Characterization and optimization of physical and chemical properties of drilling fluids are critical for the efficiency and success of drilling operations.In particular,maintaining the optimal levels of solids conten...Characterization and optimization of physical and chemical properties of drilling fluids are critical for the efficiency and success of drilling operations.In particular,maintaining the optimal levels of solids content is essential for achieving the most effective fluid performance.Proper management of solids content also reduces the risk of tool failures.Traditional solids content analysis methods,such as retort analysis,require substantial human intervention and time,which can lead to inaccuracies,time-management issues,and increased operational risks.In contrast to human-intensive methods,machine learning may offer a viable alternative for solids content estimation due to its pattern-recognition capability.In this study,a large set of laboratory reports of drilling-fluid analyses from 130 oil wells around the world were compiled to construct a comprehensive data set.The relationships among various rheological parameters were analyzed using statistical methods and machine learning algorithms.Several machine learning algorithms of diverse classes,namely linear(linear regression,ridge regression,and ElasticNet regression),kernel-based(support vector machine)and ensemble tree-based(gradient boosting,XGBoost,and random forests)algorithms,were trained and tuned to estimate solids content from other readily available drilling fluid properties.Input variables were kept consistent across all models for interpretation and comparison purposes.In the final stage,different evaluation metrics were employed to evaluate and compare the performance of different classes of machine learning models.Among all algorithms tested,random forests algorithm was found to be the best predictive model resulting in consistently high accuracy.Further optimization of the random forests model resulted in a mean absolute percentage error(MAPE)of 3.9%and 9.6%and R^(2) of 0.99 and 0.93 for the training and testing sets,respectively.Analysis of residuals,their histograms and Q-Q normality plots showed Gaussian distributions with residuals that are scattered around a mean of zero within error ranges of±1%and±4%,for training and testing,respectively.The selected model was further validated by applying the rheological measurements from mud samples taken from an offshore well from the Gulf of Mexico.The model was able to estimate total solids content in those four mud samples with an average absolute error of 1.08% of total solids content.The model was then used to develop a web-based graphical-user-interface(GUI)application,which can be practically used at the rig site by engineers to optimize drilling fluid programs.The proposed model can complement automation workflows that are designed to measure fundamental rheological properties in real time during drilling operations.While a standard retort test can take approximately 2 h at the rig site,such kind of real-time estimations can help the rig personnel to timely optimize drilling fluids,with a potential of saving 2920 man-hours in a given year for a single drilling rig.展开更多
High-pressure homogenization （HPH） technology was applied as a pretreatment to disintegrate sewage sludge. The effects of homogenization pressure, homogenization cycle number, and total solid content on sludge disin...High-pressure homogenization （HPH） technology was applied as a pretreatment to disintegrate sewage sludge. The effects of homogenization pressure, homogenization cycle number, and total solid content on sludge disintegration were investigated. The sludge disintegration degree （DDCOD）, protein concentration, and polysaccharide concentration increased with the increase of homogenization pressure and homogenization cycle number, and decreased with the increase of sludge total solid （TS） content. The maximum DDCOD of 43.94% was achieved at 80 MPa with four homogenization cycles for a 9.58 g/L TS sludge sample. A HPH sludge disintegration model of DDCOD= kNaPb was established by multivariable linear regression to quantify the effects of homogenization parameters. The homogenization cycle exponent a and homogenization pressure exponent b were 0.4763 and 0.7324 respectively, showing that the effect of homogenization pressure （P） was more significant than that of homogenization cycle number （N）. The value of the rate constant k decreased with the increase of sludge total solid content. The specific energy consumption increased with the increment of sludge disintegration efficiency. Lower specific energy consumption was required for higher total solid content sludge.展开更多
