Error Model Research Articles

Short-Term Photovoltaic Power Prediction Based on a Digital Twin Model

Due to the influence of meteorological conditions, shipboard photovoltaic (PV) systems have problems such as large fluctuation and inaccurate prediction of the output power. In this paper, a short-term PV power prediction method based on a novel digital twin (DT) model and BiLSTM is proposed. Firstly, a PV mechanism model and a data-driven model were established, in which the data-driven model was updated iteratively in real time using the sliding time window update method; then, these two models were converged to construct a PV DT model according to the DS evidence theory. Secondly, a BiLSTM model was built to make short-term predictions of the PV power using the augmented dataset of the DT model as an input. Finally, the method was tested and verified by experiments and further compared with main PV prediction methods. The research results indicate the following: firstly, the absolute error of the DT model was smaller than that of the mechanism model and the data-driven model, being as low as 5.62 W after the data update of the data-driven model; thus, the DT model realized data augmentation and high fidelity. Secondly, compared to several main PV prediction models, the PV DT model combined with BiLSTM had the lowest RMSE, MAE, and MAPE; the best followability; and the smallest absolute error under different weather conditions, which was especially obvious under cloudy weather conditions. In summary, the method can accurately predict the shipboard PV power, which has great theoretical significance and application value for improving the economy and reliability of solar ship operation.

Predicting Structural Temperature of Ancient City Walls: A Case Study Using Ambient Temperature and Deep Learning

ABSTRACT Accurate prediction of the structural temperature field is required for the performance assessment of engineering structures and for maintaining structural safety. However, there are complex time-lag effects between ambient temperature and structural temperature. It is difficult to accurately describe and predict the structural temperature using traditional linear and nonlinear fitting methods. Hence, this study uses the advantages of gated recurrent unit (GRU) neural networks in dealing with complex nonlinear mapping issues, further considering solar radiation and developing a structural temperature prediction framework using ambient temperature and the deep learning model. The framework is applied in structural temperature prediction for an ancient city wall in China. The solar altitude angle at the location of the city wall is calculated, and the input and output structures of the deep learning neural network model are optimized. In the long-term prediction of the structural temperature of the city wall, the root mean square error (RMSE) of the GRU model is less than 0.6°C, which solves the problem of low accuracy caused by time-lag effects.

Ensemble-based sensitivity analysis of track forecasts of typhoon In-fa (2021) without and with model errors in the ECMWF, NCEP, and CMA ensemble prediction systems

Ensemble-based sensitivity analysis (ESA) has been widely applied to identifying and investigating the sources of forecast uncertainty in tropical cyclone (TC) track. The standard ESA used in most preceding studies involves calculating the time-lagged covariance of ensemble perturbations by removing the ensemble mean. This method primarily focuses on the influence of initial errors. However, such studies ignore two critical dimensions of ESA. One is how ensemble sensitivity is influenced by the varying forecast performance across different ensemble prediction systems (EPSs). Secondly, the impact of model errors on forecast uncertainties remains unaddressed. An in-depth examination of these two aspects provides a more comprehensive understanding of the factors contributing to TC track uncertainties.This study employed the standard ESA to analyze and compare the sources of uncertainty in the track forecast of Typhoon In-fa (2021) in three representative operational EPSs. Our findings reveal that the EPS's specific performance in ensemble spread markedly influence ensemble sensitivity. We identified that variations in the shape and location of key synoptic systems, such as the western Pacific subtropical high and monsoon trough, across ensemble members were notably distinct. These variations played a significant role in shaping the uncertainty for In-fa's track forecast within each system. Furthermore, we introduced a modified ESA to better account for the influence of model errors on TC track uncertainties. The modified ESA, when applied to ensembles with substantial systematic deviations, predominantly reflects the impact of model errors on track forecast inaccuracies, offering a notably different perspective compared to the standard ESA. Significance statementEnsemble-based sensitivity analysis (ESA) serves as an effective method for identifying the origins of forecast uncertainty in tropical cyclone (TC) tracks. Most preceding studies based on the standard ESA have predominantly concentrated on examining key physical processes within individual ensemble prediction system (EPS) and the impact of initial condition uncertainties. They often overlook the dependence of the ensemble sensitivity on the varying forecast performance of EPSs, as well as the impact of model errors on sensitivity assessment. This study revealed that the ensemble forecasts across different EPSs exhibit different performance in terms of ensemble spread in the shape and location of primary weather systems. This variability significantly contributes to the uncertainty in TC track forecasts through diverse processes. Furthermore, this research introduced a modified ESA, which effectively identifies the impact of model systematic deviations on TC track forecast errors. This approach along with the standard ESA provide a more comprehensive and nuanced analysis of the forecast error sensitivity associated with both initial condition and model related uncertainties.

