A reliable and sensitive technique for monitoring tool condition in drilling is essential help fo... more A reliable and sensitive technique for monitoring tool condition in drilling is essential help for practising engineers. It is commonly known that the unattended use of a drill bit until it reaches the ultimate failure can potentially damage to machine tool and work-piece resulting in considerable down time and productivity loss. Thus there is a need for such tools to
ABSTRACT This paper presents a comparison of predictive models for the estimation of engine power... more ABSTRACT This paper presents a comparison of predictive models for the estimation of engine power and tailpipe emissions for a 4 kW gasoline scooter. This study forms a benchmark toward establishing an online emissions control and monitoring system to bring the emissions to within specific limits. Three emissions predictive models were investigated in this study; direct and series artificial neural network (ANN) models and a MATLAB dynamic model. The direct models takes variables lambda, throttle position, engine and vehicle speed to predict the engine power and the emissions CO, CO2 and HC. The series model first takes the mentioned input to predict the engine power and consequently using the engine power as the fifth input to predict the emissions. For the ANN models, two multilayered networks were compared and analyzed; the backpropagation (BP) and optimization layer-by-layer (OLL) algorithms. The predictive accuracy for each algorithm were compared and it was found that the OLL network is the most accurate with a maximum mean relative error (MRE) of 1.78% and 1.38% for the direct and series predictive model respectively. Comparative results showed that the series neural network model gives the most accurate predictions, with MRE of 0.63% and 0.47% for the engine power and emissions respectively. The series neural network model can be seen as generic virtual power and emissions sensors, substituting costly and cumbersome hardware. Simple obtainable process parameters together with the series neural network will contribute immensely in control and tuning of emissions for real-time vehicular applications.
International Journal of Machine Tools & Manufacture, Aug 1, 1994
The various types of rotary tool cutting operations and their performance advantage over conventi... more The various types of rotary tool cutting operations and their performance advantage over conventional machining operations are reviewed together with the simplified or fundamental operations used to study the cutting mechanics of these novel material removal operations. The fundamental driven and self-propelled rotary tool cutting operations have been simulated as wedge tool cutting processes and their relationships to the kinematically and perfectly equivalent "classical" orthogonal and oblique cutting processes explored. Based on the modified thin shear zone cutting models the simulated driven orthogonal and oblique rotary tool cutting processes have been analysed and represented as perfectly equivalent classical oblique cutting processes accounting for all the cutting forces and energy, together with a chip transportation process, owing to the tool lateral motion along its cutting edge, which involves no additional energy. By contrast, the self-propelled oblique rotary tool cutting process has been modelled as a perfectly equivalent classical orthogonal cutting process together with chip transportation. The proposed models will be experimentally verified in the next part of this investigation.
Primary aluminium is produced using a highly dynamic and unstable technique known as the Hall-Her... more Primary aluminium is produced using a highly dynamic and unstable technique known as the Hall-Heroult process. An important consideration for aluminium smelting is minimisation of process variation, which is monitored by measuring particular parameters of the Hall-Heroult process and administering corrective action as appropriate to return or maintain the process within a predetermined control range. A critical parameter to be
While there exists a broad range of neural networks for a particular task, different neural netwo... more While there exists a broad range of neural networks for a particular task, different neural network architectures are selected depending upon the nature of application in industry. The range of applications covers anything from performance estimation and pattern recognition to process modelling and control. The network selection can be carried out based on economic considerations, such as cost associated with
... 3. Neural Network Modelling Results for Electrolyte Additive Prediction Application In order ... more ... 3. Neural Network Modelling Results for Electrolyte Additive Prediction Application In order to develop an accurate model of the Hall-Heroult process using the GRNN it is necessary to study RMS error behaviour with changing network architecture. ...
Neural networks, due to their excellent capabilities for modelling process behaviour, are gaining... more Neural networks, due to their excellent capabilities for modelling process behaviour, are gaining precedence over traditional empirical modelling techniques, such as statistical methods. While neural networks have good ability to map any reasonable continuous function, they do not explain easily how the inputs are related to an output, and also whether the selected inputs have any significant relationship with an output. There is quite often a need to identify some order of influence of the input variables on the output variable. In this paper, a technique for determining the order of influence of the n elements of the input vector on the m elements of the output vector is presented and discussed. While a sample mathematical function is used to introduce the technique, a more practical application of this method in the aluminium smelting industry is considered. It is shown that using a sensitivity analysis on the backpropagation algorithm the degree of influence of input parameters on the output error can be successfully estimated.
