Machine Learning-based Method Research Articles

A machine learning-based approach for automatic motion/constraint and mobility analysis of parallel robots

Abstract Motion and constraint identification are the fundamental issue throughout the development of parallel mechanisms. Aiming at meaningful result with heuristic and visualizable process, this paper proposes a machine learning-based method for motions and constraints modeling and further develops the automatic software for mobility analysis. As a preliminary, topology of parallel mechanism is characterized by recognizable symbols and mapped to the motion of component limb through programming algorithm. A predictive model for motion and constraint with their nature meanings is constructed based on neural network. An increase in accuracy is obtained by the novel loss function, which combines the errors of network and physical equation. Based on the predictive model, an automatic framework for mobility analysis of parallel mechanisms is constructed. A software is developed with WebGL interface, providing the result of mobility analysis as well as the visualizing process particularly. Finally, five typical parallel mechanisms are taken as examples to verify the approach and its software. The method facilitates to attain motion/constraint and mobility of parallel mechanisms with both numerical and geometric features.

Determining target levels of power distribution system reliability indices using machine learning

Electrical distribution system (EDS) reliability has always been a critical subject due to the considerable share of distribution systems in power interruptions to end-users. Meeting regulatory service quality standards requires distribution companies to achieve defined reliability levels for performance improvement. Conventionally, specialists determine target reliability based on experience, experimental knowledge, and rules of thumb. However, these methods are error-prone and may result in suboptimal investments, economic inefficiency, and inadequate resource allocation in EDS operation and planning. Accordingly, this paper presents a machine learning-based method to determine the target level for EDS's reliability indices based on its historical performance. The approach involves comprehensive analysis of historical EDS reliability data, followed by a data envelopment analysis (DEA) to assess performance across subdivisions like substations or feeders, identifying the most effective units. Subsequently, machine learning models—Random Forest (RF), Support Vector Machine (SVM), and Radial Basis Function (RBF) network—are trained on optimal subdivisions from the DEA stage. These models accurately estimate target reliability for all subdivisions, contributing to the overall determination of the EDS target reliability level. A three-scenario case study validates the proposed method, suggesting its potential applicability as a valuable tool for regulatory agencies in setting EDS reliability targets. While admitting the potential challenges, notably the method's reliance on the quality and completeness of the historical data collected by distribution companies, the paper emphasizes the importance of collaborative efforts to ensure data accuracy and sufficiency. The proposed method leads to a fair mechanism to determine the target level for a real EDS's reliability indices in comparison with a business-as-usual scenario. Moreover, it provides a reliability performance ranking for the EDS's subdivisions.

Machine Learning Prediction and Evaluation for Structural Damage Comfort of Suspension Footbridge

To investigate the impact of structural damages on the comfort level of suspension footbridges under human-induced vibrations, this study addresses the limitations of traditional manual testing, which often entails significant manpower and material resources. The aim is to achieve rapid estimation and health monitoring of comfort levels during bridge operation. To accomplish this, the study combines finite-element simulation results to establish a data-driven library and introduces three distinct machine learning algorithms. Through comparative analysis, a machine learning-based method is proposed for quick evaluation of bridge comfort levels. Focusing on the Yangjiadong Suspension Bridge, the study evaluates and researches the comfort level of the structure under the influence of human-induced vibrations. The findings revealed a relatively low base frequency and high flexibility. Additionally, when considering the mass of individuals, peak acceleration decreased. The predictive performance of the Artificial Neural Network (ANN) model was found to be superior when accounting for multi-parameter damages, yielding root mean square error (RMSE), mean absolute percentage error (MAPE), and R-squared (R2) values of 0.03, 0.02, and 0.98, respectively. Moreover, the error ratio of the generalization performance analysis was below 5%. Furthermore, the study identified a damage coefficient of 0.13 for the bridge’s main cable, hanger, and steel longitudinal beam. Under a crowd density of 0.5 people per square meter, the predicted peak acceleration was 1.098 m/s2, with a model error of less than 10% compared to the observed value of 1.004 m/s2. These results underscore the model’s effectiveness in swiftly evaluating bridge comfort levels, thereby offering valuable insights for the health monitoring of bridge comfort levels.