Dynamic performance of a polygonal thermoelectric generator using sickle-shaped fins for automotive application

About 40 % of automotive fuel energy is lost in the form of heat through exhaust gas, the utilization of thermoelectric power generation technology presents a direct conversion of thermal to electric energy. A thermoelectric generator (TEG) system was designed in this study, comprising a polygonal heat exchanger using sickle-shaped fins to enhance waste heat recycling. Additionally, a transient Computational Fluid Dynamics (CFD)-analysis model was introduced to investigate its dynamic response utilizing the instantaneous variations of exhaust gas and coolant under the China Light-Duty Vehicle Test Cycle (CLTC) as transient inputs. The TEG system exhibits an average power output of 24.48 W and a conversion efficiency of 1.12 % throughout the CLTC. After heating from ambient temperature to normal operating temperature, the power rises to 34.48 W and efficiency to 1.57 %, showing enhanced output performance. A comparison of the transient and steady-state analysis results indicates that the transient analysis reflects the hysteresis effect of heat transfer more accurately. Through experimental verification, the mean absolute percentage errors of the transient model are calculated to be 5.57 % and 8.28 % respectively, proving the credibility of the simulated results. The proposed dynamic model allows for evaluating the actual behavior of the TEG system under intricate driving cycles and offers a robust reference for TEG system optimization and its automotive applications in future.

Bifurcation detection in intravascular optical coherence tomography using vision transformer based deep learning

Objective. Bifurcation detection in intravascular optical coherence tomography (IVOCT) images plays a significant role in guiding optimal revascularization strategies for percutaneous coronary intervention (PCI). We propose a bifurcation detection method using vision transformer (ViT) based deep learning in IVOCT. Approach. Instead of relying on lumen segmentation, the proposed method identifies the bifurcation image using a ViT-based classification model and then estimate bifurcation ostium points by a ViT-based landmark detection model. Main results. By processing 8640 clinical images, the Accuracy and F1-score of bifurcation identification by the proposed ViT-based model are 2.54% and 16.08% higher than that of traditional non-deep learning methods, are similar to the best performance of convolutional neural networks (CNNs) based methods, respectively. The ostium distance error of the ViT-based model is 0.305 mm, which is reduced 68.5% compared with the traditional non-deep learning method and reduced 24.81% compared with the best performance of CNNs based methods. The results also show that the proposed ViT-based method achieves the highest success detection rate are 11.3% and 29.2% higher than the non-deep learning method, and 4.6% and 2.5% higher than the best performance of CNNs based methods when the distance section is 0.1 and 0.2 mm, respectively. Significance. The proposed ViT-based method enhances the performance of bifurcation detection of IVOCT images, which maintains a high correlation and consistency between the automatic detection results and the expert manual results. It is of great significance in guiding the selection of PCI treatment strategies.

Bayesian inversion for in-situ thermal characterisation of walls in the presence of thermal anomalies

