The Experts below are selected from a list of 105 Experts worldwide ranked by ideXlab platform
Christian Holden - One of the best experts on this subject based on the ideXlab platform.
-
A High-Gain Observer for a Wet Gas Centrifugal Compressor
IFAC-PapersOnLine, 2019Co-Authors: Torstein Thode Kristoffersen, Christian HoldenAbstract:Abstract Subsea wet gas compression is an enabling boosting technology for cost-efficient production of small and remote gas condensate fields. These wet gas compressors are capable of compressing gases containing up to 5% liquid, removing the need for pre-separation and thereby reducing the investment and maintenance costs. However, the presence of liquid in the gas significantly changes the compression performance from that of dry gas, resulting in a reduced process gain and an increased Normal Operating Region for increasing liquid content. A novel backstepping process controller was previously proposed to track and stabilize a desired pressure reference inside the Normal Operating Region, requiring feedback from several variables, including the mass flow. Mass flow is an inaccurate and expensive measurement. Therefore, a full-order high-gain observer is derived for the wet gas compression process with proven local exponential convergence. The performance of the derived observer is studied in simulations.
-
CCTA - Modelling and Nonlinear Control of a Wet Gas Centrifugal Compressor
2018 IEEE Conference on Control Technology and Applications (CCTA), 2018Co-Authors: Torstein Thode Kristoffersen, Christian HoldenAbstract:Production of gas condensate from small and remote gas condensate fields require cost-efficient boosting technology for maintaining an economically satisfactory throughput. Wet gas compressors are a new boosting technology enabling boosting of gases containing up to 5% liquid per volume, removing the need for pre/bulk separation, resulting in lower investment and maintenance costs. However, introduction of liquid significantly changes the compression performance from that of dry gas compression, including the process gain and the Normal Operating Region, resulting in a challenging modelling and control problem. Therefore, in this paper, we extend the commonly applied dynamic model of Greitzer for dry gas compression with one additional state and extended polynomial approximation of the compressor characteristic. A nonlinear process control algorithm, using the angular velocity as input, is derived by applying backstepping, and local asymptotic stability proven via Lyapunov analysis. The control performance is studied in different simulations with and without saturation on the control input. The steady-state performance of the dynamic model is quantitatively validated against experimental data.
-
Applied Nonlinear Compressor Control with Gain Scheduling and State Estimation
IFAC-PapersOnLine, 2018Co-Authors: Njål Kjærnes Tengesdal, Torstein Thode Kristoffersen, Christian HoldenAbstract:Abstract Subsea gas compressors are key components for both increased and accelerated hydrocarbon production. Compressors enable the development of fields that are remote, in deep waters, or have low and/or reduced reservoir pressure. In this paper, we present a gain scheduling controller for a nonlinear centrifugal compressor model, coupled with an extended Kalman filter estimating an unmeasured system state. The controller is designed for the Normal Operating Region of a GT2252 turbocharger compressor section, planned for future small-scale experiments. The controller is developed with pole placement design on the equivalent linearized dynamics of the nonlinear model in a set of Operating points. The linearized closed-loop system in each Operating point is then provably locally asymptotically stable. The controller and estimator are analyzed in simulations, and show to achieve tracking of the desired plenum pressure using the estimate of the compressor mass flow.
-
Modelling and Nonlinear Control of a Wet Gas Centrifugal Compressor
2018 IEEE Conference on Control Technology and Applications (CCTA), 2018Co-Authors: Torstein Thode Kristoffersen, Christian HoldenAbstract:Production of gas condensate from small and remote gas condensate fields require cost-efficient boosting technology for maintaining an economically satisfactory throughput. Wet gas compressors are a new boosting technology enabling boosting of gases containing up to 5% liquid per volume, removing the need for pre/bulk separation, resulting in lower investment and maintenance costs. However, introduction of liquid significantly changes the compression performance from that of dry gas compression, including the process gain and the Normal Operating Region, resulting in a challenging modelling and control problem. Therefore, in this paper, we extend the commonly applied dynamic model of Greitzer for dry gas compression with one additional state and extended polynomial approximation of the compressor characteristic. A nonlinear process control algorithm, using the angular velocity as input, is derived by applying backstepping, and local asymptotic stability proven via Lyapunov analysis. The control performance is studied in different simulations with and without saturation on the control input. The steady-state performance of the dynamic model is quantitatively validated against experimental data.