Total milk solid(TMS)content directly reflects the quality of milk.Rumen bacteria ferment dietary components,the process of which generates the precursors for the synthesis of milk solid,therefore,the variation in rum...Total milk solid(TMS)content directly reflects the quality of milk.Rumen bacteria ferment dietary components,the process of which generates the precursors for the synthesis of milk solid,therefore,the variation in rumen bacterial community could be associated with milk solid in dairy cows.In this study,45 healthy mid-lactation Holstein dairy cows with the similar body weight,lactation stage,and milk yield were initially used for the selection of 10 cows with high TMS(HS)and 10 cows with low TMS(LS).All those animals were under the same feeding management,and the individual milk yield was recorded for 14 consecutive days before milk and rumen fluid were sampled.Rumen fluid was used to determine bacterial community by 16S rRNA gene sequencing technique.The HS cows had significantly greater feed intake and milk TMS,fat,protein content than LS cows(P<0.05).Among the volatile fatty acids(VFA),propionic acid and valeric acid concentrations were significantly greater in HS cows than those in LS cows(P<0.05).There was no significant difference in the concentrations of acetate,butyrate,isobutyrate,valerate,and the total VFA(P>0.05),nor was the acetate-to-propionate ratio,pH value,ammonia nitrogen and microbial crude protein concentrations(P>0.05).Significant differences in the relative abundances of some bacterial genera were found between HS and LS cows.Spearman’s rank correlation analysis revealed that TMS content was correlated positively with the abundances of Ruminococcaceae UCG-014,Ruminococcaceae NK4A214 group,Prevotellaceae UCG-001,Butyrivibrio 2,Prevotellaceae UCG-003,Candidatus Saccharimonas,Ruminococcus 2,Lachnospiraceae XPB1014 group,probable genus 10,Eubacterium ventriosum group,but negatively correlated with Pyramidobacte.In addition,Ruminococcaceae UCG-014,Ruminococcus 2,Ruminococcaceae UCG001,probable genus 10 and Eubacterium ventriosum group might boost the total VFA production in the rumen.In conclusion,the dry matter intake of dairy cows and some special bacteria in rumen were significantly associated with TMS content,which suggests the potential function of rumen bacteria contributing to TMS content in dairy cows.展开更多
文摘Characterization and optimization of physical and chemical properties of drilling fluids are critical for the efficiency and success of drilling operations.In particular,maintaining the optimal levels of solids content is essential for achieving the most effective fluid performance.Proper management of solids content also reduces the risk of tool failures.Traditional solids content analysis methods,such as retort analysis,require substantial human intervention and time,which can lead to inaccuracies,time-management issues,and increased operational risks.In contrast to human-intensive methods,machine learning may offer a viable alternative for solids content estimation due to its pattern-recognition capability.In this study,a large set of laboratory reports of drilling-fluid analyses from 130 oil wells around the world were compiled to construct a comprehensive data set.The relationships among various rheological parameters were analyzed using statistical methods and machine learning algorithms.Several machine learning algorithms of diverse classes,namely linear(linear regression,ridge regression,and ElasticNet regression),kernel-based(support vector machine)and ensemble tree-based(gradient boosting,XGBoost,and random forests)algorithms,were trained and tuned to estimate solids content from other readily available drilling fluid properties.Input variables were kept consistent across all models for interpretation and comparison purposes.In the final stage,different evaluation metrics were employed to evaluate and compare the performance of different classes of machine learning models.Among all algorithms tested,random forests algorithm was found to be the best predictive model resulting in consistently high accuracy.Further optimization of the random forests model resulted in a mean absolute percentage error(MAPE)of 3.9%and 9.6%and R^(2) of 0.99 and 0.93 for the training and testing sets,respectively.Analysis of residuals,their histograms and Q-Q normality plots showed Gaussian distributions with residuals that are scattered around a mean of zero within error ranges of±1%and±4%,for training and testing,respectively.The selected model was further validated by applying the rheological measurements from mud samples taken from an offshore well from the Gulf of Mexico.The model was able to estimate total solids content in those four mud samples with an average absolute error of 1.08% of total solids content.The model was then used to develop a web-based graphical-user-interface(GUI)application,which can be practically used at the rig site by engineers to optimize drilling fluid programs.The proposed model can complement automation workflows that are designed to measure fundamental rheological properties in real time during drilling operations.While a standard retort test can take approximately 2 h at the rig site,such kind of real-time estimations can help the rig personnel to timely optimize drilling fluids,with a potential of saving 2920 man-hours in a given year for a single drilling rig.
