The gas in-place(GIP)content and the ratio of adsorbed/free gas are two key parameters for the assessment of shale gas resources and have thus received extensive attention.A variety of methods have been proposed to so...The gas in-place(GIP)content and the ratio of adsorbed/free gas are two key parameters for the assessment of shale gas resources and have thus received extensive attention.A variety of methods have been proposed to solve these issues,however none have gained widespread acceptance.Carbon isotope fractionation during the methane transport process provides abundant information,serving as an effective method for differentiating the gas transport processes of adsorbed gas and free gas and ultimately evaluating the two key parameters.In this study,four stages of methane carbon isotope fractionation were documented during a laboratory experiment that simulated gas transport through shale.The four stages reflect different transport processes:the free gas seepage stage(Ⅰ),transition stage(Ⅱ),adsorbed gas desorption stage(Ⅲ)and concentration diffusion stage(Ⅳ).Combined with the results of decoupling experiments,the isotope fractionation characteristics donated by the single effect(seepage,adsorption-desorption and diffusion)were clearly revealed.We further propose a technique integrating the Amoco curve fit(ACF)method and carbon isotope fractionation(CIF)to determine the dynamic change in adsorbed and free gas ratios during gas production.We find that the gases produced in stage Ⅰ are primarily composed of free gas and that carbon isotope ratios of methane(δ13C1)are stable and equal to the ratios of source gas(13C 10).In stage Ⅱ,the contribution of free gas decreases,while the proportion of adsorbed gas increases,and the δ13C1 gradually becomes lighter.With the depletion of free gas,the adsorbed gas contribution in stage Ⅲ reaches 100%,and the δ13C1 becomes heavier.Finally,in stage Ⅳ,the desorbed gas remaining in the pore spaces diffuses out under the concentration difference,and the δ13C1 becomes lighter again and finally stabilizes.In addition,a kinetic model for the quantitative description of isotope fractionation during desorption and diffusion was established.展开更多
Monitoring and predicting marine environmental variables are important for safeguarding livelihoods and the economy.Large language models(LLMs)have shown great potential in time series prediction because of their stro...Monitoring and predicting marine environmental variables are important for safeguarding livelihoods and the economy.Large language models(LLMs)have shown great potential in time series prediction because of their strong computational capabilities,and the application of LLMs to the prediction of marine environmental var-iables is an emerging area of research.However,LLM-based approaches often exhibit oscillations in prediction outputs and large deviations from observed values.To address these issues,we propose TimeLLM-BERT,a hybrid three-stage model based on feature extraction,autoregressive prediction,and error correction.The model in-corporates a structured prompt module,a trend enhancement algorithm,and a residual-fitting optimization strategy,which significantly enhance prediction accuracy.To systematically evaluate the performance of the model,comparative experiments were conducted against LSTM,BiTCN,NBEATSx,iTransformer,NHITS,and Time-LLM models using four key variables:significant wave height(SWH),sea surface temperature(SST),temperature at 2 m above the sea surface(T2M),and wind field(WF).The results show that the performance of the model is significantly better than existing models,and the mean absolute error for SWH prediction is reduced by 24.7%.It also achieves stable performance in SST prediction and strong consistency in WF prediction compared with the existing models.Robustness and universality tests show that the error evaluation indicators exhibit low variation,demonstrating strong stability and generalization ability.In summary,TimeLLM-BERT offers significant improvements in accuracy and stability for predicting marine environmental variables,providing a new framework for modeling complex time series data.展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.41672130,41602131)the Research Project Funded by the SINOPEC Corp.(Grant No.P17027-3)the National Science and Technology Major Project(Grant No.2016ZX05061).
文摘The gas in-place(GIP)content and the ratio of adsorbed/free gas are two key parameters for the assessment of shale gas resources and have thus received extensive attention.A variety of methods have been proposed to solve these issues,however none have gained widespread acceptance.Carbon isotope fractionation during the methane transport process provides abundant information,serving as an effective method for differentiating the gas transport processes of adsorbed gas and free gas and ultimately evaluating the two key parameters.In this study,four stages of methane carbon isotope fractionation were documented during a laboratory experiment that simulated gas transport through shale.The four stages reflect different transport processes:the free gas seepage stage(Ⅰ),transition stage(Ⅱ),adsorbed gas desorption stage(Ⅲ)and concentration diffusion stage(Ⅳ).Combined with the results of decoupling experiments,the isotope fractionation characteristics donated by the single effect(seepage,adsorption-desorption and diffusion)were clearly revealed.We further propose a technique integrating the Amoco curve fit(ACF)method and carbon isotope fractionation(CIF)to determine the dynamic change in adsorbed and free gas ratios during gas production.We find that the gases produced in stage Ⅰ are primarily composed of free gas and that carbon isotope ratios of methane(δ13C1)are stable and equal to the ratios of source gas(13C 10).In stage Ⅱ,the contribution of free gas decreases,while the proportion of adsorbed gas increases,and the δ13C1 gradually becomes lighter.With the depletion of free gas,the adsorbed gas contribution in stage Ⅲ reaches 100%,and the δ13C1 becomes heavier.Finally,in stage Ⅳ,the desorbed gas remaining in the pore spaces diffuses out under the concentration difference,and the δ13C1 becomes lighter again and finally stabilizes.In addition,a kinetic model for the quantitative description of isotope fractionation during desorption and diffusion was established.
基金funded by the National Natural Science Foundation of China(grant numbers,62071279 and 41930535),the Artificial Intelli-gence Ocean Big Data Innovation Service Platform Fund.
文摘Monitoring and predicting marine environmental variables are important for safeguarding livelihoods and the economy.Large language models(LLMs)have shown great potential in time series prediction because of their strong computational capabilities,and the application of LLMs to the prediction of marine environmental var-iables is an emerging area of research.However,LLM-based approaches often exhibit oscillations in prediction outputs and large deviations from observed values.To address these issues,we propose TimeLLM-BERT,a hybrid three-stage model based on feature extraction,autoregressive prediction,and error correction.The model in-corporates a structured prompt module,a trend enhancement algorithm,and a residual-fitting optimization strategy,which significantly enhance prediction accuracy.To systematically evaluate the performance of the model,comparative experiments were conducted against LSTM,BiTCN,NBEATSx,iTransformer,NHITS,and Time-LLM models using four key variables:significant wave height(SWH),sea surface temperature(SST),temperature at 2 m above the sea surface(T2M),and wind field(WF).The results show that the performance of the model is significantly better than existing models,and the mean absolute error for SWH prediction is reduced by 24.7%.It also achieves stable performance in SST prediction and strong consistency in WF prediction compared with the existing models.Robustness and universality tests show that the error evaluation indicators exhibit low variation,demonstrating strong stability and generalization ability.In summary,TimeLLM-BERT offers significant improvements in accuracy and stability for predicting marine environmental variables,providing a new framework for modeling complex time series data.
