Accurate Performance Estimation Research Articles

Rapid Estimation of Static Capacity Based on Machine Learning: A Time-Efficient Approach

With the global surge in electric vehicle (EV) deployment, driven by enhanced environmental regulations and efforts to reduce transportation-related greenhouse gas emissions, managing the life cycle of Li-ion batteries becomes more critical than ever. A crucial step for battery reuse or recycling is the precise estimation of static capacity at retirement. Traditional methods are time-consuming, often taking several hours. To address this issue, a machine learning-based approach is introduced to estimate the static capacity of retired batteries rapidly and accurately. Partial discharge data at a 1 C rate over durations of 6, 3, and 1 min were analyzed using a machine learning algorithm that effectively handles temporally evolving data. The estimation performance of the methodology was evaluated using the mean absolute error (MAE), mean squared error (MSE), and root mean squared error (RMSE). The results showed reliable and fairly accurate estimation performance, even with data from shorter partial discharge durations. For the one-minute discharge data, the maximum RMSE was 2.525%, the minimum was 1.239%, and the average error was 1.661%. These findings indicate the successful implementation of rapidly assessing the static capacity of EV batteries with minimal error, potentially revitalizing the retired battery recycling industry.

Across Time and Space: Senju ’s Approach for Scaling Iterative Stencil Loop Accelerators on Single and Multiple FPGAs

Stencil-based applications play an essential role in high-performance systems as they occur in numerous computational areas, such as partial differential equation solving. In this context, Iterative Stencil Loops (ISLs) represent a prominent and well-known algorithmic class within the stencil domain. Specifically, ISL-based calculations iteratively apply the same stencil to a multi-dimensional point grid multiple times or until convergence. However, due to their iterative and intensive nature, ISLs are highly performance-hungry, demanding specialized solutions. Here, Field Programmable Gate Arrays (FPGAs) represent a valid architectural choice as they enable the design of custom, parallel, and scalable ISL accelerators. Besides, the regular structure of ISLs makes them an ideal candidate for automatic optimization and generation flows. For these reasons, this article introduces Senju , an automation framework for the design of highly parallel ISL accelerators targeting single-/multi-FPGA systems. Given an input description, Senju automates the entire design process and provides accurate performance estimations. The experimental evaluation shows remarkable and scalable results, outperforming single- and multi-FPGA literature approaches under different metrics. Finally, we present a new analysis of temporal and spatial parallelism trade-offs in a real-case scenario and discuss our performance through a single- and novel specialized multi-FPGA formulation of the Roofline Model.

Event-Triggered Disturbance Estimation and Output Feedback Control Design for Inner-Formation Systems

This study investigates an event-triggered disturbance estimation approach for the inner-formation system. An extended state observer is designed using an event-based sampling scheme, which offers advantages over traditional estimation methods by reducing information transmission and unnecessary output information exchange while ensuring accurate system estimation performance. Additionally, a method for designing output-feedback control is proposed. The separation of feedback control and event-based observation in the design of output feedback allows us to apply existing optimal control algorithms to the targeted plant without compromising our established event-triggered sampling methods. A numerical simulation is presented, and we demonstrate the effectiveness of our proposed approach for the inner-formation system.

State-of-charge estimation of lithium-ion battery: Joint long short-term memory network and adaptive extended Kalman filter online estimation algorithm

With the widespread use of lithium-ion batteries, estimating the State of Charge (SOC) has been one of the most critical tasks in the battery management system (BMS). In this paper, a joint long short-term memory recurrent neural network (LSTM-RNN) and adaptive extended Kalman filter (AEKF) algorithm is proposed to achieve accurate SOC estimation. The proposed algorithm generates an initial SOC estimation value through a trained LSTM-RNN with a sliding time window signal sequence as input. The initial estimation value is then used as a feedback input to the AEKF. The results show that only a 30-s time series as the input of the LSTM-RNN can achieve satisfactory estimation results. Compared with traditional adaptive extended Kalman filter and unscented Kalman filter (UKF), the proposed algorithm has a better performance of estimation accuracy under various discharging conditions, especially when applied at low temperatures. Finally, the SOC estimation results under four operating conditions and various temperatures show that the maximum error is less than 2.5 %, in particular, the root mean square error (RMSE) and the mean absolute error (MAE) are below 1.01 % and 0.64 %. The proposed algorithm is an effective SOC estimation method with good robustness and high estimation accuracy.