This paper develops a Bayesian inverse modelling framework for the in-situ characterisation of walls' thermal performance in the presence of thermal anomalies subject to uncertainty. For any given wall, the proposed framework uses in-situ contact measurements of temperature and heat flux in order to approximate the (posterior) distribution of a set of parameters which are inputs of a 3D heat transfer model of the wall that accounts for the presence of unknown anomalies. The set of inputs that we infer include (i) the geometry of the thermal anomaly as well was its material properties, (ii) the as-built thermophysical properties of the clear part (i.e. without anomaly) of the wall (iii) surface resistances and (iv) modelling errors that arise from unaccounted sources of uncertainty in the boundary conditions. The geometry of the unknown thermal anomaly is parameterised with a level-set function that is inferred within the proposed Bayesian approach. In order to incorporate prior statistical information of the thermal anomaly, a Gaussian process conditioned on measurements from thermal images is employed to build a prior on the level-set function. The posterior distribution of inferred inputs is approximated via the implementation of an Ensemble Kalman Inversion algorithm that provides stable and accurate approximations of the posterior. This paper also proposes a methodology that uses the inferred 3D model to derive a thermally equivalent 1D model of the wall that is suitable for its integration with existing building energy performance software. The proposed inverse modelling technique is experimentally validated by characterising the thermal performance of an insulation panel in the presence of a thermal anomaly. Our computations show that the proposed approach produces a 3D model that accurately describes thermal performance, and a derived equivalent 1D model that effectively reproduces the dynamic thermal performance of the insulation panel in the presence of the thermal anomaly.

Using metal oxide gas sensors to estimate the emission rates and locations of methane leaks in an industrial site: assessment with controlled methane releases

Abstract. Fugitive methane (CH4) emissions occur in the whole chain of oil and gas production, including from extraction, transportation, storage, and distribution. Such emissions are usually detected and quantified by conducting surveys as close as possible to the source location. However, these surveys are labour-intensive, are costly, and fail to not provide continuous emissions monitoring. The deployment of permanent sensor networks in the vicinity of industrial CH4 emitting facilities would overcome the limitations of surveys by providing accurate emission estimates, thanks to continuous sampling of emission plumes. Yet high-precision instruments are too costly to deploy in such networks. Low-cost sensors using a metal oxide semiconductor (MOS) are presented as a cheap alternative for such deployments due to their compact dimensions and to their sensitivity to CH4. In this study, we demonstrate the ability of two types of MOS sensors (TGS 2611-C00 and TGS 2611-E00) manufactured by Figaro® to reconstruct a CH4 signal, as measured by a high-precision reference gas analyser, during a 7 d controlled release campaign conducted by TotalEnergies® in autumn 2019 near Pau, France. We propose a baseline voltage correction linked to atmospheric CH4 background variations per instrument based on an iterative comparison of neighbouring observations, i.e. data points. Two CH4 mole fraction reconstruction models were compared: multilayer perceptron (MLP) and second-degree polynomial. Emission estimates were then computed using an inversion approach based on the adjoint of a Gaussian dispersion model. Despite obtaining emission estimates comparable with those obtained using high-precision instruments (average emission rate error of 25 % and average location error of 9.5 m), the application of these emission estimates is limited to adequate environmental conditions. Emission estimates are also influenced by model errors in the inversion process.

Study on Key Properties and Model Establishment of Innovative Recycled Aggregate Pervious Concrete.

In order to meet the needs of low-impact development and sustainable development, there is an urgent desire to develop an innovative recycled aggregate pervious concrete (I-RAPC) that is of high strength and permeability. In this study, I-RAPC was prepared based on response surface methodology (RSM) using recycled aggregate, river sand, and different types of pipes as the materials, and the effects of different pipe parameters (number, diameter, material, and distribution form) on the performance of I-RAPC were investigated. In addition, the calculation model of the compressive strength and the permeability coefficient of I-RAPC were proposed. The results showed that the frontal- and lateral-compressive strengths of I-RAPC were 39.8 MPa and 42.5 MPa, respectively, when the pipe material was acrylic, the position was 1EM, and the diameter was 10 mm-at which time the permeability coefficient was 3.02 mm/s, which was the highest in this study. The maximum relative errors of the compressive strength calculation model and the permeability coefficient calculation model were only 7.52% and 4.42%, respectively, as shown by the post hoc test. Therefore, I-RAPC has the advantages of high strength and permeability and is expected to be applied in low-impact development in cities with heavy surface sediment content and rainfall.

