基于預測控制的寬范圍輸入直流變換器
發(fā)布時間:2019-06-01 14:41
【摘要】:在燃料電池供電系統(tǒng)、新能源發(fā)電系統(tǒng)等領域,寬范圍輸入直流變換器得到了廣泛應用,其中,雙管Buck-Boost變換器因具有輸入輸出電壓極性相同、開關管電壓應力低等優(yōu)點成為研究熱點。雙管Buck-Boost變換器的傳統(tǒng)控制方法根據(jù)輸入電壓范圍選擇工作模式,本文采用模型預測控制,通過在線優(yōu)化選擇最優(yōu)的工作模式,為寬范圍輸入直流變換器的控制器設計提供一種方案。本文從雙管Buck-Boost變換器的工作原理出發(fā),分析了變換器的不同工作模式。在考慮電感電流實時變化條件下,通過求解系統(tǒng)狀態(tài)方程,建立了雙管BuckBoost變換器在Buck工作模式和Boost工作模式下的預測模型,為控制器的設計提供依據(jù)。結合雙管Buck-Boost變換器和模型預測控制的特點,為了減少開關損耗,在傳統(tǒng)三模式控制方式的基礎上,提出了混合工作模式代替Buck-Boost工作模式。根據(jù)工作模式不同,構造了以輸出電壓和電感電流為優(yōu)化量的代價函數(shù);谀P皖A測控制設計了只考慮開關管開通和關斷兩種情況的0-1簡化算法;為了解決0-1簡化算法電感電流紋波較大的問題,在不增加計算量的情況下設計了占空比估計算法;設計了根據(jù)系統(tǒng)狀態(tài)和代價函數(shù)反解出最有占空比的公式解析算法。在Simulink軟件平臺進行仿真,對比了三種算法的控制效果。設計了雙管Buck-Boost變換器的硬件參數(shù),搭建18V~72V輸入、36V輸出的雙管Buck-Boost變換器實驗平臺,采用DSP實現(xiàn)模型預測控制,對提出的控制算法進行了實驗驗證。實驗結果表明將模型預測控制應用于寬范圍直流變換器控制系統(tǒng)中的方案是可行、有效的。
[Abstract]:In the fields of fuel cell power supply system, new energy power generation system and so on, wide range input DC converter has been widely used. Among them, the double transistor Buck-Boost converter has the same polarity of input and output voltage. The advantages of low voltage stress of switch tube have become a hot research topic. The traditional control method of dual-transistor Buck-Boost converter selects the working mode according to the input voltage range. In this paper, the model predictive control is adopted, and the optimal working mode is selected by online optimization. It provides a scheme for the controller design of wide range input DC / DC converter. Based on the working principle of double-transistor Buck-Boost converter, the different working modes of the converter are analyzed in this paper. Under the condition of considering the real-time variation of inductance current, the prediction model of double-transistor BuckBoost converter in Buck mode and Boost mode is established by solving the state equation of the system, which provides the basis for the design of the controller. Combined with the characteristics of dual-transistor Buck-Boost converter and model predictive control, in order to reduce switching loss, a hybrid working mode is proposed to replace Buck-Boost mode on the basis of traditional three-mode control mode. According to the different working modes, the cost function with output voltage and inductance current as the optimized quantity is constructed. Based on the model predictive control, a simplified algorithm of 0 鈮,
本文編號:2490348
[Abstract]:In the fields of fuel cell power supply system, new energy power generation system and so on, wide range input DC converter has been widely used. Among them, the double transistor Buck-Boost converter has the same polarity of input and output voltage. The advantages of low voltage stress of switch tube have become a hot research topic. The traditional control method of dual-transistor Buck-Boost converter selects the working mode according to the input voltage range. In this paper, the model predictive control is adopted, and the optimal working mode is selected by online optimization. It provides a scheme for the controller design of wide range input DC / DC converter. Based on the working principle of double-transistor Buck-Boost converter, the different working modes of the converter are analyzed in this paper. Under the condition of considering the real-time variation of inductance current, the prediction model of double-transistor BuckBoost converter in Buck mode and Boost mode is established by solving the state equation of the system, which provides the basis for the design of the controller. Combined with the characteristics of dual-transistor Buck-Boost converter and model predictive control, in order to reduce switching loss, a hybrid working mode is proposed to replace Buck-Boost mode on the basis of traditional three-mode control mode. According to the different working modes, the cost function with output voltage and inductance current as the optimized quantity is constructed. Based on the model predictive control, a simplified algorithm of 0 鈮,
本文編號:2490348
本文鏈接:http://sikaile.net/kejilunwen/dianlidianqilunwen/2490348.html
最近更新
教材專著