Metaphase finding with deep convolutional neural networks

Design optimization of many electromagnetic and multiphysics problems have multiscale issues that require a fast, efficient, and accurate surrogate‐based model to be used. Recently, in microwave engineering field, artificial intelligence–based models are being used for modeling of complex microwave stages. In all the studies, the main aim is to form models inner structure parameters, by using the given data to predict the linear/nonlinear relationships between given inputs and outputs. Herein, a surrogate‐based model of a nonuniform microstrip transmission line (NTL) with a typical application of design optimization of a band‐pass filter for ISM band application using deep learning (DL) and meta‐heuristic optimization has been presented. In order to have a computationally efficient and accurate optimization process, firstly a 3D EM unit element model of NTL has been designed. The training and test data sets are created based on different sampling methods. A DL regression model modified multilayer perceptron M2LP have been used for prediction of scattering parameters (S) of the NTL, with respect to the variation of geometrical design parameters. The proposed S‐parameters will then be used to calculate the equivalent S‐parameters of the cascading NTL to be used to calculate the NTL‐based microstrip band‐pass filter S‐parameter response. The optimal design parameters of each line used in the filter design have been determined using a fast and powerful optimization algorithm differential evolutionary algorithm.

Section: Case Studymentioning

confidence: 99%

Artificial intelligence–based design optimization of nonuniform microstrip line band pass filter

Mahouti

Yıldırım

Kuskonmaz

2021

“…In particular, the successes of classification problems have been overcome even at human level . DL has a wide range application area such as segmentation, multi‐object tracking, biomedical applications, and even lip reading …”

Section: Modified Multi Layer Perceptronmentioning

confidence: 99%

Deep learning base modified MLP model for precise scattering parameter prediction of capacitive feed antenna

Çalık

Belen

Mahoutı

2019

The relations between the antennas' geometrical parameters and design specifications usually consist of linear and nonlinear components. Especially with the increase of the requested performance measures, the design procedure becomes much more complex due to the conflicting performance criteria or design limitations. To achieve a design with high performance with feasible design parameters, a fast, accurate, and reliable design optimization process is required. Herein, to have a fast, accurate, and high‐performance capacitive‐feed antenna model to be used in design optimization problems, a modified multi‐layer perceptron (M2LP) model has been proposed. The M2LP is an equivalent convolutional neural network (CNN) model of a standard multilayer perceptron (MLP), where instead of traditional training parameters of MLP, more advanced training parameters of CNN models such as batch‐norm layer, leaky‐rectified linear unit (ReLU) layer, and Adam training algorithm had been used. Furthermore, the M2LP model had been used in a design optimization process and the obtained optimal antenna had been prototyped using 3D printing technology for justification of the proposed M2LP model with experimental results. As can be seen from the results, the proposed M2LP model is a fast, accurate, and reliable regression model for design optimization of microwave antennas.

“…With the recent development in computer systems, the application of deep neural networks becomes more common in the modeling of complex and enormous sized problems such as large-scale off-line signature recognition, 73 large-scale sentiment classification, 74 metaphase finding, 75 temporal modeling approaches for large-scale YouTube-8m video, 76,77 segmentation of precursor lesions in cervical cancer, 78 and signature recognition application. 79 Convolutional neural The CNN architecture is similar to MLP models, where both models are trained with a version of the backpropagation algorithm and both of them are consist of input, output, and hidden layers.…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

Application of artificial intelligence algorithms on modeling of reflection phase characteristics of a nonuniform reflectarray element

Mahoutı

2019

Reflectarray antennas (RAs) have the ability to combine the advantages of both traditional parabolic reflector and phased array antennas without the need for feed network designs. Microstrip reflectarrays (MRAs) have the advantages of being small size, light weighted, easy to prototyped, high gain, low side-lobe level, and a predetermined radiation pattern. These can be achieved by precise calculation of reflection phase at each RA unit independently with a phase compensation proportional to the distance from the feed. The challenging problem is to have a fast and high accurate unit element to be used in multidimension, multiobjective design optimization. Herein, artificial intelligence algorithms (AIAs) have been used for prediction of reflection phase characterization of an X band MRA unit element with respect to the geometrical design parameters. Firstly, a nonuniform unit RA has been designed in 3D electromagnetic (EM) simulation tool for creating the training validation data sets. Then, the data sets are given to the different types of AIA regression models such as multilayer perceptron, symbolic regression, and convolutional neural network. From the results of the validation data set, it can be concluded that the proposed models have sufficient accuracy that can be used in a computationally efficient design optimization process of a large-scale RA design.