Applying Self-Supervised Learning to Image Quality Assessment in Chest CT Imaging.

Many new reconstruction techniques have been deployed to allow low-dose CT examinations. Such reconstruction techniques exhibit nonlinear properties, which strengthen the need for a task-based measure of image quality. The Hotelling observer (HO) is the optimal linear observer and provides a lower bound of the Bayesian ideal observer detection performance. However, its computational complexity impedes its widespread practical usage. To address this issue, we proposed a self-supervised learning (SSL)-based model observer to provide accurate estimates of HO performance in very low-dose chest CT images. Our approach involved a two-stage model combining a convolutional denoising auto-encoder (CDAE) for feature extraction and dimensionality reduction and a support vector machine for classification. To evaluate this approach, we conducted signal detection tasks employing chest CT images with different noise structures generated by computer-based simulations. We compared this approach with two supervised learning-based methods: a single-layer neural network (SLNN) and a convolutional neural network (CNN). The results showed that the CDAE-based model was able to achieve similar detection performance to the HO. In addition, it outperformed both SLNN and CNN when a reduced number of training images was considered. The proposed approach holds promise for optimizing low-dose CT protocols across scanner platforms.

Model-based exploration of the rationality of off-label use of cetirizine in Chinese pediatric patients: a prospective cohort study.

Aims: Cetirizine is frequently administered at an increased dosage in clinical practice and recommended by several guidelines. Nonetheless, the pharmacokinetic (PK) profile and real-world safety data remain insufficient in the Chinese pediatric population. The objective of the current study is to develop a population pharmacokinetic (PPK) model for cetirizine in Chinese pediatric patients and to investigate the rationale behind its off-label usage. Methods: A prospective cohort study was conducted, enrolling children who had been diagnosed with allergic diseases and prescribed cetirizine. The outcomes were safety and pharmacokinetic (PK) parameters. Cetirizine concentrations were measured using a pre-established analytical method. Subsequently, a PK model was developed, followed by model evaluation and simulation. The developed PK model was employed to investigate the drug exposure differences across various age groups and to simulate scenarios of potential overdose. Results: Sixty-three children were enrolled, and 24 of them received a cetirizine dose exceeding the recommended dosage. A PPK model, based on published literature, served as the basis of our analysis, with adjustment made to estimate certain parameters. The final model evaluation and validation indicated accurate predictive performance and robust parameter estimation. Simulations conducted for the label-dose among age 1-12 indicated median maximum concentration at steady state (Cmax,ss) of 7year old children could be the highest. The model was also used to predict the off-label dose scenarios and overdose patient to support the clinical decision. There were no adverse drug reactions in either group. Conclusion: This study provides evidence-based and model-based exploration for optimizing cetirizine usage in Chinese pediatric patients. The cetirizine PPK model showed accurate predictive performance and could be utilized to simulate individual patient exposure in real-world clinical scenarios.

A temporal discretization and spatial integration SCR model with dual temperature-related parameters

Developing an accurate and rapid Selective Catalytic Reduction (SCR) model is crucial for advanced control strategies, which is challenging due to the complex physicochemical processes. In this paper, a novel control-oriented SCR model is proposed, which exhibits faster and more accurate estimation performance for both NOx and NH3 emissions. Unlike traditional modeling methods, this research develops a temporal discretization and spatial integration (TDSI) SCR model with dual temperature-related parameters. By introducing NH3 recycling, the spatial integration model is improved, leading to an enhanced accuracy in model estimation. Moreover, A time discretization method based on current state can effectively address the issue of sampling time sensitivity. Additionally, dual temperature-related parameters can significantly enhance model accuracy, particularly in high-temperature regions. A simplified and clear flowchart is provided for parameter identification based on a limited number of engine bench test results. In a cold-start World Harmonized Transient Cycle (WHTC), the TDSI model can achieve NOx and NH3 accuracies of 92.26% and 92.52%, respectively, with a computation time of 1.47 s. Through comprehensive comparisons, the proposed model exhibits comparable accuracy to the GT model while significantly reducing computation time by 98.39%. Additionally, compared to the Continuous Stirred-Tank Reactor (CSTR) and θT models, there is an approximate 3% improvement in NOx and NH3 accuracies. This TDSI model provides a foundation for designing urea injection control strategies and offers new perspectives for simplifying the model design.

