method were demonstrated by considering the criteria of prediction interval width, coverage probability, and transmission reduction. The experimental results confirm that the introduced method improves the accuracy of prediction intervals and achieves a higher rate of transmission reduction compared with state-of-the-art prediction methods.

Medium- and long-term interval optimal scheduling of cascade hydropower-photovoltaic complementary systems considering multiple uncertainties

An effective way to promote the consumption of renewable energy resource generation is by forming a joint power supply system between photovoltaic (PV) power plants widely distributed along a river basin and large-scale cascade hydropower plants, using shared transmission lines to transmit power to the power grid system. However, renewable energy resource generation has significant prediction uncertainty and can flood into the power grid, which is detrimental to its safe and stable operation. To solve the above issues, this paper proposes an improved interval optimization method to model the uncertainties of the inflow runoff of a cascade hydropower plant and the photovoltaic power output. On this basis, a medium- and long-term interval optimization scheduling model is established for a cascade hydro-PV complementary system based on extreme scenarios. This model does not need to obtain an accurate probability density distribution for renewable energy, and can realize the overall optimization of runoff-PV uncertainty scenarios within a certain range, taking into account the economy and robustness of system operation. In order to ensure the solution efficiency, the original model is transformed into a mixed integer linear programming (MILP) model by a variety of deterministic transformation methods and linearization techniques. The case study shows that: 1) Under the premise that the system power output is relatively stable, the power curtailment of the interval optimized scheduling scheme proposed in this paper is reduced by 47.5% compared with the robust scheduling scheme. The ability to cope with extreme scenarios is also significantly improved compared with the deterministic scheduling scheme. 2) There is a game relationship between the robustness and economy of the obtained scheduling scheme. When the fluctuation interval width of random variables increases from 20% to 30% and 40%, the corresponding total power generation decreases by 2.4% and 4.7%, respectively. 3) A large penalty factor is conducive to reducing the amount of abandoned hydropower, but in general, the penalty factor and ecological flow level have no obvious influence on the dispatching results, and the values of both can be reasonably determined according to the actual operation requirements.

Flexible performance constraint-based control of a quadrotor UAV-suspended payload system under input saturation

In this article, a flexible performance constraint-based tracking control scheme is proposed for the anti-swing control of a quadrotor UAV-suspended payload system under external disturbance and input saturation. By applying system coupling relations and positive system theory, a self-regulating auxiliary system is constructed to generate correction signals to achieve the ability to regulate a prescribed performance function. The disturbance interval observer is employed to estimate the unknown disturbance. The interval width of the disturbance is used to design a continuous robust signal, enhancing the system's robustness. Additionally, it is with the intention of designing the controller for the under-actuated system to achieve effective tracking that the backstepping control scheme is used. It can always track the reference signal and satisfy flexible performance constraints based on theoretical analysis. Finally, the effectiveness of the proposed scheme is verified by experimental studies.

Interval prediction of vessel trajectory based on lower and upper bound estimation and attention-modified LSTM with bayesian optimization

Uncertainty prediction of vessel trajectory is essential to enhance maritime situational awareness and traffic safety. Traditional approaches for trajectory prediction face challenges in accurately quantifying uncertainties, thereby limiting effectiveness in decision-making. To address this challenge, we propose a hybrid interval prediction frame of vessel trajectory using the lower and upper bound estimation (LUBE) and attention-modified long short-term memory (LSTM) network with bayesian optimization (BO). Firstly, trajectory data is preprocessed to solve the scale irregularity. Then, a novel trajectory interval prediction model for perceiving the prediction uncertainties is designed based on an advanced attention-modified LSTM with interval prediction capability. Meanwhile, a supervised training strategy with differentiating the interval widths of latitude and longitude is put forward to devise sample labels for training the model. Additionally, a new prediction interval-based objective function is proposed considering the target of maximizing coverage and minimizing width of trajectory interval. The BO algorithm optimizes the weighted parameters in the LSTM network by minimizing the objective function value. Finally, cases from two water areas are implemented to test and verify the proposed method. The results illustrate the superiority of the proposed approach in (1) outperforming other methods used for comparison in both coverage probability and width criteria of prediction interval. (2) quantifying the prediction uncertainty and improving the reliability of trajectory predictor. (3) performing anomaly detection tasks using visualized trajectory prediction intervals. The research can help maritime traffic participants obtain more reliable trajectory data, which supports making more reasonable traffic supervision decisions.

