混合照明系统的分数阶最小能耗搜索控制方法研究

Due to the limitations of some existing lighting control methods for complex hybrid lighting system, it have been a hot and difficult issue for developing an efficient method to minimize the energy consumption in a complicate hybrid lighting system. Hence, an efficient method should be developed to minimize the energy consumption in a complicate hybrid lighting system. This project proposes a novel fractional-order minimal energy consumption seeking control method for the complicate hybrid lighting system in the time varying environment. Furthermore, the fractional calculus operator theory, variable structure control strategy and parameter perturbation based extremum seeking strategy are utilized to build and analysis the fractional-order minimal energy consumption seeking control method. There are three main research contents in the project as follows: Firstly, the convergence property and stability of the control method are proven by analyzing the fractional-order sliding regulatory device and fractional-order driving function. Secondly, an algorithm of selecting parameters in the control method is obtained by investigating the relationship between the fractional order and the other parameters. Moreover, the algorithm could be applied to improve the control performance including the convergence speed and the control precision. Thirdly, by considering a multivariate environment, a fractional-order multi-input multi-output minimum energy consumption seeking control method is designed by studying the coupled responses relationship among the inputs. This method can guarantee the minimum energy consumption as well as the comfortable light level. Successful implementation of this project can not only promote the development of optimal control theory but also improve the electricity utilizing efficiency, which has both theoretical and practical applications.

现有控制方法在解决复杂混合照明系统能耗过高的问题时都存在一定的局限性，发展更加科学合理的控制方法，以实现照明系统能耗最小化，已成为当今社会研究的热点和难点。本项目针对动态环境下复杂混合照明系统最小能耗问题，提出分数阶最小能耗极值搜索算法，结合分数阶微积分理论、变结构控制策略和参数扰动极值搜索策略，重点研究以下内容：1) 通过分析分数阶滑模调节机构和分数阶激励函数，研究分数阶最小能耗搜索控制算法的收敛特性和稳定性；2) 通过挖掘分数阶算子的阶次和其他参数的函数关系，研究如何提高动态收敛速度和稳态跟踪精度，并建立此算法的参数调节法则；3) 针对多变量环境，探讨多输入信号间耦合响应关系，研究最小能耗问题的分数阶多变量极值搜索控制算法。项目预期在保证目标区域达到设定照度的同时，实现时变环境中混合照明系统能耗最小化。研究成果既能丰富优化控制理论，又能有效提高能源利用率，具有重大的理论意义和应用价值。

现有控制方法在解决复杂混合照明系统能耗过高的问题时都存在一定的局限性，发展更加科学合理的控制方法，以实现照明系统能耗最小化，已成为当今社会研究的热点和难点。本项目针对动态环境下复杂混合照明系统最小能耗问题，提出分数阶最小能耗极值搜索算法，结合分数阶微积分理论、变结构控制策略和参数扰动极值搜索策略，重.点研究以下内容：1)通过分析极值搜索控制机构和分数阶激励函数，研究了分数阶最小能耗搜索控制算法的收敛特性和稳定性；2)通过挖掘分数阶算子的阶次和其他参数的函数关系，研究了如何提高动态收敛速度和稳态跟踪精度，并建立此算法的参数调节法则；3)针对多变量环境，探讨多输入信号间耦合响应关系，提出了最小能耗问题的分数阶多变量极值搜索控制算法。项目研究成果实现了在保证目标区域达到设定照度的同时，减小时变环境中混合照明系统能耗。研究成果既能丰富优化控制理论，又能有效提高能源利用率，具有重大的理论意义和应用价值。