Operation of a dielectric elastomer(DE)transducer is limited by the way it fails,and any other limits imposed by the user.For example,material rupture and dielectric breakdown are two examples of transducer failure,wh...Operation of a dielectric elastomer(DE)transducer is limited by the way it fails,and any other limits imposed by the user.For example,material rupture and dielectric breakdown are two examples of transducer failure,while tension loss is a user-imposed limit.Electromechanical instability may or may not result in transducer failure,but is often considered as a boundary to denote the onset of state transition.For a DE undergoing homogeneous deformation,constructing operational limits on work-conjugate state diagrams are rather straight-forward.However,for DE subjected to non-homogeneous deformation,the process is more complex.We present a method to plot the operational boundaries of a DE subject to non-homogeneous deformation,known commonly as the Universal Muscle Actuator or the loudspeaker configuration.Our analysis show that tension loss need not be imposed as a hard operational limit,and that there is further operational space for the transducer,post tension loss.By plotting the operational limits on work-conjugate planes,we determine an optimal inner-to-outer ring ratio and the required pre-stretches that maximize energy conversion.展开更多
This paper develops a real-time control method based on deep reinforcement learning aimed to determine the optimal control actions to maintain a sufficient secure operating limit.The secure operating limit refers to t...This paper develops a real-time control method based on deep reinforcement learning aimed to determine the optimal control actions to maintain a sufficient secure operating limit.The secure operating limit refers to the limit to the most stressed pre-contingency operating point of an electric power system that can withstand a set of credible contingencies without violating stability criteria.The developed deep reinforcement learning method uses a hybrid control scheme that is capable of simultaneously adjusting both discrete and continuous action variables.The performance is evaluated on a modified version of the Nordic32 test system.The results show that the developed deep reinforcement learning method quickly learns an effective control policy to ensure a sufficient secure operating limit for a range of different system scenarios.The performance is also compared to a control based on a rule-based look-up table and a deep reinforcement learning control adapted for discrete action spaces.The hybrid deep reinforcement learning control managed to achieve significantly better on all of the defined test sets,indicating that the possibility of adjusting both discrete and continuous action variables resulted in a more flexible and efficient control policy.展开更多
In order to make the power level angle(PLA)have the same bijection relationship with the thrust when the environment changes before operational limit exceedance,and ensure the pilot's ability of unrestricted throt...In order to make the power level angle(PLA)have the same bijection relationship with the thrust when the environment changes before operational limit exceedance,and ensure the pilot's ability of unrestricted throttle movement as well as the safety of the engine,a method used to design a constant thrust power schedule was proposed.The research was based on a simulink turbofan engine model and the results were organized as look-up tables in different structures with limitations.Simulation was done to confirm the validity and demonstrate the effectiveness of the schedule obtained with the approach.The constant thrust takeoff schedule,constant thrust idle schedule and constant thrust part power schedule with the present constant thrust control(CTC)method were proved to be valid.The results can be applied to this specific engine but the method can be applied to any areoengine and can be even extended to the unmanned plane.展开更多
基金This work was supported by the Agency for Science,Technology and Research(Singapore)and Department of Science&Technology(India)(A*STAR-DST)[R-265-000-524-305(SERC 1425203140)]。
文摘Operation of a dielectric elastomer(DE)transducer is limited by the way it fails,and any other limits imposed by the user.For example,material rupture and dielectric breakdown are two examples of transducer failure,while tension loss is a user-imposed limit.Electromechanical instability may or may not result in transducer failure,but is often considered as a boundary to denote the onset of state transition.For a DE undergoing homogeneous deformation,constructing operational limits on work-conjugate state diagrams are rather straight-forward.However,for DE subjected to non-homogeneous deformation,the process is more complex.We present a method to plot the operational boundaries of a DE subject to non-homogeneous deformation,known commonly as the Universal Muscle Actuator or the loudspeaker configuration.Our analysis show that tension loss need not be imposed as a hard operational limit,and that there is further operational space for the transducer,post tension loss.By plotting the operational limits on work-conjugate planes,we determine an optimal inner-to-outer ring ratio and the required pre-stretches that maximize energy conversion.
文摘This paper develops a real-time control method based on deep reinforcement learning aimed to determine the optimal control actions to maintain a sufficient secure operating limit.The secure operating limit refers to the limit to the most stressed pre-contingency operating point of an electric power system that can withstand a set of credible contingencies without violating stability criteria.The developed deep reinforcement learning method uses a hybrid control scheme that is capable of simultaneously adjusting both discrete and continuous action variables.The performance is evaluated on a modified version of the Nordic32 test system.The results show that the developed deep reinforcement learning method quickly learns an effective control policy to ensure a sufficient secure operating limit for a range of different system scenarios.The performance is also compared to a control based on a rule-based look-up table and a deep reinforcement learning control adapted for discrete action spaces.The hybrid deep reinforcement learning control managed to achieve significantly better on all of the defined test sets,indicating that the possibility of adjusting both discrete and continuous action variables resulted in a more flexible and efficient control policy.
文摘In order to make the power level angle(PLA)have the same bijection relationship with the thrust when the environment changes before operational limit exceedance,and ensure the pilot's ability of unrestricted throttle movement as well as the safety of the engine,a method used to design a constant thrust power schedule was proposed.The research was based on a simulink turbofan engine model and the results were organized as look-up tables in different structures with limitations.Simulation was done to confirm the validity and demonstrate the effectiveness of the schedule obtained with the approach.The constant thrust takeoff schedule,constant thrust idle schedule and constant thrust part power schedule with the present constant thrust control(CTC)method were proved to be valid.The results can be applied to this specific engine but the method can be applied to any areoengine and can be even extended to the unmanned plane.
