Single Machine Research Articles

Predicting Response to Exclusive Combined Radio-Chemotherapy in Naso-Oropharyngeal Cancer: The Role of Texture Analysis.

The aim of this work is to identify MRI texture features able to predict the response to radio-chemotherapy (RT-CHT) in patients with naso-oropharyngeal carcinoma (NPC-OPC) before treatment in order to help clinical decision making. Textural features were derived from ADC maps and post-gadolinium T1-images on a single MRI machine for 37 patients with NPC-OPC. Patients were divided into two groups (responders/non-responders) according to results from MRI scans and 18F-FDG-PET/CT performed at follow-up 3-4 and 12 months after therapy and biopsy. Pre-RT-CHT lesions were segmented, and radiomic features were extracted. A non-parametric Mann-Whitney test was performed. A p-value < 0.05 was considered significant. Receiver operating characteristic curves and area-under-the-curve values were generated; a 95% confidence interval (CI) was reported. A radiomic model was constructed using the LASSO algorithm. After feature selection on MRI T1 post-contrast sequences, six features were statistically significant: gldm_DependenceEntropy and DependenceNonUniformity, glrlm_RunEntropy and RunLengthNonUniformity, and glszm_SizeZoneNonUniformity and ZoneEntropy, with significant cut-off values between responder and non-responder group. With the LASSO algorithm, the radiomic model showed an AUC of 0.89 and 95% CI: 0.78-0.99. In ADC, five features were selected with an AUC of 0.84 and 95% CI: 0.68-1. Texture analysis on post-gadolinium T1-images and ADC maps could potentially predict response to therapy in patients with NPC-OPC who will undergo exclusive treatment with RT-CHT, being, therefore, a useful tool in therapeutical-clinical decision making.

Investigation of the Single-Particle Scale Structure-Activity Relationship Providing New Insights for the Development of High-Performance Batteries.

As electric vehicles, portable electronic devices and tools have increasingly high requirements for battery energy density and power density, constantly improving battery performance has become a research focus. Accurate measurement of the structure-activity relationship of active materials is the key to advancing the research of high-performance batteries. However, the conventional performance tests of active materials are based on the electrochemical measurement of porous composite electrodes containing active materials, polymer binders, and conductive carbon additives, which cannot establish an accurate structure-activity relationship with the physical characterization of microregions. In this review, in order to promote the accurate measurement and understanding of the structure-activity relationship of materials, the electrochemical measurement and physical characterization of energy storage materials at single-particle scale are reviewed. The potential problems and possible improvement schemes of the single particle electrochemical measurement and physical characterization are proposed. Their potential applications in single particle electrochemical simulation and machine learning are prospected. This review aims to promote the further application of single particle electrochemical measurement and physical characterization in energy storage materials, hoping to achieve three-dimensional unified evaluation of physical characterization, electrochemical measurement, and theoretical simulation at the single particle scale to provide new inspiration for the development of high-performance batteries. This article is protected by copyright. All rights reserved.

Investigations on application of curve fitting in power system inertia estimation

ABSTRACT As the share of renewable energy sources in the power system keeps growing, a new issue with frequency stability – diminishing and changing inertia – has emerged. Estimating power system inertia has received a lot of attention lately in order to market design of synthetic inertia as an additional service and implement timely corrective steps for robust frequency control. Curve fitting technique has emerged as one of the promising techniques for estimation of power system inertia, however, many challenging aspects need to be addressed for its practical applications. This study aims to systematically assess the effectiveness of the curve fitting method under a range of operational parameters and conditions, such as the curve fitting order, the length of data to be used for curve fitting after an event begins, and the length of the moving average filter applied to the frequency data to remove oscillatory and noise from the frequency transient. The simulation studies have been performed on DIgSILENT and MATLAB platforms. The results show that very high and very low order curve fitting results are high errors and the 5th order model with mean error in estimated value of inertia less than 1% has outperformed others. The analysis on various window width of moving average filter found that increasing the window width results in more error. For IEEE-9 bus system, a filter up to 850 ms width has a mean percentage error below 1%. The analysis also revealed that the duration of the data used for curve fitting is not significant. For variation of the duration of the data used for the curve fitting from 50 ms to 500 ms, the estimated inertia values are found to be in a close range of 2.4881 s to 2.5262 s, for the single machine system with H = 2.5 s. The analysis presented in this study will be useful for effective application of curve fitting method in inertia estimation.

