Research On Early Warning And Response Strategies Of Eco-environment For Shale Gas Development

As the most promising unconventional oil and gas resource at present,shale gas has important strategic significance for alleviating the contradiction between Chinese natural gas supply and demand,ensuring national energy security,and optimizing the energy structure.However,due to the particularity,complexity and initial stage of shale gas development,coupled with the diversified manifestations of environmental damage caused by shale gas resource development,how to issue the ecological environment for shale gas development in real time Early warning signals of the situation so that relevant personnel can take early intervention measures to slow down or prevent the deterioration of the ecological environment is a hot and difficult point that the industry and academia pay attention to.Based on the analysis of the complexity of ecological early-warning for shale gas development,this thesis constructs an "early-warning-oriented" indicator system for the ecological environment of shale gas development based on the DPSIR framework,and combines AHP-entropy weight method with optimization theory for indicator selection,To ensure the integrity of the system while reducing the number of indicators and improving the practicability of early warning.In view of the problem that the use of a single prediction model will cause the omission of key information and bias the early warning results,this thesis proposes to use the STL-NAR neural network to predict the police situation.Furthermore,the early warning interval is designed from the three aspects of block characteristics,seasonal characteristics,and evolutionary trends to enhance the dynamics of the warning situation.Finally,this thesis starts from the actual situation of the C shale gas field,verifies the validity and practicability of the model,and proposes a response strategy for the C shale gas field based on the warning results.Research indicates:(1)Using the DPSIR early warning indicator system construction and optimization method can optimize the number of early warning indicators for the C shale gas field from 28 to 18,using 64.29% of the number of indicators to express the original 81.75%of the information,while ensuring the integrity of the DPSIR framework.(2)The STL data decomposition method can well retain the trend and seasonal characteristics of the driving force,pressure,and status layer;however,for the impact and response layer indicators,the seasonal characteristics are meaningless and the data fluctuates greatly.After the decomposition,the prediction will increase the error.(3)From 2020 to 2021,the warning level of C shale gas field has been stable within the middle warning range.However,in the winter of 2020 and 2021,the increase in warning value will become steep,and we should focus on preventing it from evolving to the heavy warning line.(4)Under the premise that the early warning index value of the impact layer drops by 20%,the C shale gas field development company wants to control the early warning value in November 2020 within the second degree range of the middle warning,and only needs to reduce the development intensity by 4%;to maintain In a stable state,the development intensity needs to be reduced by 18%;to achieve the ideal state where the comprehensive warning value is reduced by 10%,the development intensity needs to be reduced by 32%,but this result is difficult to achieve.