-
Modeling and Control of a Wet-Gas Centrifugal Compressor
IEEE Transactions on Control Systems Technology, 1Co-Authors: Torstein Thode Kristoffersen, Christian HoldenAbstract:The development and production of small gas condensate discoveries require cost-efficient boosting technologies for profitable production. Subsea wet-gas compression is a new boosting technology, capable of boosting gas condensates containing up to 5% liquid per volume, removing the need for preseparation of the gas condensate, and resulting in lower investment and maintenance cost. Wet gas significantly changes the compressor performance compared with dry gas, including changing process gain and the Normal Operating Region, creating a challenging modeling and control problem. A wet-gas compression model with an empirical approximation of the pressure rise, including a complex backstepping controller, was recently proposed to solve these challenges. However, the model did not correctly model negative mass flows and the empirical polynomial provided limited insight into wet-gas compression. Therefore, in this article, we extend these dynamics to correctly model negative mass flows (deep surge) and, inspired by previous work on first-principle dry-gas compression modeling, we replace the empirical approximation with a first-principle wet-gas compressor characteristic. A nonlinear controller is designed via the Lyapunov analysis, and the local asymptotic stability is proven.
Torstein Thode Kristoffersen - One of the best experts on this subject based on the ideXlab platform.
-
A High-Gain Observer for a Wet Gas Centrifugal Compressor
IFAC-PapersOnLine, 2019Co-Authors: Torstein Thode Kristoffersen, Christian HoldenAbstract:Abstract Subsea wet gas compression is an enabling boosting technology for cost-efficient production of small and remote gas condensate fields. These wet gas compressors are capable of compressing gases containing up to 5% liquid, removing the need for pre-separation and thereby reducing the investment and maintenance costs. However, the presence of liquid in the gas significantly changes the compression performance from that of dry gas, resulting in a reduced process gain and an increased Normal Operating Region for increasing liquid content. A novel backstepping process controller was previously proposed to track and stabilize a desired pressure reference inside the Normal Operating Region, requiring feedback from several variables, including the mass flow. Mass flow is an inaccurate and expensive measurement. Therefore, a full-order high-gain observer is derived for the wet gas compression process with proven local exponential convergence. The performance of the derived observer is studied in simulations.
-
CCTA - Modelling and Nonlinear Control of a Wet Gas Centrifugal Compressor
2018 IEEE Conference on Control Technology and Applications (CCTA), 2018Co-Authors: Torstein Thode Kristoffersen, Christian HoldenAbstract:Production of gas condensate from small and remote gas condensate fields require cost-efficient boosting technology for maintaining an economically satisfactory throughput. Wet gas compressors are a new boosting technology enabling boosting of gases containing up to 5% liquid per volume, removing the need for pre/bulk separation, resulting in lower investment and maintenance costs. However, introduction of liquid significantly changes the compression performance from that of dry gas compression, including the process gain and the Normal Operating Region, resulting in a challenging modelling and control problem. Therefore, in this paper, we extend the commonly applied dynamic model of Greitzer for dry gas compression with one additional state and extended polynomial approximation of the compressor characteristic. A nonlinear process control algorithm, using the angular velocity as input, is derived by applying backstepping, and local asymptotic stability proven via Lyapunov analysis. The control performance is studied in different simulations with and without saturation on the control input. The steady-state performance of the dynamic model is quantitatively validated against experimental data.