Development of microbial desalination cells for the treatment of reverse osmosis reject water: A new benchtop approach

To address the growing global demand for usable water, there is an immediate necessity to enhance wastewater treatment systems. A continuous mode adaption of the conventional three-chamber microbial desalination cell (MDC) configuration was used, with gravity facilitating the flow of residential reject water for desalination. Initially, operating in batch mode with a 100 mL treatment volume, the single microbial desalination machine was expanded to 300 mL in continuous mode, capable of treating 5 L of home refuse water over 36 days. The batch mode MDC had a maximum current and power density of 3.81 µA/cm2 and 0.337 µW/cm2, resulting in 76 % desalination and 83.9 % COD eradication rates. Scaling up increased the MDC's performance, reaching a maximum of 0.45 µW/cm2 and 5.31 µA/cm2, which was 1.3 times greater than batch mode operation. The current work demonstrates the feasibility of microbial desalination cells and their novel approach for treating much higher quantities of reverse osmosis (R.O.) saline water in a comparable period of roughly 36 days. It emphasizes the actual limits when dealing with real-world wastewater samples, presenting a unique path for biotechnology by simultaneously generating bio-electricity and tackling future contaminants. Furthermore, incorporating desalination chambers with microbial fuel cells increases efficiency and opens up new options for enhanced wastewater treatment, resource recovery, and bioenergy generation. This pioneering strategy uses innovative membrane technologies and microbial optimization approaches to push the limits of desalination.

Ensemble Meta-Learning-Based Robust Chipping Prediction for Wafer Dicing

Our previous study utilized importance analysis, random forest, and Barnes–Hut t-SNE dimensionality reduction to analyze critical dicing parameters and used bidirectional long short-term memory (BLSTM) to predict wafer chipping occurrence successfully in a single dicing machine. However, each dicing machine of the same type may produce unevenly distributed non-IID dicing signals, which may lead to the undesirable result that a pre-trained model trained by dicing machine #1 could not effectively predict chipping occurrence in dicing machine #2. Therefore, regarding the model robustness, this study introduces an ensemble meta-learning-based model that can evaluate many dicing machines for chipping prediction with high stability and accuracy. This approach constructs several base learners, such as the hidden Markov model (HMM), the variational autoencoder (VAE), and BLSTM, to form an ensemble learning. We use model-agnostic meta-learning (MAML) to train and test the ensemble learning model by several prediction tasks from machine #1. After MAML learning, we call the trained model a meta learner. Then, we successfully apply a retrieved data set from machine #2 to the meta learner for training and testing wafer chipping occurrence in this machine. As a result, our contribution to the robust chipping prediction on cross-machines can improve the yield of wafer dicing with a prediction accuracy of 93.21%, preserve the practical wearing of dicing kerfs, and significantly cut wafer manufacturing costs.

A deep residual neural network model for synchronous motor fault diagnostics

Synchronous motors play a significant role in a wide range of industrial applications. Their reliable operation is paramount. Any faults in synchronous motors can lead to costly downtime, decreased productivity, and potential safety hazards. By accurately diagnosing and classifying faults, we can proactively address issues before they escalate, ensuring the smooth operation of synchronous motors and minimizing the risk of equipment failure. The accurate diagnosis and fault detection in synchronous motors pose a significant challenge in their operation and maintenance. In the existing models, the feature data at various depths are not thoroughly extracted to maximize their feature extraction capability. Additionally, they employ a single support vector machine to make the final decision on the output. The single support vector machine may not consistently produce more accurate outcomes. Therefore, this paper proposes a novel fault diagnosis model based on a deep residual neural network and multiple support vector machines to diagnose mechanical and electrical faults of synchronous motors. The proposed model improves upon existing fault diagnosis models in two key aspects. Firstly, by employing a deep neural network, the model is able to effectively process and extract fault features from the motor fault dataset, capturing more nuanced information that may be missed by existing models. Secondly, the use of multiple support vector machines enhances the decision-making capability of the model, allowing for more accurate fault diagnosis. By combining these two aspects, the proposed model achieves superior diagnostic performance compared to single support vector machine-based models. Our proposed model has been rigorously evaluated using mechanical and electrical fault datasets, and the results of experimental tests clearly demonstrate its superior diagnostic performance when compared to existing fault diagnosis models. The synergy of deep neural networks and multiple support vector machines not only improves fault detection accuracy but also enhances the robustness and generalizability of the model, making it a valuable tool for real-world industrial applications.

