Search

In this paper, the dataset is collected from the fluidic muscle datasheet. This dataset is then used to train models predicting the pressure, force, and contraction length of the fluidic muscle, as three separate outputs. This modeling is performed with four algorithms—extreme gradient boosted trees (XGB), ElasticNet (ENet), support vector regressor (SVR), and multilayer perceptron (MLP) artificial neural network. Each of the four models of fluidic muscles (5-100N, 10-100N, 20-200N, 40-400N) is modeled separately: First, for a later comparison. Then, the combined dataset consisting of data from all the listed datasets is used for training. The results show that it is possible to achieve quality regression performance with the listed algorithms, especially with the general model, which performs better than individual models. Still, room for improvement exists, due to the high variance of the results across validation sets, possibly caused by non-normal data distributions.

Motor power models are a key tool in robotics for modeling and simulations related to control and optimization. The authors collect the dataset of motor power using the ABB IRB 120 industrial robot. This paper applies a multilayer perceptron (MLP) model to the collected dataset. Before the training of MLP models, each of the variables in the dataset is evaluated using the random forest (RF) model, observing two metrics-mean decrease in impurity (MDI) and feature permutation score difference (FP). Pearson’s correlation coefficient was also applied Based on the scores of these values, a total of 15 variables, mainly static variables connected with the position and orientation of the robot, are eliminated from the dataset. The scores demonstrate that while both MLPs achieve good scores, the model trained on the pruned dataset performs better. With the model trained on the pruned dataset achieving R¯2=0.99924,σ=0.00007 and MA¯PE=0.33589,σ=0.00955, the model trained on the original, non-pruned, data achieves R¯2=0.98796,σ=0.00081 and MA¯PE=0.46895,σ=0.05636. These scores show that by eliminating the variables with a low influence from the dataset, a higher scoring model is achieved, and the created model achieves a better generalization performance across five folds used for evaluation.

Malware detection using hybrid features, combining binary and hexadecimal analysis with DLL calls, is crucial for leveraging the strengths of both static and dynamic analysis methods. Artificial intelligence (AI) enhances this process by enabling automated pattern recognition, anomaly detection, and continuous learning, allowing security systems to adapt to evolving threats and identify complex, polymorphic malware that may exhibit varied behaviors. This synergy of hybrid features with AI empowers malware detection systems to efficiently and proactively identify and respond to sophisticated cyber threats in real time. In this paper, the genetic programming symbolic classifier (GPSC) algorithm was applied to the publicly available dataset to obtain symbolic expressions (SEs) that could detect the malware software with high classification performance. The initial problem with the dataset was a high imbalance between class samples, so various oversampling techniques were utilized to obtain balanced dataset variations on which GPSC was applied. To find the optimal combination of GPSC hyperparameter values, the random hyperparameter value search method (RHVS) was developed and applied to obtain SEs with high classification accuracy. The GPSC was trained with five-fold cross-validation (5FCV) to obtain a robust set of SEs on each dataset variation. To choose the best SEs, several evaluation metrics were used, i.e., the length and depth of SEs, accuracy score (ACC), area under receiver operating characteristic curve (AUC), precision, recall, f1-score, and confusion matrix. The best-obtained SEs are applied on the original imbalanced dataset to see if the classification performance is the same as it was on balanced dataset variations. The results of the investigation showed that the proposed method generated SEs with high classification accuracy (0.9962) in malware software detection.

Predicting the stability of a Decentralized Smart Grid is key to the control of such systems. One of the key aspects that is necessary when observing the control of DSG systems is the need for rapid control. Due to this, the application of AI-based machine learning (ML) algorithms may be key to achieving a quick and precise stability prediction. In this paper, the authors utilize four algorithms—a multilayer perceptron (MLP), extreme gradient boosting (XGB), support vector machines (SVMs), and genetic programming (GP). A public dataset containing 30,000 points was used, with inputs consisting of τ—the time needed for a grid participant to adjust consumption/generation, p—generated power, and γ—the price elasticity coefficient for four grid elements; and outputs consisting of stab—the eigenvalue of stability and stabf, the categorical stability of the system. The system was modeled using the aforementioned methods as a regression model (targeting stab) and a classification model (targeting stabf). Modeling was performed with and without the τ values due to their low correlation. The best results were achieved with the XGB algorithm for classification, with and without the τ values as inputs—indicating them as being unnecessary.