A novel method for reference parameters identification and electrical property estimation of PV modules under varying operating conditions

The parameters identification of the diode model under reference condition are crucial and have significant impact on output characteristics estimation of photovoltaic (PV) module under varying operating conditions. The traditional methods of extracting model parameters have always focused on solving the parameter optimization problem only under some certain condition, which neglects their effect of performance estimation under other operating condition. In this paper, a novel method is proposed to identify the reference physical parameters and enhance the accuracy of electrical property estimation for PV modules under varying operating conditions. New objective function with penalty function is constructed by considering the estimated performance not only under reference condition but also under multiple real operating conditions. The maximum permissible error (MPE) is proposed to constrain the accuracy for key electrical output indicators and added in penalty function to ensure the accuracy under reference condition. Taking into account the inequality constraints, the guaranteed convergence particle swarm optimization with penalty function (P-GCPSO) is introduced for parameter identification. A set of comparisons between the measured and calculated results of PV modules indicate that the proposed method substantially more accurate than other documented methods for both single and double diode model. For the six different materials tested, the root mean square error (RMSE) was compared. The RMSE of the single-diode model is reduced by at least 57.98% compared to the compared method, with a maximum reduction of 86.57%. Similarly, the RMSE of the double -diode model was reduced by at least 47.24% with a maximum reduction of 88.71% compared to the compared methods. Additionally, the tests with six different types of PV modules at varying environmental conditions for over one year prove that the proposed method is reliable and practical for real-world applications. The relative error of power generation reached a minimum of 2.20%. This proves the reliability and practicability of the proposed method in practical applications. The proposed method is envisaged to be valuable for both offline analysis and online monitoring applications where an accurate, fast and consistent performance estimation tool is required.

Improved reference condition independent method for output performance estimation of PV modules under varying operating conditions

Traditional methods for estimating output property of the photovoltaic (PV) modules are strongly influenced by the selection of reference condition and transforming equations, which determine the calculated physical parameters under real operating conditions. The differences in the carrier transport properties of PV cells under varying operating conditions, such as the number and velocity of minority carriers at the junction edge and their recombination speed, lead to large deviations in the estimation of the output characteristics, especially under low irradiance conditions. To enhance the accuracy of performance estimation, we propose an improved method that is independent of reference condition. This method eliminates the impact of reference conditions and improves the transformation equations under all irradiance levels. Transformation equations of single diode model are established in different irradiance intervals based on the dependence of physical parameter on irradiance and temperature. Especially in the low irradiance range, all effects of irradiance and temperature are considered for each physical parameter in improved transformation equations. To optimize the unknown parameters in the transformation equations, the artificial hummingbird algorithm is used to fit experimental I–V data. The experimental results of six different types PV modules under a wide range of operating conditions are used to verify the effectiveness of the proposed method. The proposed method offers immediate benefits, including independence from reference condition and a more precise relationship between physical parameters and environmental factors in the estimation of PV output properties. Comparing the results to the traditional method by Laudani, the proposed method demonstrates superior capability in estimating I–V characteristics and accurately identifies the maximum power point under various operating conditions, which is of significant value for engineering applications.

Performance prediction and performance-based task allocation

Self-perception and perception by others can diverge. Such differences can play an important role in corporate governance settings, for example in the context of task allocation. We run two experiments to study how predictions about own performance and the performance of others develop over time and how they influence allocation in performance-based decisions. Our findings indicate that people underestimate their own performance, but are too keen to take up performance-dependent tasks. Observers form accurate performance estimates about the performer based on observation. However, when making allocation decisions on the basis of the performer's self-assessment only, they are too hesitant to trust in the performer's success.