Experimental study of the effect of dedicated mechanical subcooling on the performance of a single multiparallel evaporator

In systems with a single machine and multiple parallel evaporators, issues such as uneven liquid–vapor phase refrigerant separation and low efficiency are prevalent. This study investigates the impact of distributor inlet conditions on liquid–vapor separation uniformity and proposes improving separation performance and system efficiency by adjusting the working fluid's dryness through subcooling. It introduces the use of a rectifying nozzle-type distributor (RD) instead of a centrifugal distributor (CD) and presents experimental studies on multiple parallel evaporators. A test setup featuring multiple parallel evaporators with mechanical subcooling unit was developed to investigate the subcooling degree's impact on various aspects: superheat unevenness across branches, refrigeration capacity disparity among cold storages, cooling rates, distributor pressure drop, and overall system energy efficiency. Findings indicate significant improvements with RD; at a 16 °C subcooling degree, superheat unevenness and refrigeration capacity unevenness decreased by 30.6 % and 23.0 % respectively, while cooling rates increased by 27.3 %. At a 26 °C subcooling degree, system energy efficiency improved by 15.4 %. Compared under identical conditions, the RD system outperformed the CD system. This research contributes to a better understanding of optimizing multiple parallel evaporator performance through subcooling degree adjustments and can be applied to multi-cold room cold storage systems.

Detecting Anomalies in Blockchain Transactions using Machine Learning Classifiers and Explainability Analysis

As the use of Blockchain for digital payments continues to rise in popularity, it also becomes susceptible to various malicious attacks. Successfully detecting anomalies within Blockchain transactions is essential for bolstering trust in digital payments. However, the task of anomaly detection in Blockchain transaction data is challenging due to the infrequent occurrence of illicit transactions. Although several studies have been conducted in the field, a limitation persists: the lack of explanations for the model's predictions. This study seeks to overcome this limitation by integrating eXplainable Artificial Intelligence (XAI) techniques and anomaly rules into tree-based ensemble classifiers for detecting anomalous Bitcoin transactions. The Shapley Additive exPlanation (SHAP) method is employed to measure the contribution of each feature, and it is compatible with ensemble models. Moreover, we present rules for interpreting whether a Bitcoin transaction is anomalous or not. Additionally, we have introduced an under-sampling algorithm named XGBCLUS, designed to balance anomalous and non-anomalous transaction data. This algorithm is compared against other commonly used under-sampling and over-sampling techniques. Finally, the outcomes of various tree-based single classifiers are compared with those of stacking and voting ensemble classifiers. Our experimental results demonstrate that: (i) XGBCLUS enhances TPR and ROC-AUC scores compared to state-of-the-art under-sampling and over-sampling techniques, and (ii) our proposed ensemble classifiers outperform traditional single tree-based machine learning classifiers in terms of accuracy, TPR, and FPR scores.

Combinatorial Benders decomposition for single machine scheduling in additive manufacturing with two-dimensional packing constraints

This paper addresses a single-machine scheduling problem in additive manufacturing (AM). We focus on the direct metal laser sintering (DMLS) technology of an AM process, where parts can be produced simultaneously in a batch. In this problem, we consider simultaneously the assignment of parts to batches and the packing of parts onto two-dimensional surfaces of the build chamber to minimize the makespan objective. To solve this problem, a mixed integer linear programming model is formulated, and an approximation algorithm is proposed, followed by an exact algorithm based on combinatorial Benders decomposition method with various Benders cuts and acceleration strategies (Algorithm CBD). Finally, we conduct a comprehensive computational study to verify the efficiency of our proposed Algorithm CBD. The computational results show that our proposed Algorithm CBD can effectively solve the test instances with different sizes.