A thermodynamic analysis of CO2 closed-cycle gas turbine is presented in this paper. Two processes are investigated - base process and the same process upgraded with a heat regenerator. Maximum specific useful work is 159.94 kJ/kg for both observed processes. Involving of heat regenerator inside base CO2 closed-cycle gas turbine process requires attention due to required temperature differences - high pressure ratios cannot be obtained with a high efficiency heat regenerator. Base CO2 closed-cycle gas turbine process did not reach cycle efficiency higher than 25%, while for the upgraded process the cycle efficiency can reach 40% at high pressure ratio and for high regenerator efficiency. Additionally, multilayer perceptron is trained in order to achieve high quality models for estimating specific useful work and efficiency for both, base and upgraded process. As a result, MLP with three hidden layers achieved high values of R2 score.

Predicting the quality of the robot end-effector grasp quality during an industrial robot manipulator operation can be an extremely complex task. As is often the case with such complex tasks, Artificial Intelligence methods may be applied to attempt the creation of a model - if sufficient data exists. The presented dataset uses a publicly available dataset, consisting of 992632 measurements of position, torque, and velocity - for each of the three joints of three fingers of the simulated end-effector. The dataset is first analyzed and pre-processed to prepare it for model training. The duplicate values are removed from the dataset, as well as the statistical outliers. Then, a multilayer perceptron (MLP) machine learning algorithm is applied to 80% of the data contained in the dataset, using the Grid Search algorithm to determine the best combination of MLP hyperparameters. As the dataset consists of torque, velocity, and speed measurements for separate joints and fingers of the tested end-effector the testing is performed to see if a subset of the inputs may be used to regress the robustness of the given grip. The normalization of the dataset is also applied, and its effect on the regression quality is tested. The results, evaluated with the coefficient of determination, show that while the best model is achieved using all the possible inputs, a satisfactory result can be obtained using only velocity and torque.The results also show that the normalization of the dataset improves the regression quality in all the observed cases.

This paper analyses the application of a genetic algorithm (GA) for the purpose of designing the control system with separately excited DC motor controlled according to the rotor angle. The presented research is based on the utilization of a mathematical model designed with separate electrical and mechanical sub-systems. Such an approach allows fine-tuning of PID controllers by using an evolutionary procedure, mainly GA. For purpose of PID tuning, the new fitness function which combines several step response parameters with the aim of forming a unique surface which is then minimized with a genetic algorithm. From the results, it can be seen that the elitism-based algorithm achieved better results compared to the eligibility-based selection. Such an algorithm achieved a fitness value of 0.999982 resulting in a steady-state error of 0.000584 rad. The obtained results indicate the possibility of applying a GA in the parameterization of the PID controller for DC motor control.

Machine learning applications have demonstrated the potential to generate precise models in a wide variety of fields, including marine applications. Still, the main issue with ML-based methods is the need for large amounts of data, which may be impractical to come by. To assure the quality of the models and their robustness to different inputs, synthetic data may be generated using other ML-based methods, such as Triplet Encoded Variable Autoencoder (TVAE), copulas, or a Conditional Tabular Generative Adversarial Network (CTGAN). With this approach, a dataset can be trained using ML methods such as Multilayer Perceptron (MLP) or Extreme Gradient Boosting (XGB) to improve the general performance. The methods are applied to the dataset containing mass flow, temperature, and pressure measurements in seven points of a marine steam turbine as inputs, along with the exergy efficiency (η) and destruction (Ex) of the whole turbine (WT), low-pressure cylinder (LPC) and high-pressure cylinder (HPC) as outputs. The achieved results show that models trained on synthetic data achieve slightly worse results than the models trained on original data in previous research, but allow for the use of as little as two-thirds of the dataset to achieve these results. Using R2 as the main evaluation metric, the best results achieved are 0.99 for ηWT using 100 data points and MLP, 0.93 for ηLPC using 100 data points and an MLP-based model, 0.91 for ηHPC with the same method, and 0.97 for ExWT, 0.96 for ExLPC, and 0.98 for ExHPC using a the XGB trained model with 100 data points.

A common issue with X-ray examinations (XE) is the erroneous quality classification of the XE, implying that the process needs to be repeated, thus delaying the diagnostic assessment of the XE and increasing the amount of radiation the patient receives. The authors propose a system for automatic quality classification of XE based on convolutional neural networks (CNN) that would simplify this process and significantly decrease erroneous quality classification. The data used for CNN training consist of 4000 knee images obtained via radiography procedure (KXE) in total, with 2000 KXE labeled as acceptable and 2000 as unacceptable. Additionally, half of the KXE belonging to each label are right knees and left knees. Due to the sensitivity to image orientation of some CNNs, three approaches are discussed: (1) Left-right-knee (LRK) classifies XE based just on their label, without taking into consideration their orientation; (2) Orientation discriminator (OD) for the left knee (LK) and right knee (RK) analyses images based on their orientation and inserts them into two separate models regarding orientation; (3) Orientation discriminator combined with knee XRs flipped to the left or right (OD-LFK)/OD-RFK trains the models with all images being horizontally flipped to the same orientation and uses the aforementioned OD to determine whether the image needs to be flipped or not. All the approaches are tested with five CNNs (AlexNet, ResNet50, ResNet101, ResNet152, and Xception), with grid search and k-fold cross-validation. The best results are achieved using the OD-RFK hybrid approach with the Xception network architecture as the classifier and ResNet152 as the OD, with an average AUC of 0.97 (±0.01).