A machine learning method for precise detection of spine bone mineral density

Osteoporosis is prevalent among the elderly and requires precise diagnostic approaches for effective treatment and prevention. This work introduces a machine learning-based method for the accurate grading of osteoporosis via the analysis of high-resolution computed tomography (HRCT) images of the spine. This approach allows for the precise extraction of the trabecular component of vertebral bodies, thereby enabling accurate measurement of bone density. Evaluated in 183 individuals, our method demonstrated enhanced stability in bone density prediction and reduced bias when juxtaposed with traditional random sampling techniques. Despite the inherent risk of overestimation, the machine learning model more accurately approximates the actual bone quality compared to conventional clinical methods, which typically involve the random extraction of 3–4 trabecular bone slices from the spine. These findings offer a novel approach to the clinical assessment of osteoporosis, underscoring the significant potential of integrating machine learning into existing medical image analysis workflows.

Harnessing artificial neural networks for inverse design and analysis of MEMS-based Fabry–Pérot filters

MEMS-based Fabry–Pérot (MEMS-FP) filters have become increasingly important in optical applications such as frequency selection spectroscopy and spectral analysis. Prior to the manufacturing of MEMS-FP devices, conducting a thorough theoretical analysis of their properties is crucial. Currently, Finite Element Analysis (FEA) is the primary method for investigating the theoretical performance of MEMS-FP during the design process. However, the complexity of modeling and prolonged simulation times associated with FEA software pose significant challenges, impeding the development timeline for MEMS-FP devices. To address these challenges, we proposed a machine learning-based method to computer-aided design of the MEMS-FP devices. Our approach involves using an Artificial Neural Network (ANN) to predict their physical properties and an inverse design method, utilizing optimization algorithms such as PSO, NSGA-II, and SA, to derive the optimal design scheme for MEMS-FP devices. Our method accelerates the prediction of physical properties of MEMS-FP devices and enables the design automation of optimal schemes in just minutes within a limited design space. This approach not only reduces the research time and complexity associated with MEMS-FP devices but also shortens their development cycle. Furthermore, one specific MEMS-FP device is fabricated and tested for proof-of-concept of our proposed method.

Optimization of Abnormal Hydraulic Fracturing Conditions of Unconventional Natural Gas Reservoirs Based on a Surrogate Model

Abnormal conditions greatly reduce the efficiency of hydraulic fracturing of unconventional gas reservoirs. Optimizing the fracturing scheme is crucial to minimize the likelihood of abnormal operational conditions, such as pressure channeling, casing deformation, and proppant plugging. This paper proposes a novel machine learning-based method for optimizing abnormal conditions during hydraulic fracturing of unconventional natural gas reservoirs. Firstly, the main controlling factors of abnormal conditions are selected through a hybrid controlling analysis, upon which a surrogate model is established for predicting the occurrence probability of abnormal conditions, rather than whether abnormal conditions happen or not. Subsequently, a machine learning-based optimization algorithm is developed to minimize the occurrence probability of abnormal conditions, acknowledging their inevitability during the fracturing process. The optimal results demonstrate the proposed method outperforms traditional methods, on average. The proposed methodology is more in line with the needs of practical operation in an environment full of uncertainty.

An individualized protein-based prognostic model to stratify pediatric patients with papillary thyroid carcinoma

Pediatric papillary thyroid carcinomas (PPTCs) exhibit high inter-tumor heterogeneity and currently lack widely adopted recurrence risk stratification criteria. Hence, we propose a machine learning-based objective method to individually predict their recurrence risk. We retrospectively collect and evaluate the clinical factors and proteomes of 83 pediatric benign (PB), 85 pediatric malignant (PM) and 66 adult malignant (AM) nodules, and quantify 10,426 proteins by mass spectrometry. We find 243 and 121 significantly dysregulated proteins from PM vs. PB and PM vs. AM, respectively. Function and pathway analyses show the enhanced activation of the inflammatory and immune system in PM patients compared with the others. Nineteen proteins are selected to predict recurrence using a machine learning model with an accuracy of 88.24%. Our study generates a protein-based personalized prognostic prediction model that can stratify PPTC patients into high- or low-recurrence risk groups, providing a reference for clinical decision-making and individualized treatment.

