TABLE 4 - uploaded by Hamdi Tolga Kahraman
Content may be subject to copyright.
Source publication
In the classic ANN-based approaches, the synchronous motor parameters mostly could be modeled with n-hidden layered networks. It is an important challenge in driver software development is to realize complex mathematical models in real time environments and circuits. This paper presents an Adaptive Artificial Neural Network-based (AANN) method to e...
Context in source publication
Similar publications
Direct numerical simulation of turbulent flows requires tremendous amount of computational power because it is necessary to resolve the spatial structures of the flow down to the Kolmogorov length scale. This procedure generates the socalled grids or meshes. There are two competing grids favour, namely a collocated and staggered grid. To date, it i...
This paper investigates advanced German learners’ use of the progressive in spoken English by analyzing the German error-tagged subcorpus of LINDSEI-GE compared to the native speaker counterpart LOCNEC. The analysis reveals a significant underuse of the progressive in the spoken language. Our qualitative error analysis shows that the error patterns...
This research finds the spectrum and the net intensity of the emitted radiations from the earth surface at different temperatures of 273, 283, 293, 303, 313, and
323 K. The intensity of the radiations emitted from the surface varies by 272.013 Wm-2 between 273 K and 323 K surface temperatures at emissivity 0.9. The average intensity for emissivity...
We propose a multi-precision extension of the Quadratic Regularization (R2) algorithm that enables it to take advantage of low-precision computations, and by extension to decrease energy consumption during the solve.
The lower the precision in which computations occur, the larger the errors induced in the objective value and gradient, as well as i...
In order to effectively improve the prediction accuracy with the multi-type meteorological information provided by the NWP, this paper presents a short-term wind power prediction based on the combination of the selection and optimization of the sensitive meteorological characteristics. Firstly, the Spearman, the information entropy and the Pearson...
Citations
... From three different models (Potier, ASA, and SVR), SVR performed the best with mean average error (MAE) and root mean square error (RMSE) metrics of 0.448 and 0.009, respectively. With the use of artificial neural networks (ANN), it is possible to achieve a model for estimating the parameters of an SM; in [37], the authors use an adaptive artificial neural network (AANN). The purpose of this research was to estimate the excitation current and help designers with the modulation of the excitation current while developing sophisticated software with a low degree of programming and improving the efficiency of the classic ANN-based approach. ...
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.
... Intelligent tutoring systems (ITSs) are computer-based instructional systems that provide adaptive instruction and timely feedback to learners with reduced intervention from a human teacher [1]. To provide this personalized instruction, ITSs must evaluate students to inform decisions regarding the specific adaptations needed by each individual student [2]. One major challenge for ITSs in the domain of reading is dynamically assessing students' reading skills. ...
iSTART is a game-based intelligent tutoring system (ITS) designed to improve students’ reading skills by providing training on reading comprehension strategies. Game-based practice in iSTART follows two main approaches: generative practice and identification practice. Generative practice games ask students to author self-explanations using one or more of the instructed strategies. Identification practice games require students to recognize or select appropriate strategies based on their analysis of example texts. This study explored the feasibility of implementing stealth assessments in iSTART using only an identification game. Specifically, this study examined the extent to which participants’ performance and attitudes related to a simple vocabulary game could predict the outcomes of standardized reading assessments. MTurk participants (N = 211) played identification games in iSTART and then rated their subjective gameplay experience. Participants also completed measures of their vocabulary and reading comprehension skills. Results indicated that participants’ performance in a vocabulary practice game was predictive of literacy skills. In addition, the possibility that students’ attitude towards the game moderated the relation between game performance and literacy skills was ruled out. These findings argue for the feasibility of implementing stealth assessment in simple games to facilitate the adaptivity of ITSs.
... The operation mode of synchronous motors to a large extent determines and forms the automatic excitation control (AEC). The main requirements for such systems can be formulated as follows [12][13][14][15]: 1) AEC should provide stable operation of SMs under given load conditions. ...