Scheduling with jobs at fixed positions

In this paper, we study classical single machine scheduling problems with the additional constraint that a set of special jobs must be scheduled at certain positions in the job sequence. In other words, a special job must start when a certain number of jobs have finished on the machine. We analyze several classical objective functions for this more general setting with the additional constraint of fixed positioned jobs. We show that for some of them they are still polynomially solvable. Then, we focus on the objective of minimizing the number of tardy jobs. Considering the case of just one special job, this allows for a polynomial-time algorithm.

Bayesian optimization-enhanced ensemble learning for the uniaxial compressive strength prediction of natural rock and its application

Engineering disasters, such as rockburst and collapse, are closely related to structural instability caused by insufficient bearing capacity of geological materials. Uniaxial compressive strength (UCS) holds considerable significance in rock engineering projects. Consequently, this study endeavors to devise efficient models for the expeditious and economical estimation of UCS. Using a dataset of 729 samples, including the Schmidt hammer rebound number, P-wave velocity, and point load index data, we evaluated six algorithms, namely Adaptive Boosting (AdaBoost), Gradient Boosting Decision Tree (GBDT), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Random Forest (RF), and Extra Trees (ET) and utilized Bayesian Optimization (BO) to optimize the aforementioned algorithms. Moreover, we applied model evaluation metrics such as Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Variance Accounted For (VAF), Nash-Sutcliffe Efficiency (NSE), Weighted Mean Absolute Percentage Error (WMAPE), Coefficient of Correlation (R), and Coefficient of Determination (R2). Among the six models, BO-ET emerged as the most optimal performer during training (RMSE = 4.5042, MAE = 3.2328, VAF = 0.9898, NSE = 0.9898, WMAPE = 0.0538, R = 0.9955, R2 = 0.9898) and testing (RMSE = 4.8234, MAE = 3.9737, VAF = 0.9881, NSE = 0.9875, WMAPE = 0.2515, R = 0.9940, R2 = 0.9875) phases. Additionally, we conducted a systematic comparison between ensemble and traditional single machine learning models such as decision tree, support vector machine, and K-Nearest Neighbors, thus highlighting the advantages of ensemble learning. Furthermore, the enhancement effect of BO on generalization performance was assessed. Finally, a BO-ET-based Graphical User Interface (GUI) system was developed and validated in a Tunnel Boring Machine-excavated tunnel.

Sentiment analysis of coronavirus data with ensemble and machine learning methods

The coronavirus pandemic has distanced people from social life and increased the use of social media. People's emotions can be determined with text data collected from social media applications. This is used in many fields, especially in commerce. This study aims to predict people's sentiments about the pandemic by applying sentiment analysis to Twitter tweets about the pandemic using single machine learning classifiers (Decision Tree-DT, K-Nearest Neighbor-KNN, Logistic Regression-LR, Naïve Bayes-NB, Random Forest-RF) and ensemble learning methods (Majority Voting (MV), Probabilistic Voting (PV), and Stacking (STCK)). After vectorizing the tweets using two predictive methods, Word2Vec (W2V) and Doc2Vec, and two traditional word representation methods, Term Frequency-Inverse Document Frequency (TF-IDF) and Bag of Words (BOW), classification models built using single machine learning classifiers were compared to models built using ensemble learning methods (MV, PV and STCK) by heterogeneously combining single machine classifier algorithms. Accuracy (ACC), F-measure (F), precision (P), and recall (R) were used as performance measures, with training/test separation rates of 70%-30% and 80%-20%, respectively. Among these models, the ACC of ensemble learning models ranged from 89% to 73%, while the ACC of single classifier models ranged from 60% to 80%. Among the ensemble learning methods, STCK with Doc2Vec text representation/embedding method gave the best ACC result of 89%. According to the experimental results, ensemble models built with heterogeneous machine learning classifier algorithms gave better results than single machine learning classifier algorithms.

Formulation, Evaluation &amp; Comparative Study of Effects of Super Disintegrants in Herbal Fast Dissolving Tablet of Nyctanthes Arbor Tristis.

