An Approach for Actuation Specification and Synthesis of Dynamic SystemsSource: Journal of Dynamic Systems, Measurement, and Control:;2009:;volume( 131 ):;issue: 003::page 31012DOI: 10.1115/1.3089558Publisher: The American Society of Mechanical Engineers (ASME)
Abstract: Refinement or improvement of a dynamic system to meet a frequency response specification can benefit from the option to use passive or active compensation, or a combination of both. The process becomes more effective when supplemented with methods derived from classical network theory to synthesize candidate designs for actuators and their control systems. The synthesis procedure presented here provides an explicit way to formulate system topologies that employ passive and active elements to achieve a desired targeted performance specification, i.e., frequency response. Active elements are used to represent elements that are not physically realizable, such as negative impedances and elements that have ill-defined connectivity. A working premise is that these elements indicate the need for actuation technology. Coupled with a topological description of the system, the synthesis procedure provides a systematic approach that offers design solutions not previously conceived of through insight or experience. These “first draft” designs can be improved upon by later utilizing complementary approaches, such as optimization methods, as dictated by detailed system requirements and operating regimes. The flexibility of this synthesis approach allows the consideration of design restrictions unrelated to frequency response, but critical nonetheless in assessing the viability of candidate designs. Further, the procedure does not require assumption of a particular control/compensation architecture at the outset; this renders novel architectures that depart from traditional architectures such as proportional integral, propotional-integral-derivative, etc. The procedure is couched within a simulation basis, so that extension to state-space simulation and thus growth of the system and inclusion of more complex and nonlinear representations become possible. The concept of a virtual state space is introduced, which is integral to the development of controller architectures and associated parameters. It is found that customized passive/active compensation systems can be derived using a bond graph approach, making this approach more easily applicable to multi-energetic systems. Examples are used to demonstrate the approach, including a case study of an electromechanical vehicle suspension, from which an experimental model is derived to illustrate the synthesis procedure. Comparison of results between these examples illustrate the practical utility of the synthesis procedure. In particular, simulations reveal that increasing the number of realized passive elements for a particular system does not necessarily minimize actuator energy consumption. Detailed analysis of synthesis results show that certain design candidates feature active devices that work against either passive elements or other active devices within the system.
|
Show full item record
| contributor author | Thomas J. Connolly | |
| contributor author | Raul G. Longoria | |
| date accessioned | 2017-05-09T00:32:12Z | |
| date available | 2017-05-09T00:32:12Z | |
| date copyright | May, 2009 | |
| date issued | 2009 | |
| identifier issn | 0022-0434 | |
| identifier other | JDSMAA-26493#031012_1.pdf | |
| identifier uri | http://yetl.yabesh.ir/yetl/handle/yetl/140221 | |
| description abstract | Refinement or improvement of a dynamic system to meet a frequency response specification can benefit from the option to use passive or active compensation, or a combination of both. The process becomes more effective when supplemented with methods derived from classical network theory to synthesize candidate designs for actuators and their control systems. The synthesis procedure presented here provides an explicit way to formulate system topologies that employ passive and active elements to achieve a desired targeted performance specification, i.e., frequency response. Active elements are used to represent elements that are not physically realizable, such as negative impedances and elements that have ill-defined connectivity. A working premise is that these elements indicate the need for actuation technology. Coupled with a topological description of the system, the synthesis procedure provides a systematic approach that offers design solutions not previously conceived of through insight or experience. These “first draft” designs can be improved upon by later utilizing complementary approaches, such as optimization methods, as dictated by detailed system requirements and operating regimes. The flexibility of this synthesis approach allows the consideration of design restrictions unrelated to frequency response, but critical nonetheless in assessing the viability of candidate designs. Further, the procedure does not require assumption of a particular control/compensation architecture at the outset; this renders novel architectures that depart from traditional architectures such as proportional integral, propotional-integral-derivative, etc. The procedure is couched within a simulation basis, so that extension to state-space simulation and thus growth of the system and inclusion of more complex and nonlinear representations become possible. The concept of a virtual state space is introduced, which is integral to the development of controller architectures and associated parameters. It is found that customized passive/active compensation systems can be derived using a bond graph approach, making this approach more easily applicable to multi-energetic systems. Examples are used to demonstrate the approach, including a case study of an electromechanical vehicle suspension, from which an experimental model is derived to illustrate the synthesis procedure. Comparison of results between these examples illustrate the practical utility of the synthesis procedure. In particular, simulations reveal that increasing the number of realized passive elements for a particular system does not necessarily minimize actuator energy consumption. Detailed analysis of synthesis results show that certain design candidates feature active devices that work against either passive elements or other active devices within the system. | |
| publisher | The American Society of Mechanical Engineers (ASME) | |
| title | An Approach for Actuation Specification and Synthesis of Dynamic Systems | |
| type | Journal Paper | |
| journal volume | 131 | |
| journal issue | 3 | |
| journal title | Journal of Dynamic Systems, Measurement, and Control | |
| identifier doi | 10.1115/1.3089558 | |
| journal fristpage | 31012 | |
| identifier eissn | 1528-9028 | |
| tree | Journal of Dynamic Systems, Measurement, and Control:;2009:;volume( 131 ):;issue: 003 | |
| contenttype | Fulltext |