... Meta-heuristic algorithms are a class of optimization algorithms that are inspired by physical phenomena or biological behaviors in nature, such as MFO [11], SOS [12], GA [13], PSO [14], FA [15], BA [16], GWO [17], etc. They are widely used in machine learning [18] [19] [20] [21], data mining [22] [23], engineering design [24] [25] [26], industrial control [27] [28], and power systems [29] due to superior optimization performance. The swarm intelligence optimization algorithm is a typical representative in the meta-heuristic algorithm, and these algorithms are simple, flexible, easy to implement but not easy to fall into local optimum. ...
Density-Based Spatial Clustering of Applications with Noiseis a typical kind of algorithm based on density clustering in unsupervised learning. It can cluster data of arbitrary shape and also identify noise samples in the dataset. However, an unavoidable defect of the DBSCAN algorithm exists since the clustering performance is quite sensitive to the parameter settings of MinPts and Eps, and there is no theory to guide the setting of its parameters. Therefore, a new method is proposed to optimize the DBSCAN param-eters in this paper. Multi-Verse Optimizer algorithm, a special variable updating method with excellent optimization performance, is selected and improved for optimizing the parameters of DBSCAN, which not only can quickly find out the highest clustering accuracy of DBSCAN, but also find the interval of Eps cor-responding to the highest accuracy. In order to search the range of Eps more quickly and efficiently, we design a new mechanism for the variable update of MVO. The experimental results show that the improved MVO is used to optimize DBSCAN, which not only can quickly find out its highest clustering accuracy but also can search the parameters MinPts and Eps corresponding to the highest clustering accuracy efficiently.
... The block diagram of inner structure in a PID controller is shown in Figure 3. The output of PID controller can be defined as in Equation 4 [22]. ...
The increasing need for energy requires using existing energy sources more efficiently. Because it is the active power that supplies useful power for industrial facilities, reactive power must be minimized, and supplied by another source instead of electrical grid. Therefore, reactive power supplied by the grid can be reduced via by correcting power factor of the grid. In electrical power systems, power factor correction is called reactive power compensation. Generating reactive power during excessive excitation, synchronous motors are used as dynamic compensators in power systems. Synchronous motors are more cost-effective for industrial facilities when they are used to generate mechanic power and compensate reactive power, which increases the efficiency of industrial facilities. There are various studies focusing on the efficiency, capacity and stability of the power system via reactive power compensation in the literature. In today's world, there are numerous optimization techniques inspired by biological systems. One of these techniques is Particle Swarm Optimization (PSO) inspired by the movements of swarms of birds. This study focuses on the reactive power compensation of a power system by controlling the excitation current of a synchronous motor via PSO based PID and Ziegler Nichols (Z-N) based PID controllers.
... The comparative results ANN-based technique, classic k-NN-based estimator and the proposed IKE method affirm the superior performance of IKE method over the others. To model the SM excitation current in an easier way, a simple ANN with one hidden layer and 6 hidden nodes is suggested in [13] where the activation functions of the hidden neurons are determined by using GA. After training the resulting ANN with 394 samples and testing with 200 test data, the estimation accuracies are found to be approximately similar to those reported in [9,11,12] despite having less number of hidden layers and neurons. ...
In this paper, an effective and simpler means of estimating the excitation current of a synchronous motor (SM) is presented for power factor correction task. First, a multiple linear regression model with four predictor variables such as motor load current, actual power factor, power factor error and excitation current change is formed to estimate the SM excitation current. Then, recently introduced symbiotic organisms search (SOS) algorithm is employed in the hope of searching better values of regression coefficients in that model using the data collected from the prepared experimental setup. The supremacy of SOS over some recently published algorithms such as genetic algorithm, artificial bee colony and gravitational search algorithm is widelyattested through comparative computer simulations for the similar compensation system. The results exhibited in this article show that the proposed SOS algorithm outperforms the other reported popular algorithms from the aspects of simplicity, robustness and accuracy. In view of this, the suggested tuning of regression coefficients of the multiple linear regression model yields a better estimating performance of SM excitation current than the earlier studies.