The purpose of this research work was to formulate &amp; evaluate the herbal fast dissolving tablet of Nyctanthes Arbor Tristis. Taxonomically, the plant was authenticated by the Padmashri vikhe patil college of Arts, Science, &amp; Commerce in pravaranagar. Ref.No./PVPC/Bot./2023-24/235. The aimed is to treat inflammation associated with Sciatica. A two independent variables of superdisintegrants are used such as crosspovidone &amp; croscarmellose sodium. Using DOE software, 32 full factorial design was used to examine the combined impact of two formulation variables. TLC was performed to identify the beta sitosterol with RF value of 0.77. to determine the physicochemical interaction between the drug &amp; the excipients, FTIR study was done. FTIR study revealed that there is no interaction of the drug and excipient. In this investigation, the tablets were formulated and assessed via direct compression. The powder mixture were compressed into tablet using single punch tablet machine. All the formulations were evaluated for their characteristic such as weight variation, hardness, thickness, friability, disintegration time(DT), and dissolution rate. It was determined from all evaluation parameter that the F9 batch was the optimized batch. It was found that an optimized batch (F9) has good cumulative drug release in 20 minutes (88.87%), DT of 56 seconds, and hardness of 2.923kg/cm2. The comparative studies are done with the marketed formulation. With the commercial formulation, the optimized batchF9 satisfies all standard specification.

Innovative Approach to Enhance Stability: Neural Network Control and Aquila Optimization Integration in Single Machine Infinite Bus Systems

This paper highlights the need to improve the stability of single-machine infinite-bus (SMIB) systems, which is crucial for maintaining the dependability, efficiency, and safety of electrical power systems. The changing energy environment, characterized by a growing use of renewable sources and more intricate power networks, is challenging established stability measures. SMIB systems exhibit dynamic behavior, particularly during faults or unexpected load variations, requiring sophisticated real-time stabilization methods to avert power failures and provide a steady energy supply. This paper suggests a complex approach that combines power system stability analysis with a neural network controller enhanced by the Aquila optimization algorithm (AOA) to address the dynamic issues of SMIB systems. The study shows that the AOA-optimized neural network (AOA-NN) controller outperforms in avoiding disruptions and attaining speedy stabilization by exhaustively examining electrical, mechanical, and rotor dynamics. This method improves power system resilience and operational efficiency as demands and technology expand.

Elevating Innovation: Unveiling the Twin Traction Method for a 50-Ton Load Capacity Elevator in Building and Construction Applications

Most commercial elevators for buildings exceeding four stories use a cable-driven traction system. Typically, a single traction machine operates by hoisting the main cable on a traction sheave, thus vertically transporting the elevator car through rotational motion of the sheave. This research introduces a groundbreaking advancement aimed at elevating loading capacity to an unprecedented 50 tons—the highest known in the world. The innovation involves the development of a twin traction system, wherein two traction machines collaborate to lift the elevator. This novel elevator system has demonstrated remarkable capabilities, showcasing the ability to transport up to 300 passengers in a single trip. The installation of this high-capacity elevator system has yielded substantial improvements in construction work efficiency and safety protocols, particularly in scenarios where cranes are traditionally used. The newly developed elevator could lift 50 tons of equipment 60 times a day, whereas the crane was limited to 8 times. The positive impact on labor is also noteworthy, with increased safety and health considerations, especially in adverse weather conditions. By eliminating the need for manual stair climbing, the well-being of the workforce is prioritized. Furthermore, the heightened productivity resulting from a significant reduction in wait times for conventional elevators is a key outcome of this transformative technology. This research not only unveils a groundbreaking twin traction system but also highlights its multifaceted features in enhancing efficiency, safety, and overall productivity in various industries.

