• Smart Structures and Systems
• /
• 제29권6호
• /
• pp.799-805
• /
• 2022

Permanent magnet synchronous motors (PMSMs) are widely used in systems requiring high control precision, efficiency, and reliability. Predicting the remaining useful life (RUL) with health monitoring of PMSMs prevents catastrophic failure and ensures reliable operation of system. In this study, a model-based method for predicting the RUL of PMSMs using phase current and vibration signals is proposed. The proposed method includes feature selection and RUL prediction based on a particle filter with a degradation model. The Paris-Erdogan model describing micro fatigue crack propagation is used as the degradation model. An experimental set-up to conduct accelerated life test, capable of monitoring various signals was designed in this study. Phase current and vibration data obtained from an accelerated life test of the PMSMs were used to verify the proposed approach. Features extracted from the data were clustered based on monotonicity and correlation clustering, respectively. The results identify the effectiveness of using the current data in predicting the RUL of PMSMs.

• 한국재료학회지
• /
• 제34권7호
• /
• pp.363-369
• /
• 2024

In this study, the magnetocaloric effect and transition temperature of bulk metallic glass, an amorphous material, were predicted through machine learning based on the composition features. From the Python module 'Matminer', 174 compositional features were obtained, and prediction performance was compared while reducing the composition features to prevent overfitting. After optimization using RandomForest, an ensemble model, changes in prediction performance were analyzed according to the number of compositional features. The R2 score was used as a performance metric in the regression prediction, and the best prediction performance was found using only 90 features predicting transition temperature, and 20 features predicting magnetocaloric effects. The most important feature when predicting magnetocaloric effects was the 'Fe' compositional ratio. The feature importance method provided by 'scikit-learn' was applied to sort compositional features. The feature importance method was found to be appropriate by comparing the prediction performance of the Fe-contained dataset with the full dataset.

• Ocean Systems Engineering
• /
• 제3권2호
• /
• pp.79-95
• /
• 2013

Suction anchors are widely adopted in mooring systems. However there are still challenges in predicting the failure mode and ultimate pullout capacity of the anchor. Previously published methods for predicting the inclined pullout capacity of suction anchors are mainly based on experimental data or the FEM analysis. In the present work, an analytical method that is capable of predicting the failure mode and ultimate pullout capacity of the suction anchor in clay under inclined loading is developed. This method is based on a rational mechanical model for suction anchors and the knowledge of the mechanism that the anchor fails in seabed soils. In order to examine the analytical model, the failure angle and pullout capacity of suction anchors from FEM simulation, numerical solution and laboratory tests in uniform and linear cohesive soils are employed to compare with the theoretical predictions and the agreement is satisfactory. An analytical method that can evaluate the optimal position of the attachment point is also proposed in the present study. The present work proves that the failure mode and pullout capacity of suction anchors can be reasonably determined by the developed analytical method.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권4호
• /
• pp.288-294
• /
• 2014

Understanding a ship's present position has been one of the most important tasks during a ship's voyage, in both ancient and modern times. Particularly, a ship's dead reckoning (DR) has been used for predicting traffic situations and collision avoidance actions. However, the current system that uses the traditional method of calculating DR employs the received position and speed data only. Therefore, it is not applicable for predicting navigation within the harbor limits, owing to the frequent changes in the ship's course and speed in this region. In this study, planned routes were applied for improving the reliability of the proposed system and predicting the traffic patterns in advance. The proposed method of determining the dead reckoning position (DRP) uses not only the ships' received data but also the navigational patterns and tracking data in harbor limits. The Mercator sailing formulas were used for calculating the ships' DRPs and planned routes. The data on the traffic patterns were collected from the automatic identification system and analyzed using MATLAB. Two randomly chosen ships were analyzed for simulating their tracks and comparing the DR method during the timeframes of the ships' movement. The proposed method of calculating DR, combined with the information on planned routes and DRPs, is expected to contribute towards improving the decision-making abilities of operators.

• Journal of Electrical Engineering and Technology
• /
• 제13권3호
• /
• pp.1089-1098
• /
• 2018

This study aims at developing and applying a hybrid model to the wind power prediction (WPP). The hybrid model for a very-short-term WPP (VSTWPP) is achieved through analytical data, multiple linear regressions and least square methods (MLR&LS). The data used in our hybrid model are based on the historical records of wind power from an offshore region. In this model, the WPP is achieved in four steps: 1) transforming historical data into ratios; 2) predicting the wind power using the ratios; 3) predicting rectification ratios by the total wind power; 4) predicting the wind power using the proposed rectification method. The proposed method includes one-step and multi-step predictions. The WPP is tested by applying different models, such as the autoregressive moving average (ARMA), support vector machine (SVM), and artificial neural network (ANN). The results of all these models confirmed the validity of the proposed hybrid model in terms of error as well as its effectiveness. Furthermore, forecasting errors are compared to depict a highly variable WPP, and the correlations between the actual and predicted wind powers are shown. Simulations are carried out to definitely prove the feasibility and excellent performance of the proposed method for the VSTWPP versus that of the SVM, ANN and ARMA models.

