The analysis of the time series of quarterly volumes of flaring of associated petroleum gas (APG) in the Khanty-Mansi Autonomous Area (KhMAA) for the period of 2012-2016 was carried out. It is established that the trend of reducing the volume of combustible gas for a specified period is described by an exponential equation. It is shown that, along with the exponential trend, cyclical seasonal variations in the volume of the combusted gas are observed. A model is proposed for predicting the volume of gas being combusted as a sum of two terms, the first of which is represented by the equation of the exponential trend, and the second term is intended to take into account seasonal fluctuations in gas flaring volumes. An example of a forecast of APG flaring volumes for the period 2017-2020 is given for the territory of the KhMAA.
Materials and methods
The data of reports of the department of subsoil use and natural resources of the KhMAA, data on oil production in the KhMAA. To perform the study, a method of time series analysis was used, which made it possible to develop a model for predicting the dynamics of volumes of gas being combusted in an oil-producing territory.
Based on the analysis of the time series of quarterly volumes of flaring of associated gas in the KhMAA over the 5-year period, an exponential trend of reducing the volume of flared gas and cyclical seasonal fluctuations of the volume of flared gas that are superimposed on the line of the exponential trend have been established. Based on the detected patterns of gas combustion dynamics, a model for predicting the volume of combusted gas is proposed as the sum of two terms, the first of which is represented by the equation of the exponential trend, and the second term takes into account seasonal fluctuations in the volume of gas flaring. The parameters of the model were determined according to the data on the volumes of combusted gas in the period from 2012 to 2015. To check the adequacy of the forecast model, a comparison was made with experimental (official) data on the volumes of associated gas burned in the KhMAA. Using a control sample of data on the total volumes of gas burned in 4 seasons of 2016, it was established that the forecast error does not exceed 6%. This can serve as confirmation of the practical acceptability of the proposed method for predicting the volume of gas flaring within the oil production areas. To illustrate the proposed forecasting method, an example is given of calculating forecast estimates of gas flaring volumes for the next five year period.
The developed model can be used for rapid assessment of the forecast of the total volume of gas flaring at fields in other oil-producing regions with similar geological, geographical and climatic conditions.