Verification and validation of a coupled CFD–FEM approach with overset structured grid through free decay and regular wave tests

This paper presents a comprehensive verification and validation study applied to the hydrodynamic analysis of a semi-submersible Floating Offshore Wind Turbine (FOWT) platform. The numerical simulation is conducted using a coupled Computational Fluid Dynamics-Finite Element Method (CFD–FEM) approach. The CFD program incorporates an overset grid interpolation method, facilitating a more systematic grid refinement in the estimation of discretization uncertainties. A mooring analysis program, based on the FEM scheme, is integrated with the CFD program to simulate the dynamic responses of mooring system. The necessity of the nonlinear mooring model is demonstrated by comparing solutions with a comparative test using a linear mooring model. The verification study, based on a least-squares-based extrapolation method, quantifies the spatial and temporal discretization errors of the numerical model. The numerical solutions based on the baseline model are further validated against the model test measurements. A pitch free decay model test and a regular wave model test conducted within the OC5 project are utilized for the verification and validation study. Results from both tests demonstrate the reliability and accuracy of the coupled method.

Radio Astrometry at Different Frequencies

The very long baseline interferometry technique allows us to determine the positions of thousands of radio sources using the absolute astrometry approach. I have investigated the impacts of a selection of observing frequencies in a range from 2 to 43 GHz in single-band, dual-band, and quad-band observing modes on astrometric results. I processed seven data sets in a range of 72,000 to 6.9 million observations, estimated source positions, and compared them. I found that source positions derived from dual-band, quad-band, and 23.6 GHz single-band data agree at a level below 0.2 mas. Comparison of independent data sets allowed me to assess the error levels of individual catalogs: 0.05–0.07 mas per position component. Further comparison showed that individual catalogs have systematic errors at the same level. The positions from 23.6 GHz single-band data show systematic errors related to the residual ionosphere contribution. Analysis of source position differences revealed systematic errors along jet directions at a level of 0.09 mas. Network-related systematic errors affect all the data, regardless of frequency. Comparison of position estimates allowed me to derive the stochastic error model that closes the error budget. Based on the collected evidence, I have made a conclusion that the development of frequency-dependent reference frames of the entire sky is not warranted. In most cases dual-band, quad-band, and single-band data at a frequency of 22 GHz and higher can be used interchangeably, which allows us to exploit the strength of a specific frequency setup for given objects. Mixing observations at different frequencies causes errors not exceeding 0.07 mas.

Approximate analytical solution of a soil water movement equation under different ponding radii on the basis of numerical simulation

AbstractThis study addresses the problem of 2D soil water movement under ponding radii of 1, 2, and 3 cm. The soil water movement characteristics (shape parameters of the water content profile, ratio of horizontal wetting front to vertical wetting front, relationship between infiltration time and horizontal wetting front, and relationship between infiltration time and cumulative infiltration) under the above three kinds of water ponding radius were analyzed. On the basis of the assumption that the soil wetting body is a semi‐ellipse and the analytical solution of the 1D soil water movement equation at any angle, the approximate analytical solution of the 2D soil water movement equation under ponding conditions is optimized. The function relationships between infiltration time, wetting front, and cumulative infiltration are established. We applied the numerical data simulated by HYDRUS‐3D to validate the parameters in proposed analytical solutions and evaluated the relationships between the wetting front and hydraulic parameters. The results indicate that as the water ponding radius increases, the wetting body and 2D water content distribution becomes larger. When the water ponding radius was 2 cm, the numerical and analytical solution of 1D soil water distribution showed the best comparison results, and the model error was the smallest. The ratio of wetting fronts was linearly increased with the increase of air‐entry suction with R2 = 0.9969.

Semantic-based neural network repair for AI models in dynamic environments

This paper addresses a critical challenge in the field of artificial intelligence (AI) and deep learning (DL) - the debugging and repair of AI models within dynamic environments. The rapid development of frameworks like TensorFlow or PyTorch can introduce version incompatibilities, potentially breaking existing code. To tackle this issue, it can be seen that a semantic-based neural network repair method that leverages an existing semantics called ExAIS is proposed, defined in the Prolog logic programming language. This method identifies architectural and parameter errors in AI models and generates debug messages to facilitate the repair process. Key contributions of this paper include introducing a novel semantic-based AI model repair method capable of suggesting multiple changes to rectify inefficiencies in AI models. We evaluate the effectiveness of our approach in a real-world context by collecting a set of real bugs from AI developers. The background section provides an overview of AI framework testing, machine learning, neural architecture search (NAS), and deep learning. The methodology section details the repair method based on ExAIS semantics, covering error identification and repair suggestions as the two main steps. The evaluation section showcases the repair effectiveness on automatically generated and manually written models, emphasizing the applicability and efficiency of the approach.

