Authors:Varun Barthwal, M.M.S. Rauthan, and Rohan Varma
prediction of resource utilization is not only based upon current usage patterns but also past system behavior may be considered. In this review, it was found that machinelearning (ML) models are suitable to predict resource utilization using historical data
Authors:Ádám Pintér, Balázs Schmuck, and Sándor Szénási
The aim of this paper is to present a technique, which uses machine learning to process the short text answers with Hungarian language. The processing is based on a special neural network, the convolutional neural network, which can efficiently process short text answer. To achieve precise classification for training and recall grammatically consistent answers and the conversion of the text to the input are inevitable. To convert the input, continuous bag of words and Skip-Gram models will be used, resulting in a model that will be able to evaluate the Hungarian short text answers.
During the production of gas one of the major problems is the formation of hydrate crystals in the pipeline. The growing hydrate crystals can form hydrate plugs in the pipeline. The hydrate plug effect lengthens production outages and results in the loss of money of the maintainer, because the removal of the plug is a time consuming procedure. One of the solutions used to prevent hydrate formation is the addition of modern compositions to the gas flow. The modern compositions help to dehydrate the gas, thus, the size of hydrate crystal does not increase. The substances, used in low concentrations, have to be locally injected at the gas well sites. Thus, an injector unit is required for this purpose. The production-related aspect that the consumers expect much more flexibility from gas provider cannot be neglected because of the habits of the users and the appearance of energy-saving technologies are different. The first part of the article a newly developed injection system is introduced. To achieve optimal dosage, not only the hardware of injection system is important, but also the software. In addition to the traditional control, a preventive inhibitor dosing system can be developed, based on model driven system. The nature of the model highly influences the quality of control system. In the second part of the article a machine learning based predictive detection system is introduced
Authors:Bence Gergő Barsy, Gyula Győri, and Péter Tamás Szemes
offers a number of enhancements that can be used to extract even more data from the measurement data set: Application of Deep/machinelearning algorithms to classify multiple measurements Creating a correlation matrix for the measurement parameters