Development and machine learning-based calibration of low-cost multiparametric stations for the measurement of CO2 and CH4 in air

The pressing issue of atmospheric pollution has prompted the exploration of affordable methods for measuring and monitoring air contaminants as complementary techniques to standard methods, able to produce high-density data in time and space. The main challenge of this low-cost approach regards the in-field accuracy and reliability of the sensors. This study presents the development of low-cost stations for high-time resolution measurements of CO2 and CH4 concentrations calibrated via an in-field machine learning-based method. The calibration models were built based on measurements parallelly performed with the low-cost sensors and a CRDS analyzer for CO2 and CH4 as reference instrument, accounting for air temperature and relative humidity as external variables.To ensure versatility across locations, diversified datasets were collected, consisting of measurements performed in various environments and seasons. The calibration models, trained with 70 % for modeling, 15 % for validation, and 15 % for testing, demonstrated robustness with CO2 and CH4 predictions achieving R2 values from 0.8781 to 0.9827 and 0.7312 to 0.9410, and mean absolute errors ranging from 3.76 to 1.95 ppm and 0.03 to 0.01 ppm, for CO2 and CH4, respectively. These promising results pave the way for extending these stations to monitor additional air contaminants, like PM, NOx, and CO through the same calibration process, integrating them with remote data transmission modules to facilitate real-time access, control, and processing for end-users.

A more efficient, cost-effective way to harness renewable energy

Machine learning-based method determines the optimal design for solar cells.

Pipeline damage identification in nuclear industry using a particle swarm optimization-enhanced machine learning approach

The safety of nuclear industry pipelines is of paramount importance due to their vulnerability to damage from a variety of environmental factors. Ultrasonic guided wave inspection technology presents advantages such as high efficiency, low cost, and convenience, making it a widely adopted method for detecting damage in pipelines. However, the propagation of guided waves and the sensitivity of piezoelectric sensors can be significantly affected by environmental and operational conditions, leading to interference with damage identification. To address this challenge, we propose a robust machine learning-based method for identifying pipeline damage. Our approach integrates a model that comprises a particle swarm optimization-enhanced bidirectional gated recurrent unit-attention mechanism. Leveraging the attention mechanism, our method allocates more attention to significant feature dimensions, highlighting the influence of critical damage features in the identification process. Experimental tests of our model on a nuclear industry circulating water cooling pipeline demonstrate its efficacy in identifying pipeline damage under varying temperature and pressure conditions. Our proposed model outperforms other data-driven models, such as gated recurrent networks, long short-term memory, and bidirectional gated recurrent networks. Specifically, our model achieved an increase in accuracy of 2.2%, indicating the effectiveness and superiority of our proposed method.

Machine learning-based method for predicting C–V-T characteristics and electrical parameters of GaAs/AlGaAs multi-quantum wells Schottky diodes

In this work, two models of artificial neural networks are developed to predict the electrical parameters and capacitance-voltage characteristics of GaAs/AlGaAs multi-quantum wells Schottky diodes at different temperatures. Capacitance-Voltage-Temperature (C–V-T) characteristics for voltages and temperatures in the ranges (-4 V–0 V) and (20 K–400 K), respectively, were used to assess the effectiveness of the proposed approach. The first model (Model 1) is used to evaluate how well the neural network predicts the C–V-T characteristics. The second simulation, known as Model 2, was constructed to simultaneously overcome the problems of determining the electrical parameters and predicting C–V-T characteristics. Model 2 allows the calculation of the built-in voltage, effective density, and capacitance. Three-fold cross-validation and mean square error are used to assess the effectiveness of the developed models. The results clearly demonstrate the high prediction accuracy of the electrical parameters and C–V characteristics at all temperatures. After training, Model 1 Mean Square Error performance is 1.5033×10−6 at 1450 epochs, whereas Model 2 MSE is 4.9951×10−6 at 642 epochs. According to the error distribution frequency histogram, about 95 % of errors for Model 1 and Model 2 lie between [0.00535 and 0.005608] and [0.00328 and 0.00333], respectively. The R-values that correspond to the training and validation datasets for both models are close to one (0.9999). Parameters determination results have been compared against those obtained using ant lion optimizer based method. It was found that the results obtained from the neural networks models strongly agree with the experimental data.

An improved IPA approach driven by big data and its application to customer satisfaction research of energy-saving appliance

Against the backdrop of increasing global demand for reducing greenhouse gas emissions, promoting the use of energy-saving and environmentally friendly products has become a crucial aspect of low-carbon economic development. Customer satisfaction plays a vital role in the promotion of these products. To address the challenges of dealing with big data in the conventional customer satisfaction analysis tool, Importance Performance Analysis (IPA), a machine learning-based method is proposed to improve IPA. Firstly, the Latent Dirichlet Allocation (LDA) model is used to capture users’ opinions on different product topics. Then, the Support Vector Machine (SVM) and Random Forest (RF) algorithms are employed respectively to assess the satisfaction and importance of product attributes, enabling an objective measurement of customer satisfaction and adapting to the current trend of big data. The proposed method is applied to the analysis of water heater satisfaction on the JD platform, obtaining satisfaction levels for 10 topics. The research findings demonstrate that the improved IPA method based on SVM-RF effectively explores customer satisfaction and can provide some improvement strategies for platform managers and manufacturers.

