The anodic over voltage of the carbon anode in aluminumelectrolysis is of the order of 0.6 V at normal current densities.However, it can be reduced somewhat by doping the anode carbon withvarious inorganic compounds. ...The anodic over voltage of the carbon anode in aluminumelectrolysis is of the order of 0.6 V at normal current densities.However, it can be reduced somewhat by doping the anode carbon withvarious inorganic compounds. A new apparatus was designed to improvethe precision of over voltage measurements. Anodes were doped withMgAl_2O_4 and AlF_3 both by impregnation of the coke and by addingpowder, and the measured over voltage was compared with that ofUndoped samples. For prebake type anodes baked at around 1150 deg. C,the anodic overvoltage was reduced by 40-60 mV, And for Soderbergtype anodes, baked at 950 deg. C, by 60-80 mV.展开更多
The reaction kinetics of the oxidative dehydrogenation of propane was studied at 475-550°C over a VMgO catalyst. Vanadium-magnesium-oxides are among the most selective and active catalysts for the dehydrogenation...The reaction kinetics of the oxidative dehydrogenation of propane was studied at 475-550°C over a VMgO catalyst. Vanadium-magnesium-oxides are among the most selective and active catalysts for the dehydrogenation of propane to propylene. Selectivity to propylene up to about 60% was obtained at 10% conversion, but the selectivity decreased with increasing conversion. No oxygenates were detected, the only by-products were CO and CO2. The reaction rate of propane was found to be first order in propane and close to zero order in oxygen, which is in agreement with a Mars van Krevelen mechanism with the activation of the hydrocarbon as the rate determining step. The activation energy of the conversion of propane was found to be 122±6 kJ/mol.展开更多
In this study, the undrained behaviour of silt under low stress level is studied. An effective preparation method for built-in silt samples in the triaxial test was firstly developed. By triaxial testing of samples at...In this study, the undrained behaviour of silt under low stress level is studied. An effective preparation method for built-in silt samples in the triaxial test was firstly developed. By triaxial testing of samples at low confining pressures it was found that silt easily loses stability and liquefies. Loose silt may show temporary liquefaction under static loading, and develop full liquefaction under cyclic loading. The most important factors influencing the silt behaviour are porosity, confining pressure, consolidation state, cyclic loading level and number of cycles. The maximum obtainable shear stress is primarily a function of the confining pressure and the internal frictional angle. The actual structure of the silt material is the key factor in controlling its behaviour.展开更多
Undrained behaviour of sand under low cell pressure was studied in static and cyclic triaxial tests. It was found that very loose sand liquefies under static loading with the relative density being a key parameter for...Undrained behaviour of sand under low cell pressure was studied in static and cyclic triaxial tests. It was found that very loose sand liquefies under static loading with the relative density being a key parameter for the undrained behaviour of sand. In cyclic triaxial tests, pore water pressures built up during the cyclic loading and exceeded the confining cell pressure. This process was accompanied by a large sudden increase in axial deformation. The necessary number of cycles to obtain liquefaction was related to the confining cell pressure, the amplitude of cyclic loading and the relative density of sand.In addition, the patterns of pore water pressure response are different from those of sand samples with different relative densities. The test results are very useful for expounding scour mechanism around coastal structures since they relate to the low stress behaviour of the sand.展开更多
A Bayesian approach using Markov chain Monte Carlo algorithms has been developed to analyze Smith’s discretized version of the discovery process model. It avoids the problems involved in the maximum likelihood method...A Bayesian approach using Markov chain Monte Carlo algorithms has been developed to analyze Smith’s discretized version of the discovery process model. It avoids the problems involved in the maximum likelihood method by effectively making use of the information from the prior distribution and that from the discovery sequence according to posterior probabilities. All statistical inferences about the parameters of the model and total resources can be quantified by drawing samples directly from the joint posterior distribution. In addition, statistical errors of the samples can be easily assessed and the convergence properties can be monitored during the sampling. Because the information contained in a discovery sequence is not enough to estimate all parameters, especially the number of fields, geologically justified prior information is crucial to the estimation. The Bayesian approach allows the analyst to specify his subjective estimates of the required parameters and his degree of uncertainty about the estimates in a clearly identified fashion throughout the analysis. As an example, this approach is applied to the same data of the North Sea on which Smith demonstrated his maximum likelihood method. For this case, the Bayesian approach has really improved the overly pessimistic results and downward bias of the maximum likelihood procedure.展开更多