... Hence, the particular application in this instance is to use neural networks to predict the A... more ... Hence, the particular application in this instance is to use neural networks to predict the AlF, content of the Hall-Heroult electrolyte based on ... can be seen from the resulting RMS error for the BP and GRNN models that in both instances the network was able to model the process ...
International Journal of Hydrogen Energy, Sep 1, 2010
Predictive models were built using neural network based Adaptive Neuro-Fuzzy Inference Systems fo... more Predictive models were built using neural network based Adaptive Neuro-Fuzzy Inference Systems for hydrogen flow rate, electrolyzer system-efficiency and stack-efficiency respectively. A comprehensive experimental database forms the foundation for the predictive models. It is argued that, due to the high costs associated with the hydrogen measuring equipment; these reliable predictive models can be implemented as virtual sensors. These models can also be used on-line for monitoring and safety of hydrogen equipment. The quantitative accuracy of the predictive models is appraised using statistical techniques. These mathematical models are found to be reliable predictive tools with an excellent accuracy of AE3% compared with experimental values. The predictive nature of these models did not show any significant bias to either over prediction or under prediction. These predictive models, built on a sound mathematical and quantitative basis, can be seen as a step towards establishing hydrogen performance prediction models as generic virtual sensors for wider safety and monitoring applications.
Effect of heat treatment on the mechanical properties of box and frame-type corner joints produce... more Effect of heat treatment on the mechanical properties of box and frame-type corner joints produced from modified and non-modified beech (Fagus sylvatica L.) and spruce (Picea abies L.) wood was investigated. The wood lamellas were heat-treated at 190 and 210°C for 3 h in a treatment process with initial vacuum. Samples of boxand frame-type corner joints used in the production of furniture were prepared from the heat-modified and nonmodified lamellas using dowels and biscuits. The dowel joints had higher failure load than the biscuit joints.
Journal of Materials Processing Technology, Sep 1, 1991
Rotary tool cutting processes are described and simulated as 'orthogonal' and 'obiique'. A modifi... more Rotary tool cutting processes are described and simulated as 'orthogonal' and 'obiique'. A modified thin shear zone mechanics of cutting modei for driven and self-propelied 'oblique' rotary cutting tool processes are developed and experimentally verified, qualitatively and quantitatively over a wide range of process variables. The cutting model has been represented as an 'equivalent' classical oblique cutting process, accounting for all the cutting energy, together with a chip transportation process due to the lateral tool movement along the cutting edge, which involves no additional energy. From a predictive point of view, the forces, power and absolute chip flow direction in 'oblique' rotary tool cutting processes can be obtained from the 'equivalent' classical oblique cutting process predictions and the chip transportation analysis in the proposed model. The fundamental and practical implications of the model are discussed.
... Vr will depend on both Vy and is while V will be lower than VU, such that V = Vwcosis (26) h(... more ... Vr will depend on both Vy and is while V will be lower than VU, such that V = Vwcosis (26) h(or rtr or ri) ,b(or 8") ,K?P,KI~,RIR) (23a) ( - ) Vr = Vusinit (27 ... given as: rt = rtr = ri.cos$/cosis (28 tang =tanit/rt (29 tan@ = rtcosan/( 1-rtsinan) = rLrcosan/( 1-rtrsinan) (30 (31 (32 (30a COSQ ...