Interpretable machine learning-based text classification method for construction quality defect reports

Efficient identification and remediation of construction defects are critical for ensuring the quality and success of engineering projects. However, the complexity of construction environments poses challenges to this endeavor. Current research predominantly relies on statistical and causal analyses of defect detection reports, yet these methods are time-consuming and error-prone due to the unstructured nature of such reports. To address this, machine learning techniques have been applied to classify defect texts rapidly and accurately. However, existing studies primarily focus on model performance enhancement, neglecting interpretability and the effect of imbalanced data. This study introduces RF-SMOTE, an oversampling technique based on Random Forest (RF), to address the limitations of traditional methods like SMOTE. Comparative analyses demonstrated the efficacy of RF-SMOTE in mitigating imbalanced data effects. Further, the application of SHAP-based interpretability methods in construction management decision-making was explored, filling gaps in existing research. Contributions include providing interpretable machine learning solutions, discussing the effect of imbalanced data, and proposing SHAP-based application scenarios.

Image-Based Evaluation of Cracking Degrees on Wood Fiber Bundles: A Machine Learning Approach

In this study, a machine learning-based method to assess and predict the cracking degree (CD) on wood fiber bundles (WFB) was developed, which is crucial for enhancing the quality control and refining the production process of wood scrimber (WS). By roller-cracking poplar wood one to three times, three distinct CD levels were established, and 361 WFB specimens were analyzed, using their water absorption rate (WAR) as the foundation for CD prediction. Through crack image analysis, four key quantitative parameters were identified—cracking density, coherence degree, crack count, and average width—and this study found through discriminant analysis that the discrimination accuracy on the CD levels by cracking density or coherence degree over 90%, emphasizing their significance in evaluation. Cluster analysis grouped the specimens into three clusters based on four key quantitative parameters, aligning with the CD levels. This study developed specialized prediction models for each CD level, integrating principal component analysis for dimensionality reduction with polynomial fitting, achieving mean squared error (MSE) of 0.0132, 0.0498, and 0.0204 for levels 1, 2, and 3, respectively. An integrated model, with an accuracy of 94.3% and predictions within a 20% error margin, was created, demonstrating the effectiveness of using surface crack image features to predict WAR of WFB. This research establishes a methodological framework for assessing CDs on WFB, contributing to enhancing WS product quality and helping to better understand wood cracking and water absorption mechanisms.

Construction and Research of Machine Learning-based Pedal Misoperation Recognition Method

Construction and Research of Machine Learning-based Pedal Misoperation Recognition Method

Comprehensive evaluation and performance analysis of machine learning in heart disease prediction

Heart disease is a leading cause of mortality on a global scale. Accurately predicting cardiovascular disease poses a significant challenge within clinical data analysis. The present study introduces a prediction model that utilizes various combinations of information and employs multiple established classification approaches. The proposed technique combines the genetic algorithm (GA) and the recursive feature elimination method (RFEM) to select relevant features, thus enhancing the model’s robustness. Techniques like the under sampling clustering oversampling method (USCOM) address the issue of data imbalance, thereby improving the model’s predictive capabilities. The classification challenge employs a multilayer deep convolutional neural network (MLDCNN), trained using the adaptive elephant herd optimization method (AEHOM). The proposed machine learning-based heart disease prediction method (ML-HDPM) demonstrates outstanding performance across various crucial evaluation parameters, as indicated by its comprehensive assessment. During the training process, the ML-HDPM model exhibits a high level of performance, achieving an accuracy rate of 95.5% and a precision rate of 94.8%. The system’s sensitivity (recall) performs with a high accuracy rate of 96.2%, while the F-score highlights its well-balanced performance, measuring 91.5%. It is worth noting that the specificity of ML-HDPM is recorded at a remarkable 89.7%. The findings underscore the potential of ML-HDPM to transform the prediction of heart disease and aid healthcare practitioners in providing precise diagnoses, exerting a substantial influence on patient care outcomes.

Comprehensive analysis of the interaction of antigen presentation during anti-tumour immunity and establishment of AIDPS systems in ovarian cancer.