As the oil and gas industries continue to increase their activity in deep water, gas hydrate hazards will become more serious and challenging, both at present and in the future. Accurate predictions of the hydrate-fre...As the oil and gas industries continue to increase their activity in deep water, gas hydrate hazards will become more serious and challenging, both at present and in the future. Accurate predictions of the hydrate-free zone and the suitable addition of salts and/or alcohols in preparing drilling fluids are particularly important both in preventing hydrate problems and decreasing the cost of drilling operations. In this paper, we compared several empirical correlations commonly used to estimate the hydrate inhibition effect of aqueous organic and electrolyte solutions using experiments with ethylene glycol （EG） as a hydrate inhibitor. The results show that the Najibi et al. correlation （for single and mixed thermodynamic inhibitors） and the Ostergaard et al. empirical correlation （for single thermodynamic inhibitors） are suitable for estimating the hydrate safety margin of oil-based drilling fluids （OBDFs） in the presence of thermodynamic hydrate inhibitors. According to the two correlations, the OBDF, composed of 1.6 L vaporizing oil, 2% emulsifying agent, 1% organobentonite, 0.5% SP-1, 1% LP-1, 10% water and 40% EG, can be safely used at a water depth of up to 1900 m. However, for more accurate predictions for drilling fluids, the effects of the solid phase, especially bentonite, on hydrate inhibition need to be considered and included in the application of these two empirical correlations.展开更多
Artificial neural networks (ANNs) have been widely used to solve a number of problems to which analytical solutions are difficult to obtain using traditional mathematical approaches.Such problems exist also in the ana...Artificial neural networks (ANNs) have been widely used to solve a number of problems to which analytical solutions are difficult to obtain using traditional mathematical approaches.Such problems exist also in the analysis of industrial robots. This paper presents an overview of ANN applications to robot kinematics, dynamics,control, trajectory and path planning, and sensing. Reasons for using or not using ANNs to industrial robots are explained as well.展开更多
Technology development and implementation on the manufacturing arena has, throughout the years, frustrated employees over the requirement of constantly adjusting to new technologies. The employee's ability to lear...Technology development and implementation on the manufacturing arena has, throughout the years, frustrated employees over the requirement of constantly adjusting to new technologies. The employee's ability to learn is bounded, and thus limited. Looking at a modern production system, it is often a complicated array of technologies that have to be coordinated; or the technologies are coordinating themselves through artificial intelligence. This paper discusses the need for a holistic knowledge management approach to manufacturing,combining in-place technology with process management philosophy, in order for the firm's human resources to create, share, store and apply new knowledge.展开更多
Traditionally, the objective of a manufacturing facility layout problem is to minimize the material handling cost of the manufacturing systems. Because of the combination of the facility layout problems, the genetic a...Traditionally, the objective of a manufacturing facility layout problem is to minimize the material handling cost of the manufacturing systems. Because of the combination of the facility layout problems, the genetic algorithms (GA) technique is the most promising approach for solving practical layout problems. Much of the previous work has been done for identical problems, where all departments are equal in area. In this paper, non-identical problems are dealt with. A new coding approach - the Comer Attachmer Structure (CAS) is introduced.展开更多
基金the Norwegian Aluminium Industry, the Norwegian Research Council and the China National Key Fundamental Research Development Project (No.G1999064903)
文摘The anodic over voltage of the carbon anode in aluminumelectrolysis is of the order of 0.6 V at normal current densities.However, it can be reduced somewhat by doping the anode carbon withvarious inorganic compounds. A new apparatus was designed to improvethe precision of over voltage measurements. Anodes were doped withMgAl_2O_4 and AlF_3 both by impregnation of the coke and by addingpowder, and the measured over voltage was compared with that ofUndoped samples. For prebake type anodes baked at around 1150 deg. C,the anodic overvoltage was reduced by 40-60 mV, And for Soderbergtype anodes, baked at 950 deg. C, by 60-80 mV.
文摘The reaction kinetics of the oxidative dehydrogenation of propane was studied at 475-550°C over a VMgO catalyst. Vanadium-magnesium-oxides are among the most selective and active catalysts for the dehydrogenation of propane to propylene. Selectivity to propylene up to about 60% was obtained at 10% conversion, but the selectivity decreased with increasing conversion. No oxygenates were detected, the only by-products were CO and CO2. The reaction rate of propane was found to be first order in propane and close to zero order in oxygen, which is in agreement with a Mars van Krevelen mechanism with the activation of the hydrocarbon as the rate determining step. The activation energy of the conversion of propane was found to be 122±6 kJ/mol.
基金This research is partly supported by the National Natural Science Foundation(No.50009002)863(No.2001AA616020)projects.
文摘In this study, the undrained behaviour of silt under low stress level is studied. An effective preparation method for built-in silt samples in the triaxial test was firstly developed. By triaxial testing of samples at low confining pressures it was found that silt easily loses stability and liquefies. Loose silt may show temporary liquefaction under static loading, and develop full liquefaction under cyclic loading. The most important factors influencing the silt behaviour are porosity, confining pressure, consolidation state, cyclic loading level and number of cycles. The maximum obtainable shear stress is primarily a function of the confining pressure and the internal frictional angle. The actual structure of the silt material is the key factor in controlling its behaviour.
