Development of a DSP-FPGA-based resolver-to-digital converter for stabilized gun platforms

---

Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Elektrik ve Elektronik Mühendisliği Bölümü, Türkiye

Tezin Onay Tarihi: 2010

Öğrenci: YASİN ZENGİN

Danışman: İSMET ERKMEN

Özet:

Resolver, due to its reliability and durability, has been used for the aim of shaft position sensing of military rotary systems such as tank turrets and gun stabilization platforms for decades. Ready-to-use resolver-to-digital converter integrated circuits which convert the resolver signals into position and speed measurements are utilized in servo systems most commonly. However, the ready-to-use integrated circuits increase the dependency of the servo system to hardware components which in turn decrease the efficiency and flexibility of the servo system for changing system structures such as for changing resolver carrier frequency or changing position and speed sensors. The proposed solution to increase the efficiency and flexibility of the servo system is a software-based resolver-to-digital converter which does not require aforesaid special hardware components and presents a complete software-based solution for the conversion. The proposed software-based resolver-to-digital converter makes use of common programmable hardware v components, that is, FPGA and DSP which form the heart of the servo controller technology in recent years. The proposed structure for the conversion has three components. The first component is the signal conditioner which minimizes the disturbances coming from the resolver signals as harmonic distortions and noise. The second component, the phase-sensitive demodulator, as the name implies, is responsible for phase-sensitive demodulation of resolver signals. The third component is the estimator filter. In order to determine the optimal estimator filter, five different estimator filters with the aforesaid two components are implemented in ASELSAN’s stabilized gun system STAMP and they are compared in terms of both estimation performance and computational complexity. The implemented filters include nonlinear observer type filter which is already proposed in the literature for resolver conversion, tracking differentiator adapted to resolver conversion and kalman filters adapted to resolver conversion in different forms such as linear kalman filter, extended kalman filter and unscented kalman filter. At the end of the study, stability and sensitivity analyses are also performed for the proposed system.