An Improved Mathematical Model for Speed Sensorless Control of Fixed Pole Bearingless Induction Motor

In a fixed pole bearingless induction motor (FPBIM), the influence of the suspension winding is neglected in the traditional mathematical model, leading to the inaccurate speed estimation of speed sensorless control. Aiming at this problem, this paper proposes an improved mathematical model. First, considering the influence of the suspension winding, the resistance matrix and inductance matrix are extended based on the traditional natural coordinate model and the FPBIM's natural coordinate model (in the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a-b-c</i> stationary frame) is established. Secondly, through the rotation transformation by an extended transformation matrix, the FPBIM's two-phase stationary coordinate model (in the α-β stationary frame) is obtained and the flux linkage equations are extracted from it. Finally, to verify the effectiveness and feasibility of this improved mathematical model, it is applied to the speed sensorless control based on model reference adaptive system (MRAS). The simulations and experiments show that the FPBIM improved mathematical model gets more accurate speed estimation and better rotation performance under no-load, load, load disturbance and displacement disturbance, while the suspension performance is not affected.

A novel method of accurately characterizing heat pipes using thermoelectric modules

This study details an experimental method for the accurate and rapid thermal characterization of low-intermediate temperature heat pipes (<200 °C). The aim of this work was to provide a method of measuring the thermophysical characteristics of heat pipes while providing definitive uncertainty values that enable accurate performance estimations prior to implementation into electronic systems. To achieve this, the apparatus used two heat flow meters, constituting cubic copper interface blocks and thermoelectric modules that measured heat flow rates across the specimen while precise, well-calibrated (±0.01 K) thermistors measured axial temperature drops along its length. The method was first demonstrated by characterizing a cylindrical copper reference specimen of known thermal conductivity, indicating a high degree of accuracy, with measured effective thermal conductivity values within ± 2 % of the correlated value. Concerning the thermal characterization of heat pipe specimens, thermal resistances and corresponding effective thermal conductivities of 0.122–0.08 K/W and 61.8–90.8 kW/m.K were measured for heat flow rates between 5 and 30 W. Uncertainties within the derived characteristics were less than 3.2 % for low heat flow rates (5–15 W) and less than 1.3 % for higher heat flow rates (15–30 W). Moreover, a 450 % reduction in characterization time was achieved compared to methods that employ thermally massive heat flow meters. In many aspects, the presented method achieved superior performance compared to those presented within the literature, proving its effectiveness for the accurate characterization and subsequent implementation of heat pipes into electronic systems.

CASBA neurological assessment: Noninvasive EEG biomarkers classify amyloid‐PET, clinical diagnosis, cognitive dementia rating, and FCSRT free recall scores in patients with cognitive decline

AbstractBackgroundAlzheimer’s Disease (AD) is difficult to diagnose and stage, often relying on burdensome clinical tests for neural protein aggregation and cognitive performance. Delays and errors in the accurate diagnosis of AD reduce patient outcomes and prevent early therapeutic intervention. The CASBA electroencephalography (EEG) system leverages a custom sequence of dynamic visual stimuli to elicit and translate novel and noninvasive neural signals biomarkers into accurate estimates of amyloid burden, cognitive performance, and clinical diagnosis, to facilitate clinical decision‐making in patients with cognitive decline.MethodIn a board‐approved study, human patients with mild cognitive impairment (MCI; 71.2 ± 8.2 years old), subjective cognitive decline (SCD; 70.7 ± 6.3 years old), and AD diagnosis (71.2 ± 6.6 years old), alongside amyloid PET (Anti‐Serum Amyloid P Antibody), cognitive testing (i.e. FCSRT), and clinical dementia rating (CDR). Neural signals were collected on a 64‐channel equidistant EEG system (2048Hz, Ag/AgCl; ANT Neuro, Enschede)]. Automated preprocessing and machine learning classification pipelines were developed to extract EEG features and calculate estimates of physiological and cognitive AD metrics from these EEG biomarkers.ResultCASBA accurately classified normal vs. subnormal FCSRT free recall score (threshold of 18, 77.3% balanced accuracy, 84.0% specificity, 70.6% sensitivity), SCD v. MCI/AD clinical diagnosis (79.4% balanced accuracy, 77.3% specificity, 81.5% sensitivity), CDR (threshold of 1, 79.2% balanced accuracy, 76.9% specificity, 81.5% sensitivity), and amyloid‐PET (composite threshold of 1.25, 83.0% balanced accuracy, 75.0% specificity, 91.9% sensitivity).ConclusionThe CASBA neurological assessment accurately predicts important clinical data including patient diagnosis and staging, amyloid burden, and performance on key evaluations of cognition in AD. CASBA supports physician decision making and patient stratification and suggests a new method for noninvasive tracking and repeat assessment during clinical trials. As such, the CASBA neurological assessment offers a promising new tool to advance clinical care and innovation for patients with neurodegenerative disease.