Nowadays, with the rise of drone and satellite technology, there is a possibility for its application in sea and coastal surveillance. An advantage of this type of application is the automated recognition of marine objects, among which the most important are vessels. This paper presents the principle of vessel recognition based on the extraction of satellite image features of the vessel and the application of a multilayer perceptron (MLP). Dataset used in this research contains the total of 2750 images, where 2112 images are used as training set while the remaining 638 images are used for testing purposes. The SIFT and SURF algorithms were used to extract image features, which were later used as the input vector for MLP.The best results are achieved if a model with four hidden layers is used. These layers are constructed with 32, 128, 32, 128 neurons with ReLU activation function, respectively. Regarding the application of feature extraction, it can be observed that better results are achieved if the SIFT algorithm is used. The ROC AUC value achieved with the combination of SIFT and MLP reaches 0.99.

The importance of error detection is high, especially in modern manufacturing processes where assembly lines operate without direct supervision. Stopping the faulty operation in time can prevent damage to the assembly line. Public dataset is used, containing 15 classes, 2 types of faultless operation and 13 types of faults, with 463 force and torsion datapoints. Four different methods are used: Multilayer Perceptron (MLP) selected due to high classification performance, Support Vector Machines (SVM) commonly used for a low number of datapoints, Convolutional Neural Network (CNN) known for high performance in classification with matrix inputs and Siamese Neural Network (SNN) novel method with high performance in small datasets. Two classification tasks are performed-error detection and classification. Grid search is used for hyperparameter variation and F1 score as a metric, with a 10 fold cross-validation. Authors propose a hybrid system consisting of SNN for detection and CNN for fault classification.

The paper presents an analysis of marine Turbo-Generator Steam Turbine (TGST) energy losses at turbine gland seals. The analyzed TGST is one of two identical Turbo-Generator Steam Turbines mounted in the steam propulsion plant of a commercial LNG carrier. Research is based on the TGST measurement data obtained during exploitation at three different loads. The turbine front gland seal is the most important element which defines TGST operating parameters, energy losses and energy efficiencies. The front gland seal should have as many chambers as possible in order to minimize the leaked steam mass flow rate, which will result in a turbine energy losses’ decrease and in an increase in energy efficiency. The steam mass flow rate leakage through the TGST rear gland seal has a low or negligible influence on turbine operating parameters, energy losses and energy efficiencies. The highest turbine energy efficiencies are noted at a high load – on which TGST operation is preferable.

The paper presents an analysis of two steam turbine operation regimes - regime with all steam extractions opened (base process) and regime with all steam extractions closed. Closing of all steam extractions significantly increases turbine real developed power for 5215.88 kW and increases turbine energy and exergy losses with simultaneous decrease of turbine energy and exergy efficiencies for more than 2%. First extracted steam mass flow rate has a dominant influence on turbine power losses (in comparison to turbine maximum power when all of steam extractions are closed). Cumulative power losses caused by steam mass flow rate extractions are the highest in the fourth turbine segment and equal to 1687.82 kW.

The Super Cryogenic Dark Matter Search (SuperCDMS) experiment is used to search for Weakly Interacting Massive Particles (WIMPs)—candidates for dark matter particles. In this experiment, the WIMPs interact with nuclei in the detector; however, there are many other interactions (background interactions). To separate background interactions from the signal, it is necessary to measure the interaction energy and to reconstruct the location of the interaction between WIMPs and the nuclei. In recent years, some research papers have been investigating the reconstruction of interaction locations using artificial intelligence (AI) methods. In this paper, a genetic programming-symbolic regression (GPSR), with randomly tuned hyperparameters cross-validated via a five-fold procedure, was applied to the SuperCDMS experiment to estimate the interaction locations with high accuracy. To measure the estimation accuracy of obtaining the SEs, the mean and standard deviation (σ) values of R2, the root-mean-squared error (RMSE), and finally, the mean absolute error (MAE) were used. The investigation showed that using GPSR, SEs can be obtained that estimatethe interaction locations with high accuracy. To improve the solution, the five best SEs were combined from the three best cases. The results demonstrated that a very high estimation accuracy can be achieved with the proposed methodology.

