應用量子基因演算法求解最佳化短期火力機組排程

Application of Quantum Genetic Algorithm for Optimal Short Term Thermal Generation Unit Commitment

劉律伸(Lu-Shen Liu)

指導教授 ： 曹大鵬

國立臺北科技大學/電資學院/電機工程系研究所/碩士(2012年)

https://doi.org/10.6841/NTUT.2012.00066

摘要

機組排程是一個在排程的時間內以最經濟方式來排定機組狀態與決定機組發電量的問題，其目的在於使總發電成本最小並同時滿足負載需求、備轉容量需求與一些個別限制條件；求解機組排程必須迅速而有效，才能做為發電策略的參考依據，因此如何選擇一個有效且穩定的求解方法來求解機組排程問題已成為一種重要的課題。本論文在機組排程上採用一種較新穎的量子基因演算法，此演算法是融合量子演算法和基因演算法所實現的一種新的演算法，此演算法採用量子機率向量的編碼方式，同時使用量子位元、量子疊加狀態的思想，而量子疊加狀態的特性能使排列更多元化，而由機率表達的特性是將解的狀態以一定的機率表達出來，藉此有效提高最佳解搜索能力。本論文同時以線性規劃法為基礎來建立一個求解最小線損的最佳化電力潮流；最佳化電力潮流主要是在滿足電力系統運轉條件下，處理如何達成其目標函數最佳化的問題，諸如發電成本、最小線損、最大傳輸功率、或是系統最佳安全運轉等問題，都可以藉由改變目標函數來求解。最後以本論文所提出的量子基因演算法結合最佳化電力潮流對三個案例做分析，分別以IEEE 14-Bus、IEEE 30-Bus、IEEE 57-Bus的電力系統做24小時負載的線路損失最佳化機組排程，並和忽略限制的動態規劃法，基因演算法做比較，模擬結果印證量子基因演算法可以達到較低的發電成本同時兼顧輸電線損失，因而適合將此方法使用在求解最佳火力機組排程。

關鍵字

機組排程 ； 最佳化電力潮流 ； 量子基因演算法 ； 線性規劃

並列摘要

The purpose of unit commitment for a specified power system is to decide the generator status, electrical total power outputs to satisfy load demands, spinning reserves and constraints in the most economic way. To be a good reference for a power generation policy, the process of unit commitments must be fast and effective. How to find out an effective and stable solution method for the best of unit commitments is a major course for modern power system. This thesis combines Quantum Algorithm and Genetic Algorithm to present a new algorithm, called Quantum Genetic Algorithm (QGA). The Quantum Genetic Algorithm uses the coding method of quantum probability vector, and also use the quantum bit and quantum superposition at the same time. The characteristic of quantum superposition can make code express more flexible. The probability expression characteristic can be expresses the solution state by certain probability. It can raise the ability of optimal solution. In this thesis, the optimal power flow with linear programming is also carried out to obtain minimum power line losses. Under certain power system operating conditions, the optimal power flow study is to obtain the optimal objective functions such as the cost of power generation, the minimum line losses, the maximum transmission power and the best security operating conditions. In this thesis, three research cases including IEEE 14-bus, IEEE 30-bus and IEEE 57-bus have been studied and analyzed for a short-term unit commitments. These study results have been comparison with other methods i.e. Dynamic Programming and Genetic Algorithm. It is indicated that Quantum Genetic Algorithm which proposed in this thesis is more useful and efficient in short-term unit commitments.

並列關鍵字

Unit Commitment ； Optimal Power Flow ； Quantum Generation Algorithm ； Linear Programming

參考文獻

[2] P. G. Lowery, “Generation unit commitment by dynamic programming,” IEEE Transactions on Power Apparatus and system, Vol. 85, No. 5, 1996, pp. 422-426.

[5] H. Sasaki, M. Watanabe, and R. Yokoyama, “A Solution Method of Unit Commitment by Artificial Neural Networks,” IEEE Trans. on Power Systems, Vol. 7, No. 3, 1992. pp. 974-981.

[6] T. Saksornchai, Wei-Jen Lee, K. Methaprayoon, J. R. Liao and R. J. Ross, “Improve the unit commitment scheduling by using the neural-network-based short-term load forecasting,” IEEE Transactions on Industry Applications, Vol. 41, No. 1, 2005, pp. 169-179.

[7] F. Zhuang and F. D. Galiana, “Unit Commitment by Simulated Annealing,” IEEE Tran on Power Systems, PWRS-5, 1990, pp. 311-317.

[9] G. Xiao, S. Li, X. Wang and R. Xiao. ”A solution to unit commitment problem by ACO and PSO hybrid algorithm,” Proceedings of the 6th World Congress on Intelligent Control and Automation, Dalian, China, 2006, pp. 7475-7479.