There are hundreds of prognostic models for ovarian cancer. These genes are based on different gene classes, and there are many ways to construct the models. Therefore, this paper aims to build the most stable prognostic evaluation system known to date through 101 machine learning strategies. We combined 101 algorithm combinations with 10 machine learning algorithms to create antigen presentation-associated genetic markers (AIDPS) with outstanding precision and steady performance. The inclusive set of algorithms comprises the elastic network (Enet), Ridge, stepwise Cox, Lasso, generalized enhanced regression model (GBM), random survival forest (RSF), supervised principal component (SuperPC), Cox partial least squares regression (plsRcox), survival support vector machine (Survival-SVM). Then, in the train cohort, the prediction model was fitted using a leave-one cross-validation (LOOCV) technique, which involved 101 different possible combinations of prognostic genes. Seven validation data sets (GSE26193, GSE26712, GSE30161, GSE63885, GSE9891, GSE140082 and ICGC_OV_AU) were compared and analysed, and the C-index was calculated. Finally, we collected 32 published ovarian cancer prognostic models (including mRNA and lncRNA). All data sets and prognostic models were subjected to a univariate Cox regression analysis, and the C-index was calculated to demonstrate that the antigen presentation process should be the core criterion for evaluating ovarian cancer prognosis. In a univariate Cox regression analysis, 22 prognostic genes were identified based on the expression profiles of 283 genes involved in antigen presentation and the intersection of genes (p < 0.05). AIDPS were developed by our machine learning-based integration method, which was applied to these 22 genes. One hundred and one prediction models are fitted using the LOOCV framework, and the C-index is calculated for each model across all validation sets. Interestingly, RSF + Lasso was the best model overall since it had the greatest average C-index and the highest C-index of any combination of models tested on the validated data sets. In comparing external cohorts, we found that the C-index correlated AIDPS method using the RSF + Lasso method in 101 prediction models was in contrast to other features. Notably, AIDPS outperformed the vast majority of models across all data sets. Antigen-presenting anti-tumour immune pathways can be used as a representative gene set of ovarian cancer to track the prognosis of patients with cancer. The antigen-presenting model obtained by the RSF + Lasso method has the best C-INDEX, which plays a key role in developing antigen-presenting targeted drugs in ovarian cancer and improving the treatment outcome of patients.

Prediction and simulation of PEEP setting effects with machine learning models

ObjectiveTo establish a new machine learning-based method to adjust positive end-expiratory pressure (PEEP) using only already routinely measured data. DesignRetrospective observational study. SettingIntensive care unit (ICU). Patients or participants51811 mechanically ventilated patients in multiple ICUs in the USA (data from MIMIC-III and eICU databases). InterventionsNo interventions. Main variables of interestSuccess parameters of ventilation (arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance) ResultsThe multi-tasking neural network model performed significantly best for all target tasks in the primary test set. The model predicts arterial partial pressures of oxygen and carbon dioxide and respiratory system compliance about 45 min into the future with mean absolute percentage errors of about 21.7%, 10.0% and 15.8%, respectively. The proposed use of the model was demonstrated in case scenarios, where we simulated possible effects of PEEP adjustments for individual cases. ConclusionsOur study implies that machine learning approach to PEEP titration is a promising new method which comes with no extra cost once the infrastructure is in place. Availability of databases with most recent ICU patient data is crucial for the refinement of prediction performance.

An application of machine learning for material crack diagnosis using nonlinear ultrasonics

Crack diagnosis in non-destructive testing often requires reference data from the structure before damage or a considerable amount of response data. Also, detecting compression cracks is challenging. In this study, a machine learning-based method is proposed for diagnosing cracks in structures under compression. The method consists of convolutional neural networks (CNN) and fully connected networks (FCN). The CNN extracts features from nonlinear ultrasonic signal data, and the features determine the occurrence of fatigue cracks in a target specimen. Four types of input data are defined in accordance with the number of input frequency combinations. The performance of the proposed method is investigated using each data type to secure efficiency and accuracy in diagnosing aluminum specimens under various compression conditions. As a result, the F1 score, a measure of accuracy, of the proposed method depends on the number of input frequency combinations. The method detects high-compression cracks with high accuracy compared to the present technology specialized for compression cracks in a certain data type. A high accuracy of more than 96% is achieved with less computation time. The proposed method will provide an accurate crack diagnosis for compression cracks with reduced time and effort.

Machine Learning-Based Rapid Prediction Method of Failure Mode for Reinforced Concrete Column

Machine Learning-Based Rapid Prediction Method of Failure Mode for Reinforced Concrete Column