Path planning is one of the key steps in the application of industrial robotic manipulators. The process of determining trajectories can be time-intensive and mathematically complex, which raises the complexity and error proneness of this task. For these reasons, the authors tested the application of a genetic algorithm (GA) on the problem of continuous path planning based on the Ho–Cook method. The generation of trajectories was optimized with regard to the distance between individual segments. A boundary condition was set regarding the minimal values that the trajectory parameters can be set in order to avoid stationary solutions. Any distances between segments introduced by this condition were addressed with Bezier spline interpolation applied between evolved segments. The developed algorithm was shown to generate trajectories and can easily be applied for the further path planning of various robotic manipulators, which indicates great promise for the use of such algorithms.

In the case of pandemics such as COVID-19, the rapid development of medicines addressing the symptoms is necessary to alleviate the pressure on the medical system. One of the key steps in medicine evaluation is the determination of pIC50 factor, which is a negative logarithmic expression of the half maximal inhibitory concentration (IC50). Determining this value can be a lengthy and complicated process. A tool allowing for a quick approximation of pIC50 based on the molecular makeup of medicine could be valuable. In this paper, the creation of the artificial intelligence (AI)-based model is performed using a publicly available dataset of molecules and their pIC50 values. The modeling algorithms used are artificial and convolutional neural networks (ANN and CNN). Three approaches are tested—modeling using just molecular properties (MP), encoded SMILES representation of the molecule, and the combination of both input types. Models are evaluated using the coefficient of determination (R2) and mean absolute percentage error (MAPE) in a five-fold cross-validation scheme to assure the validity of the results. The obtained models show that the highest quality regression (R2¯=0.99, σR2¯=0.001; MAPE¯=0.009%, σMAPE¯=0.009), by a large margin, is obtained when using a hybrid neural network trained with both MP and SMILES.

Imaging is one of the main tools of modern astronomy—many images are collected each day, and they must be processed. Processing such a large amount of images can be complex, time-consuming, and may require advanced tools. One of the techniques that may be employed is artificial intelligence (AI)-based image detection and classification. In this paper, the research is focused on developing such a system for the problem of the Magellan dataset, which contains 134 satellite images of Venus’s surface with individual volcanoes marked with circular labels. Volcanoes are classified into four classes depending on their features. In this paper, the authors apply the You-Only-Look-Once (YOLO) algorithm, which is based on a convolutional neural network (CNN). To apply this technique, the original labels are first converted into a suitable YOLO format. Then, due to the relatively small number of images in the dataset, deterministic augmentation techniques are applied. Hyperparameters of the YOLO network are tuned to achieve the best results, which are evaluated as mean average precision (mAP@0.5) for localization accuracy and F1 score for classification accuracy. The experimental results using cross-vallidation indicate that the proposed method achieved 0.835 mAP@0.5 and 0.826 F1 scores, respectively.

Since the outbreak of coronavirus disease-2019 (COVID-19), the whole world has taken interest in the mechanisms of its spread and development. Mathematical models have been valuable instruments for the study of the spread and control of infectious diseases. For that purpose, we propose a two-way approach in modeling COVID-19 spread: a susceptible, exposed, infected, recovered, deceased (SEIRD) model based on differential equations and a long short-term memory (LSTM) deep learning model. The SEIRD model is a compartmental epidemiological model with included components: susceptible, exposed, infected, recovered, deceased. In the case of the SEIRD model, official statistical data available online for countries of Belgium, Netherlands, and Luxembourg (Benelux) in the period of March 15 2020 to March 15 2021 were used. Based on them, we have calculated key parameters and forward them to the epidemiological model, which will predict the number of infected, deceased, and recovered people. Results show that the SEIRD model is able to accurately predict several peaks for all the three countries of interest, with very small root mean square error (RMSE), except for the mild cases (maximum RMSE was 240.79 ± 90.556), which can be explained by the fact that no official data were available for mild cases, but this number was derived from other statistics. On the other hand, LSTM represents a special kind of recurrent neural network structure that can comparatively learn long-term temporal dependencies. Results show that LSTM is capable of predicting several peaks based on the position of previous peaks with low values of RMSE. Higher values of RMSE are observed in the number of infected cases in Belgium (RMSE was 535.93) and Netherlands (RMSE was 434.28), and are expected because of thousands of people getting infected per day in those countries. In future studies, we will extend the models to include mobility information, variants of concern, as well as a medical intervention, etc. A prognostic model could help us predict epidemic peaks. In that way, we could react in a timely manner by introducing new or tightening existing measures before the health system is overloaded.