High-performance real-time fire detection and forecasting framework for industrial cables

In industrial scenarios, cable fires have always been the most common threat, and traditional fire detection systems often rely on a large number of sensors, and the detection range is very limited, and it is impossible to effectively predict the fire situation in time. In this paper, we propose a detection and prediction scheme for industrial cable fire, which breaks the limitations of the previous research on multi-sensor signal input, and highly couples the detection and prediction modules to realize fire prediction based on video image input only. In fire detection, we design an object detection model using HSV for flame feature enhancement based on YOLOv8, and in fire prediction aspect, we use iTransformer as a time series prediction model to mine the correlation between various parameters to predict the spread of fire. In the experiments, the average absolute percentage error of the flame detection model for the detection of flame height, width and longitudinal position was 3.49%–10.64 %, 2.45%–8.89 % and 1.61%–9.31 %, respectively, and the MAPE of the time series prediction model for the above three parameters was 11.18%–15.06 % and 4.35%–8.18 %, 3.37%–6.62 %.The results of the above experiments verify that the proposed model has the ability to quantitatively analyze the fire spread trend in the actual fire and help firefighters make decisions.

Using Reinforcement Learning and Error Models for Drone Precise Landing

We propose a novel framework for achieving precision landing in drone services. The proposed framework consists of two distinct decoupled modules, each designed to address a specific aspect of landing accuracy. The first module is concerned with intrinsic errors, where new error models are introduced. This includes a spherical error model that takes into account the orientation of the drone. Additionally, we propose a live position correction algorithm that employs the error models to correct for intrinsic errors in real time. The second module focuses on external wind forces and presents an aerodynamics model with wind generation to simulate the drone’s physical environment. We utilize reinforcement learning to train the drone in simulation with the goal of landing precisely under dynamic wind conditions. Experimental results, conducted through simulations and validated in the physical world, demonstrate that our proposed framework significantly increases landing accuracy while maintaining a low onboard computational cost.

Water quality ensemble prediction model for the urban water reservoir based on the hybrid long short-term memory (LSTM) network analysis

ABSTRACT The water quality of drinking water reservoirs directly impacts the water supply safety for urban residents. This study focuses on the Da Jing Shan Reservoir, a crucial drinking water source for Zhuhai City and the Macau Special Administrative Region. The aim is to establish a prediction model for the water quality of drinking water reservoirs, which can serve as a vital reference for water plants when formulating their water supply plans. In this research, after smoothing the data using the Hodrick-Prescott filter, we utilized the long short-term memory (LSTM) network model to create a water quality prediction model for the Da Jing Shan Reservoir. Simulation calculations reveal that the model's fitting degree is consistently above 60%. Specifically, the prediction accuracy for pH, dissolved oxygen (DO), and biochemical oxygen demand (BOD) in the water quality prediction model aligns with actual results by more than 70%, effectively simulating the reservoir's water quality changes. Moreover, for parameters like pH, DO, BOD, and total phosphorus, the relative forecasting error of the LSTM model is less than 10%, confirming the model's validity. The results of this study offer an essential model reference for predicting water quality for the Da Jing Shan Reservoir.

A Method for Simulating the Positioning Errors of a Robot Gripper

The research is aimed at creating a methodology for increasing the positioning accuracy of an industrial robot and minimizing the vibration of the robot gripper by applying machine learning based on the developed mathematical model for estimating the positioning error. Two components of positioning accuracy are considered: geometric and kinematic errors and elastic static deformations. The dynamic error in the partial system of motion of the robot manipulator links is analyzed. The equation of partial motions is obtained from Lagrange’s differential equation of motion of the II kind. The system of differential equations for the positioning error was solved analytically by Euler’s method. An example of modeling the position and orientation error of the gripper due to temperature deformations of the third link for the manipulator scheme is given. An example of the modeling of static deformations and errors of the manipulator with elastic pliability of the robot links is given. An example of dynamic error modeling in a partial system of motion of the robot links is given. The proposed method of modeling robot gripper positioning errors makes it possible to increase the positioning accuracy of the industrial robot and minimize the vibration of the gripper. Having a mathematical model of positioning errors, it is possible to compensate for the positioning error by changing the speed of movement of the gripper reference point before determining the direct kinematic task.