Industrial Big Data‐Driven Modeling and Prediction for Hot‐Rolled Strip Crown with Multigrade and Multispecification Data

In the field of hot rolling big data, the presence of different steel types, specifications, and data heterogeneity poses significant challenges to the accuracy and stability of using single machine learning regression technology for prediction. Therefore, this study proposes a hot‐rolled strip crown prediction method that combines data clustering and fusion modeling. First, this article introduces a relevant mechanism for designing cluster strategies. The optimal clustering strategy is determined through comparative experiments using rolling process parameters, strip size, and main material components as the clustering features. Subsequently, the K‐Means++ algorithm is used to effectively cluster the training and testing datasets based on this strategy, generating multiple clusters for both datasets. Finally, this study establishes seven different training models to match the most suitable regression prediction model for each cluster, and matching between each cluster and the model is determined through rigorous testing. The evaluation of the fusion model shows an R2 value of 0.829 and a root mean square error value of 3.974. The experimental results show that the proposed method outperforms traditional methods in solving the challenges of multiclass classification and data heterogeneity, providing strong data support for the intelligent control of the hot‐rolled strip crown in the future.

Single machine scheduling problems with truncated learning effects and exponential past-sequence-dependent delivery times

Single machine scheduling problems with truncated learning effects and exponential past-sequence-dependent delivery times

Minimizing of Forecasting Error in Fuzzy Time Series Model Using Graph-Based Clustering Method

In recent years, numerous fuzzy time series (FTS) forecasting models have been developed to address complex and incomplete problems. However, the accuracy of these models is specific to the problem at hand and varies across datasets. Despite claims of superiority over traditional statistical and single machine learning-based models, achieving improved forecasting accuracy remains a formidable challenge. In FTS models, the lengths of intervals and fuzzy relationship groups are considered crucial factors influencing forecasting accuracy. Hence, this study introduces an FTS forecasting model based on the graph-based clustering technique. The clustering algorithm, utilized during the fuzzification stage, enables the derivation of unequal interval lengths. The proposed model is applied to forecast two numerical datasets: enrollment data from the University of Alabama and the datasets of Gas prices RON95 in Vietnam. Comparisons of forecasting results between the proposed model and others are conducted for enrollment forecasts at the University of Alabama. The findings reveal that the proposed model achieves higher forecasting accuracy across all orders of fuzzy relationships when compared to its counterparts

CDFRS: A scalable sampling approach for efficient big data analysis

The sampling-based approximation method has demonstrated its potential in various domains such as machine learning, query processing, and data analysis. Most preceding sampling algorithms generate samples at the record level, making it impractical to apply them to very large datasets using a single machine. Even distributed solutions encounter efficiency issues when dealing with terabyte-scale datasets. In this paper, we introduce a scalable sampling approach named CDFRS, which can generate samples with a distribution-preserving guarantee from extensive datasets. CDFRS exhibits significantly improved speed compared to existing sampling algorithms when dealing with terabyte-scale datasets. We provide theoretical guarantees and empirical justifications, demonstrating that samples generated by the CDFRS approach maintain the distribution characteristics of the original dataset. Additionally, we propose a sample size determination algorithm, denoted as A2. Experiment results indicate that the running time of CDFRS shows at least an order of magnitude improvement over other distributed sampling methods. Notably, sampling a 10TB dataset using CDFRS only takes hundreds of seconds, while the compared method requires more than ten thousand seconds. In the context of big data analysis, including tasks such as classification and clustering, models trained with samples generated by CDFRS closely match those trained with the entire training set. Furthermore, the proposed A2 algorithm efficiently determines an appropriate sample size compared with traditional methods.

Big Data Scheduling with Order Acceptance Consideration in Supply Chain

The efficient job scheduling schemes can make full use of resources, and then achieve different goals, such as maximizing efficiency, minimizing cost and saving energy. In supply chain, there are a lot of members and enormous data. Therefore, a suitable scheduling scheme has become the most common and effective method to optimize the execution of big data in supply chain. A scheduling problem with production and delivery consideration, which is usually in supply chain has been considered. For the reason of highly time emergency and random coming orders in quick production and delivery system, the effective algorithms for order acceptance scheduling problem are required. This paper addressed the production and delivery problem which has one manufacturer and multiple customers. There is single machine for order production at the manufacturer. For the consideration of order acceptance or not, the manufacturer need to choose order set to be accepted for processing. The paper aims at finding a balance between orders profit, delivery cost and time-based cost to maximize the total revenue. We consider three main objective functions in scheduling theory, analyze the problem complexity. The complexity of two of the problems are weakly NP hard, and the other one is strongly NP hard. For three weakly NP hard problems, we give pseudo-polynomial time optimal algorithms.