• 한국안전학회지
• /
• 제19권3호
• /
• pp.89-94
• /
• 2004

Most reinforcements in concrete are constructed by steel. Corrosion of reinforcement is the main cause of damage and early failure of reinforced concrete structures. The corrosion is mainly professed by the chloride ingress. In general, chloride in concrete can be discriminated by two components, total chloride and fire chloride. This paper provides a testing method on the coefficient of chloride diffusion in concrete and the relationship between total chloride and free chloride in concrete for the composition of predicting model on diffusion rate of chloride. In order to complete this predicting model, this study will use chloride penetration characteristic, diffusion coefficient and experiment of color change on silver nitrate solution. This predicting model is going to help that grasp special quality on salt content inclusion of concrete structure that is exposed in chloride environment. Accurate predicting model can be effectively used not only in selecting of repair time but also in preventing from various deteriorations.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2002년도 춘계학술대회 논문집
• /
• pp.140-145
• /
• 2002

At present, wind tunnel test or CFD is used for predicting aerodynamic drag coefficient in motor company. But, wind tunnel test requires much cost and time, and CFD has about 30% error. In this study a predicting program of the aerodynamic drag coefficient based on empirical techniques was developed. Also a mathematical optimization method using GRG method was added to the program. The program was applied to six cars. Aerodynamic drag coefficient values of six cars were Predicted with 4.857% average error. The optimization method was also applied to six cars. Three parameters selected from sensitivity analysis were determined to reduce the afterbody drag coefficient to the value established by a designer and when some parameters were changed for a developing automotive, optimal modifiable parameters were determined to preserve the same drag coefficient as the original automotive. It was verified that this program could predict the aerodynamic drag coefficient effectively and accurately, and this program with GRG method could determine optimal values of parameters.

• 한국생물정보학회:학술대회논문집
• /
• 한국생물정보시스템생물학회 2005년도 BIOINFO 2005
• /
• pp.221-227
• /
• 2005

THEMATICS is a simple computational method for predicting functional sites in proteins. The method computes the theoretical titration curves of the ionizable residues of a protein using its 3D structure, determines the residues with perturbed, non-Henderson-Hasselbalch titration behavior, and identifies clusters of these perturbed residues in physical proximity. We have shown previously that this method is highly successful in predicting catalytic sites in enzymes. In the present study, we apply the method to non-catalytic ligand-binding proteins. It is shown that THEMATICS can predict non-catalytic binding sites. The success rate is better than 80 % for a set of 30 non-catalytic, ligand-binding proteins. The application of the method to Glutamine-binding protein from E. coli is discussed in detail.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 춘계학술대회논문집
• /
• pp.877-882
• /
• 2002

A new estimation model of predicting the sound absorption performance for multiple perforated plate sound absorbing system was developed using transfer matrix method. The proposed method was validated by comparing the calculated absorption coefficients of a single layer perforated plate with the values measured by the two-microphone impedance tube method far various porosity and cavity depth. The developed transfer matrix method was further applied to estimate the multiple layer perforated plates and it is shown that the estimated absorption coefficients generally agree well with the measured values.

• Steel and Composite Structures
• /
• 제20권3호
• /
• pp.469-486
• /
• 2016

Codes EN 1993 and EN 1994 require to take into account actual joint characteristics in the global analysis. In order to implement the semi-rigid connection effects in frame design, knowledge of joint rotation characteristics ($M-{\phi}$ relationship), or at least three basic joint properties, namely the moment resistance $M_R$, the rotational stiffness $S_j$ and rotation capacity, is required. To avoid expensive experimental tests many methods for predicting joint parameters were developed. The paper presents a comprehensive analytical model that has been developed for predicting the moment resistance $M_R$, initial stiffness $S_{j.ini}$ and rotation capacity of the minor axis, composite, semi-rigid joint. This model is based on so-called component method included in EN 1993 and EN 1994. Comparison with experimental test results shows that a quite good agreement was achieved. A computer program POWZ containing proposed procedure were created. Based on the numerical simulation made with the use of this program and applying regression analysis, simplified equations for main joint properties were also developed.