A synchronous machine is an electro-mechanical converter consisting of a stator and a rotor. The stator is the stationary part of a synchronous machine that is made of phase-shifted armature windings in which voltage is generated and the rotor is the rotating part made using permanent magnets or electromagnets. The excitation current is a significant parameter of the synchronous machine, and it is of immense importance to continuously monitor possible value changes to ensure the smooth and high-quality operation of the synchronous machine itself. The purpose of this paper is to estimate the excitation current on a publicly available dataset, using the following input parameters: Iy: load current; PF: power factor; e: power factor error; and df: changing of excitation current of synchronous machine, using artificial intelligence algorithms. The algorithms used in this research were: k-nearest neighbors, linear, random forest, ridge, stochastic gradient descent, support vector regressor, multi-layer perceptron, and extreme gradient boost regressor, where the worst result was elasticnet, with R2 = −0.0001, MSE = 0.0297, and MAPE = 0.1442; the best results were provided by extreme boosting regressor, with R2¯ = 0.9963, MSE¯ = 0.0001, and MAPE¯ = 0.0057, respectively.

Fire is usually detected with fire detection systems that are used to sense one or more products resulting from the fire such as smoke, heat, infrared, ultraviolet light radiation, or gas. Smoke detectors are mostly used in residential areas while fire alarm systems (heat, smoke, flame, and fire gas detectors) are used in commercial, industrial and municipal areas. However, in addition to smoke, heat, infrared, ultraviolet light radiation, or gas, other parameters could indicate a fire, such as air temperature, air pressure, and humidity, among others. Collecting these parameters requires the development of a sensor fusion system. However, with such a system, it is necessary to develop a simple system based on artificial intelligence (AI) that will be able to detect fire with high accuracy using the information collected from the sensor fusion system. The novelty of this paper is to show the procedure of how a simple AI system can be created in form of symbolic expression obtained with a genetic programming symbolic classifier (GPSC) algorithm and can be used as an additional tool to detect fire with high classification accuracy. Since the investigation is based on an initially imbalanced and publicly available dataset (high number of samples classified as 1-Fire Alarm and small number of samples 0-No Fire Alarm), the idea is to implement various balancing methods such as random undersampling/oversampling, Near Miss-1, ADASYN, SMOTE, and Borderline SMOTE. The obtained balanced datasets were used in GPSC with random hyperparameter search combined with 5-fold cross-validation to obtain symbolic expressions that could detect fire with high classification accuracy. For this investigation, the random hyperparameter search method and 5-fold cross-validation had to be developed. Each obtained symbolic expression was evaluated on train and test datasets to obtain mean and standard deviation values of accuracy (ACC), area under the receiver operating characteristic curve (AUC), precision, recall, and F1-score. Based on the conducted investigation, the highest classification metric values were achieved in the case of the dataset balanced with SMOTE method. The obtained values of ACC¯±SD(ACC), AUC¯±SD(ACU), Precision¯±SD(Precision), Recall¯±SD(Recall), and F1-score¯±SD(F1-score) are equal to 0.998±4.79×10−5, 0.998±4.79×10−5, 0.999±5.32×10−5, 0.998±4.26×10−5, and 0.998±4.796×10−5, respectively. The symbolic expression using which best values of classification metrics were achieved is shown, and the final evaluation was performed on the original dataset.

Hepatitis C is an infectious disease which is caused by the Hepatitis C virus (HCV) and the virus primarily affects the liver. Based on the publicly available dataset used in this paper the idea is to develop a mathematical equation that could be used to detect HCV patients with high accuracy based on the enzymes, proteins, and biomarker values contained in a patient’s blood sample using genetic programming symbolic classification (GPSC) algorithm. Not only that, but the idea was also to obtain a mathematical equation that could detect the progress of the disease i.e., Hepatitis C, Fibrosis, and Cirrhosis using the GPSC algorithm. Since the original dataset was imbalanced (a large number of healthy patients versus a small number of Hepatitis C/Fibrosis/Cirrhosis patients) the dataset was balanced using random oversampling, SMOTE, ADSYN, and Borderline SMOTE methods. The symbolic expressions (mathematical equations) were obtained using the GPSC algorithm using a rigorous process of 5-fold cross-validation with a random hyperparameter search method which had to be developed for this problem. To evaluate each symbolic expression generated with GPSC the mean and standard deviation values of accuracy (ACC), the area under the receiver operating characteristic curve (AUC), precision, recall, and F1-score were obtained. In a simple binary case (healthy vs. Hepatitis C patients) the best case was achieved with a dataset balanced with the Borderline SMOTE method. The results are ACC¯±SD(ACC), AUC¯±SD(AUC), Precision¯±SD(Precision), Recall¯±SD(Recall), and F1−score¯±SD(F1−score) equal to 0.99±5.8×10−3, 0.99±5.4×10−3, 0.998±1.3×10−3, 0.98±1.19×10−3, and 0.99±5.39×10−3, respectively. For the multiclass problem, OneVsRestClassifer was used in combination with GPSC 5-fold cross-validation and random hyperparameter search, and the best case was achieved with a dataset balanced with the Borderline SMOTE method. To evaluate symbolic expressions obtained in this case previous evaluation metric methods were used however for AUC, Precision, Recall, and F1−score the macro values were computed since this method calculates metrics for each label, and find their unweighted mean value. In multiclass case the ACC¯±SD(ACC), AUC¯macro±SD(AUC), Precision¯macro±SD(Precision), Recall¯macro±SD(Recall), and F1−score¯macro±SD(F1−score) are equal to 0.934±9×10−3, 0.987±1.8×10−3, 0.942±6.9×10−3, 0.934±7.84×10−3 and 0.932±8.4×10−3, respectively. For the best binary and multi-class cases, the symbolic expressions are shown and evaluated on the original dataset.