... Thus, they have been commonly used in the power factor correction task [1][2][3]. The wide variety of applications of SMs as reactive power compensators makes it necessary to achieve a fast and reliable parameter modeling system design [4][5][6][7]. The usage of SMs in industrial applications leads to a poor power factor. ...
... It is an important problem in terms of cost, efficiency, electrical overload, and capacity. This configuration in the power line also requires the increasing of the capacities of power breakers, transformers, relays, and isolations [2][3][4][5][8][9][10][11]. There are a number of methods used in the power factor correction task in order to reduce cost and improve efficiency. ...
... The relationships among the SM parameters are mostly complex and nonlinear [5,[12][13][14][15]. Researchers have suggested artificial intelligence (AI)-based nonlinear modeling techniques, such as proportional plus integral plus derivative [16], pulse width modulation [17][18][19][20], fuzzy logic [2,3], Kalman filter-based methods [7,15,21], artificial neural networks (ANNs) [22,23], particle swarm optimization (in real-time applications) [24], intuitive k-nearest neighbor (k-NN) estimator and genetic algorithm (GA) [5,25], and adaptive ANNs [4] for modeling the parameters and/or predicting the excitation current of SMs and permanent magnet synchronous machines. The modeling of SM parameters using modern AI-based methods for excitation current estimation was realized in recently published studies [4,5,17]. ...
... Thus, they have been commonly used in the power factor correction task [1][2][3]. The wide variety of applications of SMs as reactive power compensators makes it necessary to achieve a fast and reliable parameter modeling system design [4][5][6][7]. The usage of SMs in industrial applications leads to a poor power factor. ...
... It is an important problem in terms of cost, efficiency, electrical overload, and capacity. This configuration in the power line also requires the increasing of the capacities of power breakers, transformers, relays, and isolations [2][3][4][5][8][9][10][11]. There are a number of methods used in the power factor correction task in order to reduce cost and improve efficiency. ...
... The relationships among the SM parameters are mostly complex and nonlinear [5,[12][13][14][15]. Researchers have suggested artificial intelligence (AI)-based nonlinear modeling techniques, such as proportional plus integral plus derivative [16], pulse width modulation [17][18][19][20], fuzzy logic [2,3], Kalman filter-based methods [7,15,21], artificial neural networks (ANNs) [22,23], particle swarm optimization (in real-time applications) [24], intuitive k-nearest neighbor (k-NN) estimator and genetic algorithm (GA) [5,25], and adaptive ANNs [4] for modeling the parameters and/or predicting the excitation current of SMs and permanent magnet synchronous machines. The modeling of SM parameters using modern AI-based methods for excitation current estimation was realized in recently published studies [4,5,17]. ...
The subject of modeling and estimating of synchronous motor (SM) parameters is a challenge mathematically.Although effective solutions have been developed for nonlinear systems in artificial intelligence (AI)-based models,problems are faced with the application of these models in power circuits in real-time. One of these problems is thedelay time resulting from a complex calculation process and thus the difficulties faced in the design of real-time motordriving circuits. Another important problem regards the difficulty in the realization of a complex AI-based model inmicroprocessor-based real-time systems. In this study, a new hybrid technique is developed to solve the problems inAI-based nonlinear modeling approaches. Through this method, the relationships among the motor parameters can bedescribed in a linear/quadratic SM form. The most effective and modern metaheuristic methods are utilized in thecreation of SM forms. The SM forms developed in this study lead to an easy design and application of the SM driversoftware. Therefore, a model that is faster, more effective, and more easily applicable than AI-based popular methods isdeveloped for SMs. The proposed techniques can also be applied to many other industrial modeling problems that havenonlinear features.
