Design and implementation of predictive current controllers for permanent magnet synchronous motors in air-conditioning systems

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

MOST105-2623-E019-002-ET

Translated Name/計畫中文名

空調系統中永磁同步電動機的預測電流控制器之設計與實現

Project Coordinator/計畫主持人

Cheng-Kai Lin

Funding Organization/主管機關

National Science and Technology Council

Department/Unit

Department of Electrical Engineering

Website

https://www.grb.gov.tw/search/planDetail?id=11721684

Year

2016

Start date/計畫起

01-01-2016

Expected Completion/計畫迄

31-12-2016

Bugetid/研究經費

951千元

ResearchField/研究領域

電子電機工程

臺灣夏季氣候炎熱難耐，這將大量地增加冷氣空調的使用率，使得尖峰用電量屢創新高。為了達到節能減碳的目的，脈波寬度調變技術已被廣泛地應用至空調系統中。然而，此調變技術已經被發明了至少20年以上。因此，開發新的電流控制技術是值得研究的，以提供永磁同步電動機空調系統其他可行的選擇方案。上述的可能性激發了我們去提出這個研究計畫。本專題研究申請案為一年期計畫(2016/1/1~2016/12/31)。本研究計畫擬於上半年(2016/1/1~ 2016/6/30)將模型式預測電流控制器實現在永磁同步電動機空調系統上。所提用的模型式預測電流控制器是以永磁同步電動機的離散時間數學模型為基礎，需使用電動機的定子電感參數、定子電阻參數以及所測得的電動機的線對線電壓和線對線電流，透過結合反電動勢的估測，來達到預測電流控制的目的。而在下半年(2016/7/1~2016/12/31)中，我們擬將無模型式預測電流控制器實現在永磁同步電動機空調系統上，所提用的無模型式預測電流控制器是以電流感測器為基礎。在每次取樣週期內，定子電流將會被偵測兩次。接著，使用減法運算，在不同開關切換模式下的電流變化量便可被計算和記錄。根據偵測的電流值和之前的電流變化量，便可以預測出在下一次取樣週期內，所有可能的開關切換模式下所對應的未來定子電流值。類似於模型式預測電流控制器所使用的規則，在目前取樣週期內選擇出具有最小成本函數的開關切換狀態，該切換狀態將在下一次取樣週期內被採用。如此，即可達成控制電流的目的而不需要採用任何的脈波寬度調變技術。無模型式預測電流控制器的主要特色，包括:計算簡單，不需要加裝電壓感測器，不需要使用空調系統的離散時間數學模型，不需要知道空調系統的任何參數。相較於模型式預測電流控制器，所提無模型式預測電流控制器更適合用於空調系統中，即使空調系統的參數隨溫度和壓力等外在環境因素而改變，仍不會影響電流預測的準確性。最後，驗證時將採用TMS320F28335為核心的數位訊號處理器搭配業界生產的變頻器和永磁同步電動機，實現這二種電流控制器並做相關的比較實驗以驗證其可行性與正確性。所得的實驗結果將可能為目前現有的空調系統的開關切換策略帶來新的視角，即不需要使用任何的脈波寬度調變技術仍可達到節能省電的目的，並有機會進一步將此技術移轉至工業界，以提升國內空調系統的研究水平。"Taiwan sweats over extreme hot weather in summer, so that the utilization rates of air conditioners are massively increased. It breaks the record of peak electricity consumption again and again. In the past 2 decades, in order to save energy and reduce carbon footprint, the pulse width modulation (PWM) technology has been widely applied to the air-conditioning systems. Therefore, it is valuable to develop new current control techniques, which would be another feasible options for the permanent magnet synchronous motor (PMSM) air-conditioning systems. The above possibilities inspired us to propose this research project. This research proposal will plan to last a year (2016/1/1~2016/12/31). During the first half year (2016/1/1~2016/6/30), we will design a model-based predictive current controller (MBPCC) for the PMSM air-conditioning system. Based on the discrete-time mathematical model of the PMSM, the MBPCC will achieve the purpose of predictive current control by these parameters and signals: the estimated back-EMF, the stator resistance, and the stator inductance, as well as the sampled line-to-line voltages and currents. During the second half year (2016/7/1~2016/12/31), we will design a model-free predictive current controller (MFPCC) for the PMSM air-conditioning system. The proposed MFPCC is based on the current sensors. First, the stator current will be detected twice in each sampling period. Next, by using the subtraction method, the current variations under the different switching states can be calculated and recorded. Then, according to the sampled stator current and the previous current variations, the future stator currents under all the possible switching states in the next sampling period can be predicted. By following the similar rule used in the MBPCC, the switching state that minimizes a cost function will be selected in the present sampling period and applied in the next sampling period. Therefore, any PWM technology is no longer required, but the purpose of current control can still be achieved by using the MFPCC. The main advantages of the MFPCC are that the calculation is simple, and voltage sensor is not required. Besides, any discrete-time mathematical model of the PMSM air-conditioning system is not used in the MFPCC, and its parameters are not needed to be known. Compared to the MBPCC, the proposed MFPCC is more suitable for the air-conditioning systems. Moreover, even if their parameters are changed according to the extrinsic environmental factors such as temperature and pressure, the accuracy of current prediction would not be much affected because of the natural benefits of the model-free approach. Finally, to verify their feasibility and functionality, we will do some comparative experiments by using a digital signal processor (TMS320F28335), a PMSM, and an industrial inverter to realize the MBPCC and the MFPCC. The obtained results will bring a new perspective of the switching strategies for the air-conditioning systems, achieving the purpose of energy conservation without using any PWM technology. The MFPCC is hopeful to further transfer to industry, enhancing the research level of the air-conditioning systems in Taiwan."

Keyword(s)

永磁同步電動機
空調系統
數位訊號處理器
預測電流控制
permanent magnet synchronous motor
air-conditioning system
digital signal processor
predictive current control

DSpace CRIS

Design and implementation of predictive current controllers for permanent magnet synchronous motors in air-conditioning systems

Details

Description