This article describes the implementation of the You Only Look Once (YOLO) detection algorithm for the detection of returnable packaging. The method of creating an original dataset and creating an augmented dataset is shown. The model was evaluated using mean Average Precision (mAP), F1score, Precision, Recall, Average Intersection over Union (Average IoU) score, and Average Loss. The training was conducted in four cycles, i.e., 6000, 8000, 10,000, and 20,000 max batches with three different activation functions Mish, ReLU, and Linear (used in 6000 and 8000 max batches). The influence train/test dataset ratio was also investigated. The conducted investigation showed that variation of hyperparameters (activation function and max batch sizes) have a significant influence on detection and classification accuracy with the best results obtained in the case of YOLO version 4 (YOLOV4) with the Mish activation function and max batch size of 20,000 that achieved the highest mAP of 99.96% and lowest average error of 0.3643.

Inverse kinematic equations allow the determination of the joint angles necessary for the robotic manipulator to place a tool into a predefined position. Determining this equation is vital but a complex work. In this article, an artificial neural network, more specifically, a feed-forward type, multilayer perceptron (MLP), is trained, so that it could be used to calculate the inverse kinematics for a robotic manipulator. First, direct kinematics of a robotic manipulator are determined using Denavit–Hartenberg method and a dataset of 15,000 points is generated using the calculated homogenous transformation matrices. Following that, multiple MLPs are trained with 10,240 different hyperparameter combinations to find the best. Each trained MLP is evaluated using the R 2 and mean absolute error metrics and the architectures of the MLPs that achieved the best results are presented. Results show a successful regression for the first five joints (percentage error being less than 0.1%) but a comparatively poor regression for the final joint due to the configuration of the robotic manipulator.

In this paper, a comparison of two different methods for a steam turbine energy analysis is presented. A high-pressure steam turbine from a supercritical thermal power plant (HPT) was analysed at three different turbine loads using the energy flow stream (EFS) method and isentropic (IS) method. The EFS method is based on steam turbine input and output energy flow streams and on the real steam turbine produced power. The method is highly dependable on the steam mass flow rate lost through the turbine gland seals. The IS method is based on a comparison of turbine steam expansion processes. Observed energy analysis methods cannot be directly compared because they are based on different sources of steam turbine energy losses, so, an overall steam turbine energy analysis is presented. Unlike most steam turbines from the literature, the analysed HPT did not have the highest overall energy efficiency at a full load due to exceeding the water/steam critical pressure at the turbine inlet during such operation.

Vaccinations are one of the most important steps in combat against viral diseases such as COVID-19. Determining the influence of the number of vaccinated patients on the infected population represents a complex problem. For this reason, the aim of this research is to model the influence of the total number of vaccinated or fully vaccinated patients on the number of infected and deceased patients. Five separate modeling algorithms are used: Linear Regression (LR), Logistic Regression (LogR), Least Absolute Shrinkage and Selection Operator (LASSO), Multilayer Perceptron (MLP), and Support Vector Regression (SVR). Cross-correlation analysis is performed to determine the optimal lags in data to assist in obtaining better scores. The cross-validation of models is performed, and the models are evaluated using Mean Absolute Percentage Error (MAPE). The modeling is performed for four different countries: Germany, India, the United Kingdom (UK), and the United States of America (USA). Models with an error below 1% are found for all the modeled cases, with the best models being achieved either by LR or MLP methods. The obtained results indicate that the influence of vaccination rates on the number of confirmed and deceased patients exists and can be modeled using ML methods with relatively high precision.