The importance of oblique cutting as a representative for many practical machining operations is ... more The importance of oblique cutting as a representative for many practical machining operations is discussed. A few of the existing oblique cutting models and their deficiencies are discussed from a predictive point of view. A neural network architecture is developed to predict the forces and power in single edged oblique cutting operation. Experiments are carried out over a comprehensive range
International Journal of Machine Tools & Manufacture, Aug 1, 1994
The various types of rotary tool cutting operations and their performance advantage over conventi... more The various types of rotary tool cutting operations and their performance advantage over conventional machining operations are reviewed together with the simplified or fundamental operations used to study the cutting mechanics of these novel material removal operations. The fundamental driven and self-propelled rotary tool cutting operations have been simulated as wedge tool cutting processes and their relationships to the kinematically and perfectly equivalent "classical" orthogonal and oblique cutting processes explored. Based on the modified thin shear zone cutting models the simulated driven orthogonal and oblique rotary tool cutting processes have been analysed and represented as perfectly equivalent classical oblique cutting processes accounting for all the cutting forces and energy, together with a chip transportation process, owing to the tool lateral motion along its cutting edge, which involves no additional energy. By contrast, the self-propelled oblique rotary tool cutting process has been modelled as a perfectly equivalent classical orthogonal cutting process together with chip transportation. The proposed models will be experimentally verified in the next part of this investigation.
The importance and industrial applications of filled polymers is reported. In recent years cast n... more The importance and industrial applications of filled polymers is reported. In recent years cast nylon with various fillers has been extensively used in industry, where machining is carried out as a secondary process to achieve necessary surface finish and shape. There is a pressing need to develop various cutting techniques to enhance the machining performance features for these polymers. An understanding of the effect of major process variables on forces, power, and surface finish help practicing engineers in selection of cutting tools and machine tools for efficient manufacturing. In this paper, the qualitative and quantitative effects of major process variables such as tool, cut geometry, and cutting speeds are studied for cast nylon with two different fillers, namely MoS2 and vulcanized vegetable oil (VVO). The data processing is carried out by a suite of statistical routines.
... Regression Neural Networks V. Karri School of Engineering, University of Tasmania GPO Box 252... more ... Regression Neural Networks V. Karri School of Engineering, University of Tasmania GPO Box 252-65, Hobart, Tasmania, 7001, Australia [email protected] Abstract. ... There are i input neurodes, j pattern neurodes, k+1 summation neurodes and k output neurodes. A 1 B sj ...
A reliable and sensitive technique for monitoring tool condition in drilling is essential help fo... more A reliable and sensitive technique for monitoring tool condition in drilling is essential help for practising engineers. It is commonly known that the unattended use of a drill bit until it reaches the ultimate failure can potentially damage to machine tool and work-piece resulting in considerable down time and productivity loss. Thus there is a need for such tools to
ABSTRACT This paper presents a comparison of predictive models for the estimation of engine power... more ABSTRACT This paper presents a comparison of predictive models for the estimation of engine power and tailpipe emissions for a 4 kW gasoline scooter. This study forms a benchmark toward establishing an online emissions control and monitoring system to bring the emissions to within specific limits. Three emissions predictive models were investigated in this study; direct and series artificial neural network (ANN) models and a MATLAB dynamic model. The direct models takes variables lambda, throttle position, engine and vehicle speed to predict the engine power and the emissions CO, CO2 and HC. The series model first takes the mentioned input to predict the engine power and consequently using the engine power as the fifth input to predict the emissions. For the ANN models, two multilayered networks were compared and analyzed; the backpropagation (BP) and optimization layer-by-layer (OLL) algorithms. The predictive accuracy for each algorithm were compared and it was found that the OLL network is the most accurate with a maximum mean relative error (MRE) of 1.78% and 1.38% for the direct and series predictive model respectively. Comparative results showed that the series neural network model gives the most accurate predictions, with MRE of 0.63% and 0.47% for the engine power and emissions respectively. The series neural network model can be seen as generic virtual power and emissions sensors, substituting costly and cumbersome hardware. Simple obtainable process parameters together with the series neural network will contribute immensely in control and tuning of emissions for real-time vehicular applications.