Measured turbulence statistics from two NASA sonic boom test campaigns

Turbulence in the lower atmosphere can have substantial effects on measured sonic boom waveforms on the ground. These turbulent distortions vary spatially and temporally and are best understood statistically. Simultaneous atmospheric and acoustic data were acquired as part of the NASA Sonic Booms in Atmospheric Turbulence (SonicBAT) and Carpet Determination in Entirety Measurements (CarpetDIEM) tests. Data from these tests, both conducted at NASA Armstrong Flight Research Center, enable a detailed study of atmospheric effects on sonic boom waveforms. The present study expands on previous work [Anderson et al., J. Acoust. Soc. Am. 152, A127 (2022)] by analyzing more booms with additional statistical analyses including metric distributions, standard deviations, and confidence intervals. Results from CarpetDIEM indicate that the confidence interval widths of mean metric levels computed across a seven-microphone array spanning 122 m vary by 1-5 dB, depending on the loudness metric. The effects of peak clipping are also considered, and an analysis is conducted to determine how to best recover from peak clipping for different metrics. [Work supported by NASA Langley Research Center through the National Institute of Aerospace and Analytical Mechanics Associates.]

Recommendations for Setting a Criterion and Assessing Commutability of Sample Materials Used in External Quality Assessment/Proficiency Testing Schemes.

It is important for external quality assessment materials (EQAMs) to be commutable with clinical samples; i.e., they should behave like clinical samples when measured using end-user clinical laboratory in vitro diagnostic medical devices (IVD-MDs). Using commutable EQAMs makes it possible to evaluate metrological traceability and/or equivalence of results between IVD-MDs. The criterion for assessing commutability of an EQAM between 2 IVD-MDs is that its result should be within the prediction interval limits based on the statistical distribution of the clinical sample results from the 2 IVD-MDs being compared. The width of the prediction interval is, among other things, dependent on the analytical performance characteristics of the IVD-MDs. A presupposition for using this criterion is that the differences in nonselectivity between the 2 IVD-MDs being compared are acceptable. An acceptable difference in nonselectivity should be small relative to the analytical performance specifications used in the external quality assessment scheme. The acceptable difference in nonselectivity is used to modify the prediction interval criterion for commutability assessment. The present report provides recommendations on how to establish a criterion for acceptable commutability for EQAMS, establish the difference in nonselectivity that can be accepted between IVD-MDs, and perform a commutability assessment. The report also contains examples for performing a commutability assessment of EQAMs.

Bayesian extreme learning machines for hydrological prediction uncertainty

In recent years, extreme learning machines (ELM) have been used to accurately predict a variety of hydrological variables (e.g., streamflow, precipitation, river water quality). Using the same model structure, ELM often obtains similar performance to multi-layer perceptron (MLP) networks without the need for an iterative learning process (backpropagation), resulting in faster training. However, despite the increasing popularity of ELM, the hydrology literature has not focused on training algorithms that can be used to generate probabilistic predictions for this method. This is an interesting research gap, as it is generally accepted that quantifying hydrological prediction uncertainty and producing probabilistic predictions (instead of point predictions or mean value predictions) is a prerequisite for reliable water resource management. Thus, for the first time, Bayesian ELM (BELM) and sparse BELM (SBELM) methods are adopted and applied for probabilistic streamflow simulation and multi-step ahead forecasting (1–3 days), using as a case study four watersheds from Mexico, Germany, Canada, and Belgium. Using deterministic and probabilistic metrics, BELM and SBELM are compared against Bayesian linear regression (BLR), MLP combined with Monte-Carlo dropout weights, and a deep learning method: long short-term memory network (LSTM) coupled with Monte-Carlo dropout weights. Adding time-lagged observations of streamflow and meteorological variables (up to 14 days), such as precipitation and potential evapotranspiration, improves the simulation accuracy up to a factor of 10. In general, both BELM and SBELM show more accurate point predictions than MLP, BLR, and LSTM. BELM and SBELM show similar performance in terms of accuracy and reliability. Although, BELM marginally outperforms SBELM by generating, on average, a narrower prediction interval width. The sparsity feature of SBELM reduces (on average) the network size by 14–83 %. BELM and SBELM produce more accurate and reliable predictions than LSTM and are up to 122 and 125 times more computationally efficient to train, respectively. The case study suggests that BELM and SBELM are promising probabilistic machine learning models for hydrological prediction that are attractive alternatives to physical (e.g., lumped conceptual) models and common deep learning models (e.g., LSTM).