-
Applied Nonlinear Compressor Control with Gain Scheduling and State Estimation
IFAC-PapersOnLine, 2018Co-Authors: Njål Kjærnes Tengesdal, Torstein Thode Kristoffersen, Christian HoldenAbstract:Abstract Subsea gas compressors are key components for both increased and accelerated hydrocarbon production. Compressors enable the development of fields that are remote, in deep waters, or have low and/or reduced reservoir pressure. In this paper, we present a gain scheduling controller for a nonlinear centrifugal compressor model, coupled with an extended Kalman filter estimating an unmeasured system state. The controller is designed for the Normal Operating Region of a GT2252 turbocharger compressor section, planned for future small-scale experiments. The controller is developed with pole placement design on the equivalent linearized dynamics of the nonlinear model in a set of Operating points. The linearized closed-loop system in each Operating point is then provably locally asymptotically stable. The controller and estimator are analyzed in simulations, and show to achieve tracking of the desired plenum pressure using the estimate of the compressor mass flow.
-
Modelling and Nonlinear Control of a Wet Gas Centrifugal Compressor
2018 IEEE Conference on Control Technology and Applications (CCTA), 2018Co-Authors: Torstein Thode Kristoffersen, Christian HoldenAbstract:Production of gas condensate from small and remote gas condensate fields require cost-efficient boosting technology for maintaining an economically satisfactory throughput. Wet gas compressors are a new boosting technology enabling boosting of gases containing up to 5% liquid per volume, removing the need for pre/bulk separation, resulting in lower investment and maintenance costs. However, introduction of liquid significantly changes the compression performance from that of dry gas compression, including the process gain and the Normal Operating Region, resulting in a challenging modelling and control problem. Therefore, in this paper, we extend the commonly applied dynamic model of Greitzer for dry gas compression with one additional state and extended polynomial approximation of the compressor characteristic. A nonlinear process control algorithm, using the angular velocity as input, is derived by applying backstepping, and local asymptotic stability proven via Lyapunov analysis. The control performance is studied in different simulations with and without saturation on the control input. The steady-state performance of the dynamic model is quantitatively validated against experimental data.
-
Modeling and Control of a Wet-Gas Centrifugal Compressor
IEEE Transactions on Control Systems Technology, 1Co-Authors: Torstein Thode Kristoffersen, Christian HoldenAbstract:The development and production of small gas condensate discoveries require cost-efficient boosting technologies for profitable production. Subsea wet-gas compression is a new boosting technology, capable of boosting gas condensates containing up to 5% liquid per volume, removing the need for preseparation of the gas condensate, and resulting in lower investment and maintenance cost. Wet gas significantly changes the compressor performance compared with dry gas, including changing process gain and the Normal Operating Region, creating a challenging modeling and control problem. A wet-gas compression model with an empirical approximation of the pressure rise, including a complex backstepping controller, was recently proposed to solve these challenges. However, the model did not correctly model negative mass flows and the empirical polynomial provided limited insight into wet-gas compression. Therefore, in this article, we extend these dynamics to correctly model negative mass flows (deep surge) and, inspired by previous work on first-principle dry-gas compression modeling, we replace the empirical approximation with a first-principle wet-gas compressor characteristic. A nonlinear controller is designed via the Lyapunov analysis, and the local asymptotic stability is proven.
P. Lindskog - One of the best experts on this subject based on the ideXlab platform.
-
Control Schemes in Polynomial Time
1993Co-Authors: I. Klein, P. LindskogAbstract:Of all hard- and software developed for industrial control purposes, the majority is devoted to sequential, or binary valued, control and only a minor part to classical linear control. The sequential parts of the controller are typically invoked during startup or shut-down phases to bring the system either into its Normal Operating Region or into some safe standby Region. Despite its importance, fairly little theoretical research has been devoted to this area, and sequential control programs are still created manually without much support for a systematic approach. We propose a method to create sequential control programs automatically and on-line upon request, for example when a plant fault has occurred. The main idea is to spend some effort off-line on modeling the process, and from this model generate the control strategy, i.e. the plan. Here we present a planning tool implemented in a real-time expert system called G2. The planning system contains algorithms for creating plans in form of minimal GRAFCET charts that show maximal parallelism. The algorithms can handle a restricted class of problems and for this class the complexity only increases polynomially with the number of state variables.
