**Author details**

192 Simulated Annealing – Single and Multiple Objective Problems

with negligible degradation in the RD performance.

**6. Conclusion** 

ASP

**Figure 12.** Rate-Distortion performance and complexity cost comparison versus frame

This chapter presents a novel fast motion estimation algorithm, Simulated Annealing Adaptive Search algorithm. As mv field has heavy correlation, the proposed algorithm takes the advantage of MV correlation information, which is statistically calculated and plays a significant role in SAAS process. In the SA search step, the search region is adaptively divided and the divisions are searched indexed by MV correlation probabilities in descending order. Furthermore, by utilizing Boltzmann probability concept, the minima acceptation or rejection condition of each SA search is controlled by this correlation

<sup>0</sup> <sup>10</sup> <sup>20</sup> <sup>30</sup> <sup>40</sup> <sup>50</sup> <sup>60</sup> <sup>70</sup> <sup>80</sup> <sup>90</sup> <sup>100</sup> <sup>10</sup>

coastguard.cif

SAAS UMHexagonS

Frame number

Considering the large reduction in the computational complexity, the quality degradation is very small. Rate-Distortion performances are presented in Table 4. In all cases, the FS outperforms the others in image quality and bit-rate. Compared to FS, average PSNR degradation of the proposed algorithm is only 0.010. In term of bit-rate, SAAS has a slightly higher degradation than PIDS and UMHexagonS. SAAS has bit-rate decreasing of 0.64% in average. Further information can be obtained in Figure 9, which compares the ratedistortion performance among FS, UMHexagonS, PIDS and SAAS against different QPs (16, 20, 24, 28, 32, 36 and 40). Figure 10 compares the simulation results versus frame number of video sequences *Coastguard*. It is clearly reveal the superiority of SAAS to UMHexagonS in computational reduction, which more than 50% of search points are saved while the PSNR and bit-rate performance are very similar. From the results above, it can be confirmed that the SAAS algorithm has the capability to dramatically reduce the computational burden

Zhiru Shi, W.A.C. Fernando and A. Kondoz *I-Lab, CVSSP, University of Surrey, Guildford, United Kingdom* 

## **7. References**

	- [14] Sullivan G. J, Wiegand T(1998) Rate-distortion optimization for video compression. IEEE Magazine Signal Processing 15:74-90.

**Section 2** 

**Multiple Objectives** 

**Multiple Objectives** 

194 Simulated Annealing – Single and Multiple Objective Problems

IEEE Magazine Signal Processing 15:74-90.

[14] Sullivan G. J, Wiegand T(1998) Rate-distortion optimization for video compression.

**Chapter 10** 

© 2012 Lobato et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2012 Lobato et al., licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**Design and Identification Problems of Rotor** 

The study of rotating machinery appears in the context of machines and structures due to the significant number of phenomena typical to their operation that impact their dynamic behavior and maintenance. Consequently, rotor bearing systems face numerous problems that affect a wide variety of machines, e.g., compressors, pumps, motors, centrifuge machines, large and small turbines. This type of machine finds various applications in the industry, such as, automotive, aerospace and power generation. In most applications an unpredictable stoppage can lead to considerable financial losses and risks. Therefore, there is an evident need for the complete modelling of rotating systems, including the components of the interface between fixed and moveable parts, such as the hydrodynamic bearings. Bench-scale experimental analyses provide more complete models of the main components of the rotor, with strong emphasis on the modelling of the bearings of rotary

machines, since they constitute the rotor-foundation structure connecting elements.

inverse problem represents an important alternative approach.

The machinery parameters are needed to study the dynamic behavior of the system, namely the Campbell diagram, stability analysis, critical speeds, excitation responses, control and health monitoring. The determination of unknown parameters in rotating machinery is a difficult task. To overcome this difficulty, the use of optimization techniques to solve the

In the literature, various works have been proposed to determine unknown parameters of dynamic systems. Edwards et al. [1] presented a procedure to determine unbalance and support parameters simultaneously based on the least-squares method. Xu et al. [2] proposed a rotor balancing method by using optimization techniques, which does not need

**Bearing Systems Using the Simulated** 

**Annealing Algorithm** 

http://dx.doi.org/10.5772/47833

**1. Introduction** 

Fran Sérgio Lobato, Elaine Gomes Assis,

Additional information is available at the end of the chapter

Valder Steffen Jr and Antônio José da Silva Neto
