Browsing by Author "Tomova F."
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item APPLICATION OF THE MONTE CARLO METHOD FOR FORECASTING THE DURATION OF PEIRCE-SMITH CONVERTER CAMPAIGNS(2022-01-01) Tomova F.Pierce-Smith converters are used for producing molten ‘blister’ copper. They consist of a steel drum lined with refractory bricks, a support device, a rotary mechanism for the converter, a device for supplying air with tuyeres and a device for removing the converter gases. The characteristics and features of Peirce-Smith converters are analyzed and a database containing measurement data and knowledge-based data is created. Methods and algorithms for pre-processing of raw data have been developed in order to improve the results obtained from techniques that implement predictive maintenance of technological facilities. Various methods for predicting the state of Peirce-Smith converters have been investigated and analyzed. A Monte Carlo Simulation (MCS) algorithm was developed for objects in the Peirce-Smith class of converters, by which in addition to simulating the wall thickness, the duration of the campaign was also determined by probability characteristics.Item COVERING A MAXIMUM NUMBER OF POINTS BY A FIXED NUMBER OF EQUAL DISKS VIA SIMULATED ANNEALING(2024-01-01) Tomova F.; Filipov S.; Avdzhieva A.The presented paper considers the problem of covering a maximum number of n given points in the plane by m equal disks of radius r. A point is covered if it is inside one or more than one disk. The disks need to be placed in the plane in such a way that a maximum number of points are covered. To solve the problem, an objective function, called energy, is introduced in such a way that the greater the covering is, the lower the energy is. Thus, a configuration of disks with minimum energy is a configuration with maximum covering. To find a configuration of disks that minimizes the energy, a stochastic algorithm based on the Monte Carlo simulated annealing technique is proposed. The algorithm overcomes potential local minima, which, as shown in the paper, are quite likely to occur. The computational complexity of the algorithm is O(mn). The algorithm is tested on several cases demonstrating its efficiency in finding global minima of the energy, i.e. configurations with maximum covering.Item MULTIMODAL SYSTEM FOR FACIAL EMOTION RECOGNITION BASED ON DEEP LEARNING(2024-01-01) Atanassov A.; Pilev D.; Tomova F.Emotions are one of the main ways of communication between people and of expressing attitudes towards objects, products, services, etc. They are divided to verbal and non-verbal classes. Human speech and intonation belong to the first class, and to the second (non-verbal) facial and body emotions, known as body language. The subject of this report is the development of multimodal deep learning system intended to recognize facial and body emotions and their relationship with the scene (weather) in which they occur. It is based on three deep learning neural networks (DNN) each one for recognition of facial emotion, body emotion and weather. Combining their results, we improve significantly the final facial emotion recognition (FER) results.