-
Automatic creation of sequential control schemes in polynomial time
Proceedings of 32nd IEEE Conference on Decision and Control, 1993Co-Authors: I. Klein, P. LindskogAbstract:Of all hard- and software developed for industrial control purposes, the majority is devoted to sequential, or binary valued, control and only a minor part to classical linear control. The sequential parts of the controller are typically invoked during startup or shut-down phases to bring the system either into its Normal Operating Region or into some safe standby Region. Despite its importance, fairly little theoretical research has been devoted to this area, and sequential control programs are still created manually without much support for a systematic approach. We propose a method to create sequential control programs automatically and online upon request, for example when a plant fault has occurred. The main idea is to spend some effort off-line on modeling the process, and from this model generate the control strategy, i.e. the plan. Here we present a planning tool implemented in a real-time expert system called G2. The planning system contains algorithms for creating plans in form of minimal GRAFCET charts that show maximal parallelism. The algorithms can handle a restricted class of problems and for this class the complexity only increases polynomially with the number of state variables.
Paul Migliore - One of the best experts on this subject based on the ideXlab platform.
-
variable speed operation of generators with rotor speed feedback in wind power applications
Journal of Solar Energy Engineering-transactions of The Asme, 1996Co-Authors: E Muljadi, C P Butterfield, Paul MiglioreAbstract:The use of induction generators in wind power applications has been common since the early development of the wind industry. Most of these generators operate at fixed frequency and are connected directly to the utility grid. Unfortunately, this mode of operation limits the rotor speed to a specific rpm. Variable-speed operation is preferred in order to facilitate maximum energy capture over a wide range of wind speeds. This paper explores variable-speed Operating strategies for wind turbine applications. The objectives are to maximize energy production, provide controlled start-up and reduce torque loading. This paper focuses on optimizing the energy captured by Operating at maximum aerodynamic efficiency at any wind speed. The control strategy we analyze uses rotor speed and generator power as the feedback signals. In the Normal Operating Region, rotor speed is used to compute a target power that corresponds to optimum operation. With power as the control objective, the power converter and generator are controlled to track the target power at any rpm. Thus, the torque-speed characteristic of the generator is shaped to optimize the energy capture. The target power is continuously updated at any rpm. in extreme areas of the Operating envelope, during start-up, shutdown, generator overload, ormore » overspeed, different strategies driven by other system considerations must be used.« less
-
variable speed operation of generators with rotor speed feedback in wind power applications
15. wind energy symposium Houston TX (United States) 28 Jan - 2 Feb 1996, 1995Co-Authors: E Muljadi, C P Butterfield, Paul MiglioreAbstract:The use of induction generators in wind power applications has been common since the early development of the wind industry. Most of these generators operate at fixed frequency and are connected directly to the utility grid. Unfortunately, this mode of operation limits the rotor speed to a specific rpm. Variable-speed operation is preferred in order to facilitate maximum energy capture over a wide range of wind speeds. This paper explores variable-speed Operating strategies for wind turbine applications. The objectives are to maximize energy production, provide controlled start-up and reduce torque loading. This paper focuses on optimizing the energy captured by Operating at maximum aerodynamic efficiency at any wind speed. The control strategy we analyze uses rotor speed and generator power as the feedback signals. In the Normal Operating Region, rotor speed is used to compute a target power that corresponds to optimum operation. With power as the control objective, the power converter and generator are controlled to track the target power at any rpm. Thus, the torque-speed characteristic of the generator is shaped to optimize the energy capture. The target power is continuously updated at any rpm. in extreme areas of the Operating envelope, during start-up, shutdown, generator overload, or overspeed, different strategies driven by other system considerations must be used.