International Journal of Machine Tools & Manufacture, Aug 1, 1994
The various types of rotary tool cutting operations and their performance advantage over conventi... more The various types of rotary tool cutting operations and their performance advantage over conventional machining operations are reviewed together with the simplified or fundamental operations used to study the cutting mechanics of these novel material removal operations. The fundamental driven and self-propelled rotary tool cutting operations have been simulated as wedge tool cutting processes and their relationships to the kinematically and perfectly equivalent "classical" orthogonal and oblique cutting processes explored. Based on the modified thin shear zone cutting models the simulated driven orthogonal and oblique rotary tool cutting processes have been analysed and represented as perfectly equivalent classical oblique cutting processes accounting for all the cutting forces and energy, together with a chip transportation process, owing to the tool lateral motion along its cutting edge, which involves no additional energy. By contrast, the self-propelled oblique rotary tool cutting process has been modelled as a perfectly equivalent classical orthogonal cutting process together with chip transportation. The proposed models will be experimentally verified in the next part of this investigation.
Primary aluminium is produced using a highly dynamic and unstable technique known as the Hall-Her... more Primary aluminium is produced using a highly dynamic and unstable technique known as the Hall-Heroult process. An important consideration for aluminium smelting is minimisation of process variation, which is monitored by measuring particular parameters of the Hall-Heroult process and administering corrective action as appropriate to return or maintain the process within a predetermined control range. A critical parameter to be
While there exists a broad range of neural networks for a particular task, different neural netwo... more While there exists a broad range of neural networks for a particular task, different neural network architectures are selected depending upon the nature of application in industry. The range of applications covers anything from performance estimation and pattern recognition to process modelling and control. The network selection can be carried out based on economic considerations, such as cost associated with
... 3. Neural Network Modelling Results for Electrolyte Additive Prediction Application In order ... more ... 3. Neural Network Modelling Results for Electrolyte Additive Prediction Application In order to develop an accurate model of the Hall-Heroult process using the GRNN it is necessary to study RMS error behaviour with changing network architecture. ...
Neural networks, due to their excellent capabilities for modelling process behaviour, are gaining... more Neural networks, due to their excellent capabilities for modelling process behaviour, are gaining precedence over traditional empirical modelling techniques, such as statistical methods. While neural networks have good ability to map any reasonable continuous function, they do not explain easily how the inputs are related to an output, and also whether the selected inputs have any significant relationship with an output. There is quite often a need to identify some order of influence of the input variables on the output variable. In this paper, a technique for determining the order of influence of the n elements of the input vector on the m elements of the output vector is presented and discussed. While a sample mathematical function is used to introduce the technique, a more practical application of this method in the aluminium smelting industry is considered. It is shown that using a sensitivity analysis on the backpropagation algorithm the degree of influence of input parameters on the output error can be successfully estimated.
... Hence, the particular application in this instance is to use neural networks to predict the A... more ... Hence, the particular application in this instance is to use neural networks to predict the AlF, content of the Hall-Heroult electrolyte based on ... can be seen from the resulting RMS error for the BP and GRNN models that in both instances the network was able to model the process ...
International Journal of Hydrogen Energy, Sep 1, 2010
Predictive models were built using neural network based Adaptive Neuro-Fuzzy Inference Systems fo... more Predictive models were built using neural network based Adaptive Neuro-Fuzzy Inference Systems for hydrogen flow rate, electrolyzer system-efficiency and stack-efficiency respectively. A comprehensive experimental database forms the foundation for the predictive models. It is argued that, due to the high costs associated with the hydrogen measuring equipment; these reliable predictive models can be implemented as virtual sensors. These models can also be used on-line for monitoring and safety of hydrogen equipment. The quantitative accuracy of the predictive models is appraised using statistical techniques. These mathematical models are found to be reliable predictive tools with an excellent accuracy of AE3% compared with experimental values. The predictive nature of these models did not show any significant bias to either over prediction or under prediction. These predictive models, built on a sound mathematical and quantitative basis, can be seen as a step towards establishing hydrogen performance prediction models as generic virtual sensors for wider safety and monitoring applications.
Effect of heat treatment on the mechanical properties of box and frame-type corner joints produce... more Effect of heat treatment on the mechanical properties of box and frame-type corner joints produced from modified and non-modified beech (Fagus sylvatica L.) and spruce (Picea abies L.) wood was investigated. The wood lamellas were heat-treated at 190 and 210°C for 3 h in a treatment process with initial vacuum. Samples of boxand frame-type corner joints used in the production of furniture were prepared from the heat-modified and nonmodified lamellas using dowels and biscuits. The dowel joints had higher failure load than the biscuit joints.
