SCIENTIFIC-METHODICAL APPROACH TO ASSESSING THREATS TO UKRAINE’S ENERGY SECURITY USING NEURAL NETWORKS

The article substantiates the importance of ensuring Ukraine's energy security, the state of which is determined by the conditions of the internal environment of the functioning and development of the energy system, as well as the influence of external environmental factors. The spectrum of threats to Ukraine's energy security is identified in the following areas: threats caused by war; threats of a financial, economic, technical and technological nature; cyber threats and information attacks; environmental requirements and climate challenges; threats associated with corruption and ineffective management. A scientific and methodical approach to assessing the level of threats to energy security based on probabilistic neural networks is proposed. The input parameters for assessing the degree of threats to the country's energy security are proposed to be formed on the basis of an expert survey. To check the consistency of expert judgments, it is proposed to carry out clustering of expert assessments. The authors substantiate that a clear identification of threats in terms of their types, taking into account their content, qualitative and quantitative context of threats, and assessment of the probable degree of threats using the proposed scientific and methodical approach based on neural networks will serve as the basis for making informed management decisions to prevent / minimize / eliminate the negative impact of factors on the energy security of Ukraine. Keywords: energy security, energy sector, threats, risks, assessment, management, economic and mathematical modeling, neural networks.