基金supported by the China-Israel Joint Research Program, MOST of Chinathe National Natural Science Foundation of China (No. 51178047)the Foundation of Key Laboratory for Solid Waste Management and Environment Safety,Ministry of Education of China (No. SWMES 2010-2)
文摘High-pressure homogenization （HPH） technology was applied as a pretreatment to disintegrate sewage sludge. The effects of homogenization pressure, homogenization cycle number, and total solid content on sludge disintegration were investigated. The sludge disintegration degree （DDCOD）, protein concentration, and polysaccharide concentration increased with the increase of homogenization pressure and homogenization cycle number, and decreased with the increase of sludge total solid （TS） content. The maximum DDCOD of 43.94% was achieved at 80 MPa with four homogenization cycles for a 9.58 g/L TS sludge sample. A HPH sludge disintegration model of DDCOD= kNaPb was established by multivariable linear regression to quantify the effects of homogenization parameters. The homogenization cycle exponent a and homogenization pressure exponent b were 0.4763 and 0.7324 respectively, showing that the effect of homogenization pressure （P） was more significant than that of homogenization cycle number （N）. The value of the rate constant k decreased with the increase of sludge total solid content. The specific energy consumption increased with the increment of sludge disintegration efficiency. Lower specific energy consumption was required for higher total solid content sludge.
基金the Scientific Research Project for Major Achievements of the Agricultural Science and Technology Innovation Program(CAAS ZDXT2019004)the Agricultural Science and Technology Innovation Program(ASTIP-IAS12)Modern Agro-Industry Technology Research System of the PR China(CARS-36).
文摘Total milk solid(TMS)content directly reflects the quality of milk.Rumen bacteria ferment dietary components,the process of which generates the precursors for the synthesis of milk solid,therefore,the variation in rumen bacterial community could be associated with milk solid in dairy cows.In this study,45 healthy mid-lactation Holstein dairy cows with the similar body weight,lactation stage,and milk yield were initially used for the selection of 10 cows with high TMS(HS)and 10 cows with low TMS(LS).All those animals were under the same feeding management,and the individual milk yield was recorded for 14 consecutive days before milk and rumen fluid were sampled.Rumen fluid was used to determine bacterial community by 16S rRNA gene sequencing technique.The HS cows had significantly greater feed intake and milk TMS,fat,protein content than LS cows(P<0.05).Among the volatile fatty acids(VFA),propionic acid and valeric acid concentrations were significantly greater in HS cows than those in LS cows(P<0.05).There was no significant difference in the concentrations of acetate,butyrate,isobutyrate,valerate,and the total VFA(P>0.05),nor was the acetate-to-propionate ratio,pH value,ammonia nitrogen and microbial crude protein concentrations(P>0.05).Significant differences in the relative abundances of some bacterial genera were found between HS and LS cows.Spearman’s rank correlation analysis revealed that TMS content was correlated positively with the abundances of Ruminococcaceae UCG-014,Ruminococcaceae NK4A214 group,Prevotellaceae UCG-001,Butyrivibrio 2,Prevotellaceae UCG-003,Candidatus Saccharimonas,Ruminococcus 2,Lachnospiraceae XPB1014 group,probable genus 10,Eubacterium ventriosum group,but negatively correlated with Pyramidobacte.In addition,Ruminococcaceae UCG-014,Ruminococcus 2,Ruminococcaceae UCG001,probable genus 10 and Eubacterium ventriosum group might boost the total VFA production in the rumen.In conclusion,the dry matter intake of dairy cows and some special bacteria in rumen were significantly associated with TMS content,which suggests the potential function of rumen bacteria contributing to TMS content in dairy cows.