... Therefore, the heuristic unit has been extended. In the proposed heuristic unit, the activation function types of the nodes are heuristically changed as well as previous studies (Ustun 2009;Bayindir et al. 2012). Additionally, the number of hidden layers and nodes are also adaptable by the genetic algorithm-based (GA) heuristic exploring unit. ...
... Additionally, the number of hidden layers and nodes are also adaptable by the genetic algorithm-based (GA) heuristic exploring unit. In this study, a GA-based heuristic approach was extended to explore the optimum parameters of ANN (Ustun 2009;Bayindir et al. 2012;Goldberg 1989;Tu and Lu 2004;Mitchell 1997;Babu and Suresh 2013;Iyer and Sharda 2009;Peng and Dubay 2012). The first aim of the study is to determine the using possibilities of IKE and AANN methods for prediction of bonding strength of plywood panels. ...
... Upper limit value was chosen as 5% for error rate before the study was launched, because this value was supported as acceptable upper limit in several studies about the using of AI techniques (Kahraman et al. 2012a;Aksoy et al. 2012;Kahraman et al. 2012b;Bayindir et al. 2012). Since the calculated values were less than 5%, it may be recommended that both of the proposed methods could be used in applying to the estimation of bonding strength values of plywood panels. ...
En onemli ahsap kokenli levha urunlerinden biri olan kontrplak trafik levhalarindan insaata kadar pek cok kullanim yerine sahiptir. Yuksek kalitede kontrplak uretimi icin tutkal turune bagli olarak optimum pres kosullari altinda iyi bir yapismanin saglanmasi gerektigi bilinen bir gercektir. Bu calismada kontrplagin yapisma direncinin tahmin edilmesi icin modern meta bulussal tekniklerden IKE ve AANN metotlarinin kullanim imkanlari arastirilmistir. Calisma deneysel ve analitik olarak iki kisimdan olusmaktadir. Calismada agac turu olarak saricam, sahil cami ve karacam kullanilmistir. Kaplamalar 2 farkli sicaklikta (32°C ve 50°C) soyulmus ve 3 farkli sicaklikta (110°C, 140°C ve 160°C) kurutulmustur. Kontrplak uretimi icin fenol formaldehit ve melamin ure formaldehit tutkallari olmak uzere iki farkli tutkal turu kullanilmistir. Deneysel olarak kontrplaklarin yapisma direnci degerleri EN 314-1 standardina gore yapilmistir. IKE ve AANN teknikleri analitik olarak yapisma direnci tahmininde kullanilmislardir. En iyi tahmin performansi k degeri 10 icin elde edilmistir. Yapisma direnci uzerine en etkili faktor olarak tutkal turu belirlenmistir. IKE ve AANN icin belirlenen hata oranlari %5’in altinda bulunmustur. Calisma neticesinde uygulanan tekniklerin kontrpalklarda yapisma direnci tahmininde kullanilabilir olduklari tespit edilmistir.
Interturn faults (ITF) at the main machine field winding in brushless synchronous machines (BSM) are not easy to detect as the scarcity of measurements in the rotor. This paper presents a new field winding ITF severity estimation technique based on deep learning algorithms. The theoretical exciter field current is calculated through an artificial neural network, from the machine output values, and then compared to the measured exciter field current. The estimation and the comparison are carried out without need of other than the machine electrical outputs and the exciter field current measurements. To verify the proposed technique, numerous healthy and faulty condition tests were performed on a special laboratory testbench, with more than 7 million measurements in healthy and more than 1 million in different faulty conditions. So as firstly to train the neural network with healthy operation datasets, and secondly to test its capability to estimate the fault severity level. The use of deep neural networks has proven to enhance the accuracy of the exciter field current estimation with respect to previous theoretical model-based methods, enabling to increase the sensitivity and the reliability of the fault detection and fault severity calculation.
