Search

The paper presents a study on the application of a system for pattern design of women's dresses from knitted fabrics containing elastomeric threads from 1% to 6%. The system uses a different approach to account the stretch of the knitted fabric when designing clothing. In this approach are used for initial information for determining the values of the reduction factors only the data on the composition of the knitted fabrics according to the current standards. The reduction factors are calculated depending on the percentage content of elastomeric threads in the knitted fabric. Three models of women's dresses were designed. Knitted fabrics containing elastomeric threads from 1% to 6% have been chosen for their sewing. The compatibility has been studied between the dresses' details when they are located in different directions during cutting, when they are with different loop structure or with different stretch. The shaping of the models by three-dimensional elements, such as by the fixed draperies, has been analysed. Approaches are proposed for resizing the dresses' details (located in different directions of the stretch of the knitted fabric), for the correct selection of compatible knitted fabrics for dress details and for the determination of the size of the knot and the width of the draped sections. The fit of the made dresses corresponds to the designed silhouettes and no creases or displaced seams are monitored. The accuracy of the mathematical models, which are determining the value of the reduction factors and the applicability of the system for pattern designing of the knitted dresses has been proven. [ABSTRACT FROM AUTHOR]

In this paper an Information and Communication Technology (ICT) prototype of Big Data Analysis System based on Data Mining Techniques, Hadoop information infrastructure platform for distributed data collection and MATLAB analytical environment have were proposed. The obtained results of test activities related to some system modules for clustering analysis, recognition and classification tasks about defined economic objectives and processes are presented. Achieved indicators for revenues, expenditures and procedures for excessive deficit for individual Economic regions in Europe, covering the period from 1995 to 2021, according to a report by the European Commission were defined as a complex object of study. Approaches based on k-Means Clustering, Linear and Quadratic Discriminant Analysis, Resubstitution and Cross-validation Techniques, Feed-Forward Neural Networks (FFNN) with Scaled Conjugate Gradient training algorithm were applied. Models with observed high quality indicators were synthesized in the process of the exposed procedures. [ABSTRACT FROM AUTHOR]

In this paper a hybrid approach for spectral analysis and voice profiles recognition by techniques on the base of Machine Learning and Artificial Intelligence (AI) have been proposed. For voice-processing procedures was applied the algorithm of Fast Fourier Transformation (FFT) with several different window types, respectively Hamming, 4 Term BHarris, Flat Top and Hanning. The spectral feature extraction is used for pre-processing training sets about k-Nearest Neighbours (k-NN) classification models and Feed-Forward Neural Networks (FFNN) for individual's personal identity about target group of people. Euclidean, Cityblock, Minkowski and Chebychev metric distances were applied in k-NN model creation. The design models are evaluated through resubstitution and cross-validation techniques. Levenberg- Marquardt learning algorithm was used to FFNN architectures with Linear and Tangent Sigmoid activation functionsin network outputs. High quality k-NN and FFNN models in regard to personal voice identification with level of accuracy achieved 97.68 % and 100.0 % were synthesized. [ABSTRACT FROM AUTHOR]

The paper presents quantitative and qualitative diagnosing approach for analysis of the sensory elements used to force measurements force at metal deformation and predicting of the degree of influence of applied force "F" using Feed- Forward Neural Networks (FFNN). This approach consists in the design and selection of neural models by backpropagation algorithms, respectively Scaled Conjugate Gradient (SCG) and Levenberg-Marquardt with functional procedures based on set of evaluation criteria, including Accuracy, Cross-Entropy (CE) and Mean Squared Error (MSE). The registered output electrical voltages from bridge circuits of switching on one and two strain gauge transducers from performed laboratory experiments were applied to investigate of FNN target models. According to SCG algorithm, neural architecture for quantitative sensory identification with 100 % achieved accuracy was synthesized. Using LM training rules, an architecture with a minimum MSE = 0.0029 for predictive analysis of the force effects on the test specimens was found. [ABSTRACT FROM AUTHOR]

Future climate scenarios of the Global Circulation model (GCM) show an increased frequency of heavy rainfall events, which may lead to more severe floods. It is also expected that more and more areas will suffer due to flooding as a result of growing urbanization. Public attention has increased in many parts of the world in recent years and calls have been made to improve flood warnings, including the United States, the European Union and Australia (Hapuarachchi, H.A.P, and Q.J. Wang 2008). To respond and manage flood hazard there is a need to provide a high spatial resolution flood forecast and with sufficient lead time. This study presents an approach for creation of a forecast model based on the analysis of historical hydrometeorological data from conventional and automatic monitoring networks of the National Institute of Meteorology and Hydrology in Bulgaria. The study area is the downstream Kamchia river watershed. Real-time water level observations and calculated discharges based on temporary rating curves are used to dynamically adjust the runoff forecasting. In this paper an approach for combining a hydrological model (TOPKAPI) and a twodimensional hydraulic model (HEC-RAS) for flood simulation is presented. Hydrological modelling is used for forecasting the outflow at a hydrometric station (43800) on Kamchia River near the village of Grozdyovo. The 2D hydraulic HEC-RAS model is used for simulating rainfall - runoff process in Kamchia watershed downstream of the village of Grozdyovo and the results from the hydrological modeling are used as an input data. In this paper the results of using operational hydrological data and forecast precipitation totals for flood simulation is presented. [ABSTRACT FROM AUTHOR]