Channel Estimation in MIMO OFDM Systems with Tapped Delay Line Model

The continuous increase in the user demands fornew-generation communication systems, is making the wireless channel more complex and challenging for estimation, developing a simulation model for the channel,and evaluating the performance of different MIMO systems. In this work, a simulation model for multipath fading channels in wireless communication is performed. The model includes a selection of typical Tapped-Delay-Line channel models that can be implemented to reproduce the effects of representative channel distortion and interference. Based on the simulation results, the proposed method exhibits accurate channel estimation performance for frequency-selective fading channels. The proposed work employed LS, MMSE, and ML methods for channel estimation, using 16 and 32 pilots and fixed pilot locations in each frame. Results are obtained for 4x4, 8x8, 16x16, 16x8, and 16x4 MIMO systems and tapped delay line systems.

Discordant pair analysis for sample efficient model evaluation

We present a new technique for assessing the effectiveness of a classification algorithm using discordant pair analysis. This method utilizes a known performance baseline algorithm and a large unlabeled dataset with an assumed class distribution to obtain overall performance estimates by only assessing the subset of examples that the algorithms classify discordantly. Our approach offers an efficient way to evaluate the performance of an algorithm that minimizes the human adjudications needed while also maintaining precision in the evaluation and in some cases improving the evaluation quality by reducing human adjudication errors. This approach is a computationally efficient alternative to the traditional exhaustive method of performance evaluation and has the potential to improve the accuracy of performance estimates. Simulation studies show that the discordant pair method reduces the number of adjudications by over 90%, while maintaining the same level of sensitivity and specificity.

Performance analysis and optimization of a solar assisted heat pump concept

Solar-assisted heat pumps have become a popular choice for efficient and environmentally-friendly heating and hot water production. This article introduces a unique concept of integrating a solar thermal facade into buildings to provide the cold source for a heat pump, which is a novel approach compared to other available solutions. While accurate sizing is crucial for efficient system operation, evaluating the potential benefits of better control strategies is also important because these strategies can bring significant benefits to the system. However, at present, there is a shortage of tools that can facilitate accurate sizing and effective evaluation of various control strategies for such systems. This article explores the benefits and challenges associated with this innovative technology, highlighting the need for developing models with a good representation of the system behavior and accurate performance estimation. A parametric study will be conducting on the principal components sizing for various climates in France. The aim is to gain a comprehensive understanding of the installation’s potential performance, with the optimization of sizing and control strategies to be performed at a later stage.

Machine learning approaches in microbiome research: challenges and best practices.

Microbiome data predictive analysis within a machine learning (ML) workflow presents numerous domain-specific challenges involving preprocessing, feature selection, predictive modeling, performance estimation, model interpretation, and the extraction of biological information from the results. To assist decision-making, we offer a set of recommendations on algorithm selection, pipeline creation and evaluation, stemming from the COST Action ML4Microbiome. We compared the suggested approaches on a multi-cohort shotgun metagenomics dataset of colorectal cancer patients, focusing on their performance in disease diagnosis and biomarker discovery. It is demonstrated that the use of compositional transformations and filtering methods as part of data preprocessing does not always improve the predictive performance of a model. In contrast, the multivariate feature selection, such as the Statistically Equivalent Signatures algorithm, was effective in reducing the classification error. When validated on a separate test dataset, this algorithm in combination with random forest modeling, provided the most accurate performance estimates. Lastly, we showed how linear modeling by logistic regression coupled with visualization techniques such as Individual Conditional Expectation (ICE) plots can yield interpretable results and offer biological insights. These findings are significant for clinicians and non-experts alike in translational applications.

