- 學位論文
考慮阻塞效應與尾流效應之風場風機排列最佳化研究
Wind Farm Layout Optimization Research Accounting for Blockage Effect and Wake Effect
摘要
一風力發電場(簡稱「風場」)通常布設許多風力發電機(簡稱「風機」)，如 風機間距離太近，其受到彼此影響的程度必然增加，包含「尾流效應」導致下游 風機的迎風速度大幅下降及「阻塞效應」使得上游風機迎風速小於原始自由風速， 降低整個風場的年發電總量(Annual Energy Production, AEP)。針對一具歷史 風速資料的特定風場，本研究同時考慮尾流效應與阻塞效應，在經選定型號的風 機數量以及風機間最小間距的限制條件下，使用粒子群演算法(Particle Swarm Optimization, PSO)尋找風機在風場內的最佳排列方式，並據以估算風場年發電 總量。在風場內風機總數已固定、且各風機位置已經初步設計好的考量下，本研 究特別探討兩種情境，一為從場址內發電量低落的風機逐次更動其位置，尋求獲 致風場最大年發電量各風機之最佳布設位置;但礙於有限的運算資源，假設能更 動位置的風機不超過六支;第二種情境亦假設位置能被更動的風機為六支，但探 討兩風機最小間距之限制條件如何影響風機之最佳排列與風場之年發電量。綜合 前述兩種情境的案例分析與討論，以最佳化方式排列的風機位置，皆可使得風機 之間的綜合干擾效應降至最低，並最大化風場之年發電量。在第一種情境中，風 場內被更動的風機數量與風場年發電量之增長幅度呈現近似線性的正相關，可據 此估算若將場址內全數風機做更動時，可獲得 20%年發電量改善程度;此外，根據第二種情境的結果，當風機之間的最小間距越低時，風場年發電量的改善幅度 越高，觀察可發現原因為當風機彼此最小間距越小時，可以使得風機盡可能面對 盛行風向一字排開，以降低阻塞效應與尾流效應之影響，當前環評法規所規定之 最小間距為 500 公尺，此情境之結果也提供了更多的討論空間。
並列摘要
There are usually a quite amount of wind turbines installed in a wind farm. If the distance between two turbines is too close, the power generation efficiency of the turbines will decrease. It is owing to both the ”wake effect” and the ”blockage effect.” The former will cause the wind speed of the downstream turbine to drop significantly, and the latter will cause the wind speed of the upstream turbine to be less than the original free-wind speed. The total annual energy production (AEP) of the wind farm will decrease accordingly. For a specific wind farm having designated area and historical wind data in Taiwan, by assuming the number of wind turbines is determined already and the original locations of turbines are designed preliminary, this study adopts the particle swarm optimization (PSO) algorithm to find the optimal geographical layout of the turbines. The objective function is the AEP of the wind farm and the design variable are locations of the turbines which, in turn, are influenced by the annual wind data, the wake effect, and the blockage effect. The minimum distance between any two turbines is considered an additional constraint. Owing to the shortage of computation resource, only a limited amount of wind turbines are allowed to change their preliminarily designed locations in our illustrated two cases. In both cases, the number of turbines allowed to be re-allocated is limited to six, but the influence of minimum distance between two turbines is examined additionally in the second case. The result of the first case indicates there is an approximate linear relation between the increment of AEP of the wind farm and the re-allocated number of wind turbines; and there is a 20% increase of AEP if all turbines can be re-allocated based on our computational experience. The result of the second case shows the AEP of the wind farm can be increased significantly if the minimum distance between two turbines is decreased. It violates the current rule of 500-m minimum distance based on primarily the environmental consideration though. The increase of AEP in both cases is attributed to the reduction of wake effect and blockage effect.