Journal of Materials Processing Technology, Sep 1, 1991
Rotary tool cutting processes are described and simulated as 'orthogonal' and 'obiique'. A modifi... more Rotary tool cutting processes are described and simulated as 'orthogonal' and 'obiique'. A modified thin shear zone mechanics of cutting modei for driven and self-propelied 'oblique' rotary cutting tool processes are developed and experimentally verified, qualitatively and quantitatively over a wide range of process variables. The cutting model has been represented as an 'equivalent' classical oblique cutting process, accounting for all the cutting energy, together with a chip transportation process due to the lateral tool movement along the cutting edge, which involves no additional energy. From a predictive point of view, the forces, power and absolute chip flow direction in 'oblique' rotary tool cutting processes can be obtained from the 'equivalent' classical oblique cutting process predictions and the chip transportation analysis in the proposed model. The fundamental and practical implications of the model are discussed.
... Vr will depend on both Vy and is while V will be lower than VU, such that V = Vwcosis (26) h(... more ... Vr will depend on both Vy and is while V will be lower than VU, such that V = Vwcosis (26) h(or rtr or ri) ,b(or 8") ,K?P,KI~,RIR) (23a) ( - ) Vr = Vusinit (27 ... given as: rt = rtr = ri.cos$/cosis (28 tang =tanit/rt (29 tan@ = rtcosan/( 1-rtsinan) = rLrcosan/( 1-rtrsinan) (30 (31 (32 (30a COSQ ...
The importance of oblique cutting as a representative for many practical machining operations is ... more The importance of oblique cutting as a representative for many practical machining operations is discussed. A few of the existing oblique cutting models and their deficiencies are discussed from a predictive point of view. A neural network architecture is developed to predict the forces and power in single edged oblique cutting operation. Experiments are carried out over a comprehensive range
International Journal of Machine Tools & Manufacture, Aug 1, 1994
The various types of rotary tool cutting operations and their performance advantage over conventi... more The various types of rotary tool cutting operations and their performance advantage over conventional machining operations are reviewed together with the simplified or fundamental operations used to study the cutting mechanics of these novel material removal operations. The fundamental driven and self-propelled rotary tool cutting operations have been simulated as wedge tool cutting processes and their relationships to the kinematically and perfectly equivalent "classical" orthogonal and oblique cutting processes explored. Based on the modified thin shear zone cutting models the simulated driven orthogonal and oblique rotary tool cutting processes have been analysed and represented as perfectly equivalent classical oblique cutting processes accounting for all the cutting forces and energy, together with a chip transportation process, owing to the tool lateral motion along its cutting edge, which involves no additional energy. By contrast, the self-propelled oblique rotary tool cutting process has been modelled as a perfectly equivalent classical orthogonal cutting process together with chip transportation. The proposed models will be experimentally verified in the next part of this investigation.
The importance and industrial applications of filled polymers is reported. In recent years cast n... more The importance and industrial applications of filled polymers is reported. In recent years cast nylon with various fillers has been extensively used in industry, where machining is carried out as a secondary process to achieve necessary surface finish and shape. There is a pressing need to develop various cutting techniques to enhance the machining performance features for these polymers. An understanding of the effect of major process variables on forces, power, and surface finish help practicing engineers in selection of cutting tools and machine tools for efficient manufacturing. In this paper, the qualitative and quantitative effects of major process variables such as tool, cut geometry, and cutting speeds are studied for cast nylon with two different fillers, namely MoS2 and vulcanized vegetable oil (VVO). The data processing is carried out by a suite of statistical routines.
... Regression Neural Networks V. Karri School of Engineering, University of Tasmania GPO Box 252... more ... Regression Neural Networks V. Karri School of Engineering, University of Tasmania GPO Box 252-65, Hobart, Tasmania, 7001, Australia [email protected] Abstract. ... There are i input neurodes, j pattern neurodes, k+1 summation neurodes and k output neurodes. A 1 B sj ...
Uploads
Papers by Vishy Karri