A Design of Position Control System with Model Error Compensator

A Design of Position Control System with Model Error Compensator

Strategic Geosteering Workflow with Uncertainty Quantification and Deep Learning: Initial Test on the Goliat Field Data

Continuous integration of real-time logging-while-drilling data into a subsurface model with relevant geological uncertainties enables strategic geosteering: a field-level optimization of the well-placement strategy. Model errors arising from oversimplified conceptual geological models and imperfect simulation of measurements result in unreliable subsurface-model updates. The model errors are particularly pronounced when synthetic measurements are approximated with a fast but imperfect model, such as a deep neural network (DNN).#xD;We present a practical data-assimilation workflow consisting of offline and online phases. The offline phase involves DNN training and building an uncertain prior near-well geo-model. The online phase utilizes the flexible iterative ensemble smoother (FlexIES) to perform real-time assimilation of extra-deep electromagnetic data while accounting for the model errors in the approximate DNN model. We demonstrate the proposed workflow on historic well-log data from the Goliat Field (Barents Sea). #xD;The median of our probabilistic estimation is on par with proprietary inversion, regardless of the number of layers in the chosen prior or the approximate DNN model. By estimating model errors, FlexIES automatically quantifies the uncertainty in the boundaries and resistivities of layers, which is not standard in proprietary inversion. #xD;This capability allows us to capture uncertainties more efficiently, thus providing input for future quantitative decision support methods. We demonstrate the potential of quantitive decision support by visually estimating the ahead-of-bit risk of reservoir exit that has occurred during the considered operation.

Research and application on in-situ monitoring of strata movement in submarine cold seeps area based on three-dimensional MEMS sensor network

Accurate monitoring of strata movement is crucial for understanding the geomorphological evolution of the cold seeps. However, the complex environments and inherent challenges of deep-sea operations present significant obstacles for existing observation methods. In this paper, we present a strata movement monitoring system based on a three-dimensional micro-electromechanical system (MEMS) sensor network that includes multiple short and long sensor arrays. The movement vector is reconstructed using triaxial-acceleration and cubic Hermite interpolation. In laboratory simulations with vertical and horizontal displacement of 250 mm and 100 mm, respectively, the root mean square error of the reconstruction model is 3.74 mm. Two sensor arrays were deployed and operated for 4 days in cold seeps in South China Sea, and the observed displacement showed micro-tremors, and the displacement ranges in the x- and y- directions were (−0.4 mm, 0.18 mm) and (−0.43 mm, 0.19 mm), respectively. These results affirm the feasibility of the proposed system.

Modal spectrum analysis for an imperfect beam model with random vibration

The cross-Spectral Density (SD) is a method of analyzing the relationship between randomly-input forces and their responses in the frequency domain in spectrum analysis. It is essential for analyzing antiresonant eigenfrequencies that are difficult to understand using only direct-SD. Additionally, the mass in a beam is often analyzed in modern engineering to control the vibration characteristics of a simple beam or as a major model for error. Especially, the scalability can be expected in various fields such as nanosensors as well as macroscopic scale by studying imperfect mass. Firstly, the correlation function and SD are defined for an arbitrary vibration input in a beam. Then, the response function is established according to a harmonic input of some unit force amplitude. Moreover, the direct- and cross-SDs for the response are calculated and mathematically compared with both results. Further analysis is performed by comparing the variation of the SD with response position, the standard deviation (StDev) of the mean of maximum responses, and the bandwidth. The developed cross-SD of the imperfect beam extends the previously developed SD models of the perfect beam and can provide ideas for new types of eigenfrequency analysis, vibration isolation, etc.