Lowering joint torques of a robotic manipulator enables lowering the energy it uses as well as increase in the longevity of the robotic manipulator. This article proposes the use of evolutionary computation algorithms for optimizing the paths of the robotic manipulator with the goal of lowering the joint torques. The robotic manipulator used for optimization is modelled after a realistic six-degree-of-freedom robotic manipulator. Two cases are observed and these are a single robotic manipulator carrying a weight in a point-to-point trajectory and two robotic manipulators cooperating and moving the same weight along a calculated point-to-point trajectory. The article describes the process used for determining the kinematic properties using Denavit–Hartenberg method and the dynamic equations of the robotic manipulator using Lagrange–Euler and Newton–Euler algorithms. Then, the description of used artificial intelligence optimization algorithms is given – genetic algorithm using random and average recombination, simulated annealing using linear and geometric cooling strategy and differential evolution. The methods are compared and the results show that the genetic algorithm provides best results in regard to torque minimization, with differential evolution also providing comparatively good results and simulated annealing giving the comparatively weakest results while providing smoother torque curves.

COVID-19 is an infectious disease that mostly affects the respiratory system. At the time of this research being performed, there were more than 1.4 million cases of COVID-19, and one of the biggest anxieties is not just our health, but our livelihoods, too. In this research, authors investigate the impact of COVID-19 on the global economy, more specifically, the impact of COVID-19 on the financial movement of Crude Oil price and three US stock indexes: DJI, S&P 500, and NASDAQ Composite. The proposed system for predicting commodity and stock prices integrates the stationary wavelet transform (SWT) and bidirectional long short-term memory (BDLSTM) networks. Firstly, SWT is used to decompose the data into approximation and detail coefficients. After decomposition, data of Crude Oil price and stock market indexes along with COVID-19 confirmed cases were used as input variables for future price movement forecasting. As a result, the proposed system BDLSTM + WT-ADA achieved satisfactory results in terms of five-day Crude Oil price forecast.

This paper describes the implementation of magnetic bearings and their active control system. The working principle of the special combined axial and radial magnetic bearings and their implementation is explained. The overall laboratory system consists of two magnetic bearings with their housings, the suspended shaft and control system with sensors and power electronics. Sensor electronics are used to measure the position of the shaft within a bearing and to determine the appropriate bearing current with outer position control loops and nested current control loops. The motor control board featuring the TMS320F28335 DSP and LDC1000 proximity sensors is used for control and power electronics. In order to implement the control system the X2C tool is used. This is an open source model based development and code generation tool embedded in the Scilab/Xcos environment. The sensor data acquisition time is optimized to increase the speed of the control system.

Research on the implementation of inductive proximity sensors in active magnetic bearings is performed in this paper. LDC1000 digital sensors manufactured by Texas Instruments are used since they claim to provide precision measured in microns and simple acquisition of the output data. The Sensor chip is connected with the sensing coil which is placed inside the magnetic bearing, close to electromagnets. This newly developed system created a platform for research in signal quality and its resistance to electrical noise. The controller used in this application was Texas Instruments TMS320F28335 digital signal processor (DSP). Software for data acquisition from LDC1000 was written and implemented in DSP using Code Composer Studio development environment. Two different output data are acquired and processed: proximity data and frequency count. Graphs presented in this paper show different resistance to electrical noise. A conclusion derived from this research can be applied in an industry where electromagnetic noise exists, together with the need for precise distance measurement.

Urinary bladder cancer is one of the most common cancers of the urinary tract. This cancer is characterized by its high metastatic potential and recurrence rate. Due to the high metastatic potential and recurrence rate, correct and timely diagnosis is crucial for successful treatment and care. With the aim of increasing diagnosis accuracy, artificial intelligence algorithms are introduced to clinical decision making and diagnostics. One of the standard procedures for bladder cancer diagnosis is computer tomography (CT) scanning. In this research, a transfer learning approach to the semantic segmentation of urinary bladder cancer masses from CT images is presented. The initial data set is divided into three sub-sets according to image planes: frontal (4413 images), axial (4993 images), and sagittal (996 images). First, AlexNet is utilized for the design of a plane recognition system, and it achieved high classification and generalization performances with an AUCmicro¯ of 0.9999 and σ(AUCmicro) of 0.0006. Furthermore, by applying the transfer learning approach, significant improvements in both semantic segmentation and generalization performances were achieved. For the case of the frontal plane, the highest performances were achieved if pre-trained ResNet101 architecture was used as a backbone for U-net with DSC¯ up to 0.9587 and σ(DSC) of 0.0059. When U-net was used for the semantic segmentation of urinary bladder cancer masses from images in the axial plane, the best results were achieved if pre-trained ResNet50 was used as a backbone, with a DSC¯ up to 0.9372 and σ(DSC) of 0.0147. Finally, in the case of images in the sagittal plane, the highest results were achieved with VGG-16 as a backbone. In this case, DSC¯ values up to 0.9660 with a σ(DSC) of 0.0486 were achieved. From the listed results, the proposed semantic segmentation system worked with high performance both from the semantic segmentation and generalization standpoints. The presented results indicate that there is the possibility for the utilization of the semantic segmentation system in clinical practice.