I. Klein - One of the best experts on this subject based on the ideXlab platform.
-
Automatic Synthesis of Sequential Control Schemes
2013Co-Authors: I. KleinAbstract:Of all hard- and software developed for industrial control purposes, the majority is devoted to sequential, or binary valued, control and only a minor part to classical linear control. Typically, the sequential parts of the controller are invoked during startup and shut-down to bring the system into its Normal Operating Region and into some safe standby Region, respectively. Despite its importance, fairly little theoretical research has been devoted to this area, and sequential control programs are therefore still created manually without much theoretical support to obtain a systematic approach.We propose a method to create sequential control programs automatically. The main idea is to spend some eort off-line modelling the plant, and from this model generate the control strategy, that is the plan. The plant is modelled using action structures, thereby concentrating on the actions instead of the states of the plant. In general the planning problem shows exponential complexity in the number of state variables. However, by focusing on the actions, we can identify problem classes as well as algorithms such that the planning complexity is reduced to polynomial complexity. We prove that these algorithms are sound, i.e., the generated solution will solve the stated problem, and complete, i.e., if the algorithms fail, then no solution exists. The algorithms generate a plan as a set of actions and a partial order on this set specifying the execution order. The generated plan is proven to be minimal and maximally parallel.For a larger class of problems we propose a method to split the original problem into a number of simpler problems that can each be solved using one of the presented algorithms. It is also shown how a plan can be translated into a GRAFCET chart, and to illustrate these ideas we have implemented a planning tool, i.e., a system that is able to automatically create control schemes. Such a tool can of course also be used on-line if it is fast enough. This possibility opens up completely new applications such as operator supervision and simplied error recovery and restart procedures after a plant fault has occurred.Additionally we analyze reachability for a restricted class of problems. For this class we state a reachability criterion that may be checked using a slightly modified version of one of the above mentioned algorithms.
-
Control Schemes in Polynomial Time
1993Co-Authors: I. Klein, P. LindskogAbstract:Of all hard- and software developed for industrial control purposes, the majority is devoted to sequential, or binary valued, control and only a minor part to classical linear control. The sequential parts of the controller are typically invoked during startup or shut-down phases to bring the system either into its Normal Operating Region or into some safe standby Region. Despite its importance, fairly little theoretical research has been devoted to this area, and sequential control programs are still created manually without much support for a systematic approach. We propose a method to create sequential control programs automatically and on-line upon request, for example when a plant fault has occurred. The main idea is to spend some effort off-line on modeling the process, and from this model generate the control strategy, i.e. the plan. Here we present a planning tool implemented in a real-time expert system called G2. The planning system contains algorithms for creating plans in form of minimal GRAFCET charts that show maximal parallelism. The algorithms can handle a restricted class of problems and for this class the complexity only increases polynomially with the number of state variables.
-
Automatic creation of sequential control schemes in polynomial time
Proceedings of 32nd IEEE Conference on Decision and Control, 1993Co-Authors: I. Klein, P. LindskogAbstract:Of all hard- and software developed for industrial control purposes, the majority is devoted to sequential, or binary valued, control and only a minor part to classical linear control. The sequential parts of the controller are typically invoked during startup or shut-down phases to bring the system either into its Normal Operating Region or into some safe standby Region. Despite its importance, fairly little theoretical research has been devoted to this area, and sequential control programs are still created manually without much support for a systematic approach. We propose a method to create sequential control programs automatically and online upon request, for example when a plant fault has occurred. The main idea is to spend some effort off-line on modeling the process, and from this model generate the control strategy, i.e. the plan. Here we present a planning tool implemented in a real-time expert system called G2. The planning system contains algorithms for creating plans in form of minimal GRAFCET charts that show maximal parallelism. The algorithms can handle a restricted class of problems and for this class the complexity only increases polynomially with the number of state variables.