基金supported by the National Natural Science Foundation of China(No.50009002)Minstry of Science and Technology Project(No.2001AA616020)Parts of this work were supported by the EU SCARCOST project
文摘Undrained behaviour of sand under low cell pressure was studied in static and cyclic triaxial tests. It was found that very loose sand liquefies under static loading with the relative density being a key parameter for the undrained behaviour of sand. In cyclic triaxial tests, pore water pressures built up during the cyclic loading and exceeded the confining cell pressure. This process was accompanied by a large sudden increase in axial deformation. The necessary number of cycles to obtain liquefaction was related to the confining cell pressure, the amplitude of cyclic loading and the relative density of sand.In addition, the patterns of pore water pressure response are different from those of sand samples with different relative densities. The test results are very useful for expounding scour mechanism around coastal structures since they relate to the low stress behaviour of the sand.
文摘A Bayesian approach using Markov chain Monte Carlo algorithms has been developed to analyze Smith’s discretized version of the discovery process model. It avoids the problems involved in the maximum likelihood method by effectively making use of the information from the prior distribution and that from the discovery sequence according to posterior probabilities. All statistical inferences about the parameters of the model and total resources can be quantified by drawing samples directly from the joint posterior distribution. In addition, statistical errors of the samples can be easily assessed and the convergence properties can be monitored during the sampling. Because the information contained in a discovery sequence is not enough to estimate all parameters, especially the number of fields, geologically justified prior information is crucial to the estimation. The Bayesian approach allows the analyst to specify his subjective estimates of the required parameters and his degree of uncertainty about the estimates in a clearly identified fashion throughout the analysis. As an example, this approach is applied to the same data of the North Sea on which Smith demonstrated his maximum likelihood method. For this case, the Bayesian approach has really improved the overly pessimistic results and downward bias of the maximum likelihood procedure.
基金supported by the National Natural Science Foundation (No. 50704028, 50904053)the Project 863 (No.2006AA09Z316)+1 种基金the Fundamental Research Funds for the Central Universities (No. CUGL100410)supported by the Opening Project of National Laboratory on Scientific Drilling, China University of Geosciences at Beijing (No. NLSD200901)
文摘As the oil and gas industries continue to increase their activity in deep water, gas hydrate hazards will become more serious and challenging, both at present and in the future. Accurate predictions of the hydrate-free zone and the suitable addition of salts and/or alcohols in preparing drilling fluids are particularly important both in preventing hydrate problems and decreasing the cost of drilling operations. In this paper, we compared several empirical correlations commonly used to estimate the hydrate inhibition effect of aqueous organic and electrolyte solutions using experiments with ethylene glycol （EG） as a hydrate inhibitor. The results show that the Najibi et al. correlation （for single and mixed thermodynamic inhibitors） and the Ostergaard et al. empirical correlation （for single thermodynamic inhibitors） are suitable for estimating the hydrate safety margin of oil-based drilling fluids （OBDFs） in the presence of thermodynamic hydrate inhibitors. According to the two correlations, the OBDF, composed of 1.6 L vaporizing oil, 2% emulsifying agent, 1% organobentonite, 0.5% SP-1, 1% LP-1, 10% water and 40% EG, can be safely used at a water depth of up to 1900 m. However, for more accurate predictions for drilling fluids, the effects of the solid phase, especially bentonite, on hydrate inhibition need to be considered and included in the application of these two empirical correlations.
文摘Artificial neural networks (ANNs) have been widely used to solve a number of problems to which analytical solutions are difficult to obtain using traditional mathematical approaches.Such problems exist also in the analysis of industrial robots. This paper presents an overview of ANN applications to robot kinematics, dynamics,control, trajectory and path planning, and sensing. Reasons for using or not using ANNs to industrial robots are explained as well.
文摘Technology development and implementation on the manufacturing arena has, throughout the years, frustrated employees over the requirement of constantly adjusting to new technologies. The employee's ability to learn is bounded, and thus limited. Looking at a modern production system, it is often a complicated array of technologies that have to be coordinated; or the technologies are coordinating themselves through artificial intelligence. This paper discusses the need for a holistic knowledge management approach to manufacturing,combining in-place technology with process management philosophy, in order for the firm's human resources to create, share, store and apply new knowledge.
文摘Traditionally, the objective of a manufacturing facility layout problem is to minimize the material handling cost of the manufacturing systems. Because of the combination of the facility layout problems, the genetic algorithms (GA) technique is the most promising approach for solving practical layout problems. Much of the previous work has been done for identical problems, where all departments are equal in area. In this paper, non-identical problems are dealt with. A new coding approach - the Comer Attachmer Structure (CAS) is introduced.