In this paper, the publicly available dataset for the Combined Diesel-Electric and Gas (CODLAG) propulsion system was used to obtain symbolic expressions for estimation of fuel flow, ship speed, starboard propeller torque, port propeller torque, and total propeller torque using genetic programming (GP) algorithm. The dataset consists of 11,934 samples that were divided into training and testing portions in an 80:20 ratio. The training portion of the dataset which consisted of 9548 samples was used to train the GP algorithm to obtain symbolic expressions for estimation of fuel flow, ship speed, starboard propeller, port propeller, and total propeller torque, respectively. After the symbolic expressions were obtained the testing portion of the dataset which consisted of 2386 samples was used to measure estimation performance in terms of coefficient of correlation (R2) and Mean Absolute Error (MAE) metric, respectively. Based on the estimation performance in each case three best symbolic expressions were selected with and without decay state coefficients. From the conducted investigation, the highest R2 and lowest MAE values were achieved with symbolic expressions for the estimation of fuel flow, ship speed, starboard propeller torque, port propeller torque, and total propeller torque without decay state coefficients while symbolic expressions with decay state coefficients have slightly lower estimation performance.

Urinary bladder cancer is one of the most common urinary tract cancers. Standard diagnosis procedure can be invasive and time-consuming. For these reasons, procedure called optical biopsy is introduced. This procedure allows in-vivo evaluation of bladder mucosa without the need for biopsy. Although less invasive and faster, accuracy is often lower. For this reason, machine learning (ML) algorithms are used to increase its accuracy. The issue with ML algorithms is their sensitivity to the amount of input data. In medicine, collection can be time-consuming due to a potentially low number of patients. For these reasons, data augmentation is performed, usually through a series of geometric variations of original images. While such images improve classification performance, the number of new data points and the insight they provide is limited. These issues are a motivation for the application of novel augmentation methods. Authors demonstrate the use of Deep Convolutional Generative Adversarial Networks (DCGAN) for the generation of images. Augmented datasets used for training of commonly used Convolutional Neural Network-based (CNN) architectures (AlexNet and VGG-16) show a significcan performance increase for AlexNet, where AUCmicro reaches values up to 0.99. Average and median results of networks used in grid-search increases. These results point towards the conclusion that GAN-based augmentation has decreased the networks sensitivity to hyperparemeter change.

This article presented an improvement of marine steam turbine conventional exergy analysis by application of neural networks. The conventional exergy analysis requires numerous measurements in seven different turbine operating points at each load, while the intention of MLP (Multilayer Perceptron) neural network-based analysis was to investigate the possibilities for measurements reducing. At the same time, the accuracy and precision of the obtained results should be maintained. In MLP analysis, six separate models are trained. Due to a low number of instances within the data set, a 10-fold cross-validation algorithm is performed. The stated goal is achieved and the best solution suggests that MLP application enables reducing of measurements to only three turbine operating points. In the best solution, MLP model errors falling within the desired error ranges (Mean Relative Error) MRE < 2.0% and (Coefficient of Correlation) R2 > 0.95 for the whole turbine and each of its cylinders.

In this paper a genetic algorithm (GA) approach to design of multi-layer perceptron (MLP) for combined cycle power plant power output estimation is presented. Dataset used in this research is a part of publicly available UCI Machine Learning Repository and it consists of 9568 data points (power plant operating regimes) that is divided on training dataset that consists of 7500 data points and testing dataset containing 2068 data points. Presented research was performed with aim of increasing regression performances of MLP in comparison to ones available in the literature by utilizing heuristic algorithm. The GA described in this paper is performed by using mutation and crossover procedures. These procedures are utilized for design of 20 different chromosomes in 50 different generations. MLP configurations that are designed with GA implementation are validated by using Bland - Altman (B-A) analysis. By utilizing GA, MLP with five hidden layers of 80,25,65,75 and 80 nodes, respectively, is designed. For aforementioned MLP, k - fold cross-validation is performed in order to examine its generalization performances. The Root Mean Square Error ( R M S E ) value achieved with aforementioned MLP is 4.305 , that is significantly lower in comparison with MLP presented in available literature, but still higher than several complex algorithms such as KStar and tree based algorithms.