Extreme learning machine model with honey badger algorithm based state-of-charge estimation of lithium-ion battery

Accurate state-of-charge (SOC) detection was still a challenging task to complete due to complex battery dynamics and constantly changing external conditions. The formula for SOC was difficult to determine since external parameters including voltage, current, temperature, and battery arrangement were complex. Also the methods for estimating SOC that were already in use were not always appropriate for the same car operating in various road and climatic conditions. In all situations, the conventional methodologies did not deliver an accurate estimation performance. Here, a unique optimization-based Extreme Learning Machine (ELM) was created to accurately determine a battery's SOC and enhance the operation and safety of battery systems. A lithium ion battery was first created, and data on its current, voltage, SOC, capacity, duration, and discharge rate were gathered to produce a real-time dataset at several temperatures, including 00,250 and 450. The dataset underwent additional pre-processing to standardize the values and enhance the accuracy of the data. To determine the precise state of the battery, these pre-data were loaded into the ELM model. However, the performance of ELM was significantly influenced by the length of training and the number of neurons in a hidden layer. An advanced Honey Badger Optimization Algorithm (HBA) was used to choose the appropriate hidden neurons and increase the estimation accuracy in order to overcome this problem. The proposed SOC estimation model provides 97% accuracy in the FUDS drive cycle and 99% accuracy in the US06 drive cycle. The proposed model provides a well performance for estimating SOC in lithium-ion battery at various temperature, also the proposed model was fit for real time implementation.

ASPECT-BASED SENTIMENT ANALYSIS TERHADAP ULASAN APLIKASI FLIP MENGGUNAKAN PEMBOBOTAN TERM FREQUENCY-INVERSE DOCUMENT FREQUENCY (TF-IDF) DENGAN METODE KLASIFIKASI K-NEAREST NEIGHBORS (K-NN)

The rapid growth of online transactions in Indonesia has increased the demand for efficient interbank transfer solutions. However, the costs associated with such transactions have become a significant obstacle. Flip, a company with a vision to become a global leader in customer satisfaction-driven services, offers a solution to this challenge. This study proposes an aspect-based sentiment analysis method using the K-Nearest Neighbors (K-NN) algorithm to analyze user sentiment on key aspects, namely speed, security, and the cost of using the Flip application. The results of this research provide valuable information that can be used as a basis to provide insights, suggestions and recommendations to businesses, so they can create better solutions and promote optimal user experience. The research results show that the K-NN model has the ability to predict user psychology well in all aspects, with a significant level of accuracy, specifically speed (73.04%), security (86, 05%) and costs (80.11%). In addition, this study also compares two model validation methods: simple data splitting method and K-Fold cross-validation. Although the simple data splitting method has a higher average accuracy, K-fold cross-validation is considered superior as it provides a more accurate and reliable estimate of the overall performance of the model. Sentiment analysis results show that Flip app users tend to give negative feedback on speed and security, while they give positive feedback on cost. Therefore, the main recommendation is that the company PT Fliptech Lentera Inspirasi Pertiwi improves the speed and security aspects to increase user satisfaction with the Flip application. Therefore, this customer-centric service will continue to prioritize user satisfaction as its primary goal.

A Numerical Study on the Hydrodynamic Performance of a Tanker in Bow Sea Conditions Depending on Restraint Conditions

The importance of accurate ship performance estimation is increasing for efficient ship operation. Ship performance has been evaluated through model tests in the past, but there are limitations in terms of facilities and costs. With the spread of high-performance computers, the method of evaluating the performance of a ship by numerical analysis, especially computational fluid dynamics (CFD), has become common. There have been many numerical studies on added resistance under various wave conditions for many years, showing a high reliability. Meanwhile, most of the studies were conducted under conditions where the degree of freedom (DOF) of the ship was limited due to computational complexity. In this study, we tried to compare the added resistance performance and fluid dynamics of S-VLCC with 6 DOFs in the regular wave conditions. One of the methods for utilizing the 6 DOFs is the soft-mooring system, which allows springs to be attached to the bow and stern to recover the non-restoring force of the hull. The second method considers the free-running condition. The virtual disk is used for the self-propulsion of the ship, and the rudder can be rotated to maintain its course. The propeller rotation speed and rudder angle are controlled through PID control. The bow wave (ψ = 180°) and oblique wave (ψ = 150°, 120°) conditions were considered, and various regular wave conditions from short to long wavelengths were regarded. The effects of restraint conditions on the added resistance and motion response amplitude operator (RAO), according to each wave condition, were compared. As a result, there was a difference in the roll motion for each restraint condition, and the y-direction force and yaw moment generated on the hull were compared to analyze the cause. In addition, we observed the change in flow characteristics by comparing the streamlines around the hull and the nominal wake on the propeller plane.