| 摘要: |
| 【目的】探讨不同湍流模型对库区水温分布模拟的差异，为大型深水水库水温的准确模拟提供科学依据。【方法】以黄河上游李家峡水库为例，分别选取Smagorinsky、标准k-ε、混合k-ε等3种湍流模型，模拟预测李家峡水库升温期和高温期库区水温的分布，利用实测的水温资料，对比分析3种湍流模型水温模拟的精度，总结3种湍流模型模拟库区水温的分布规律。【结果】Smagorinsky模型、混合k-ε模型模拟的坝前垂向水温与实测水温吻合程度较高，标准k-ε模型的模拟结果与实测水温有一定差异；3种湍流模型模拟的发电引水口水温均高于实测水温，混合k-ε湍流模型模拟的5月水温与实测水温的最大误差为5.2%，Smagorinsky模型、标准k-ε模型模拟该月水温的相应误差为11.1%和8.2%，以混合k-ε湍流模型的误差最小、精度最高；升温期3种模型模拟的纵垂向水温均呈混合状态分布，高温期均呈弱分层状态分布，其中混合k-ε模型模拟水温在水下20 m以内出现较为明显的温跃层，与实测水温更为接近。【结论】针对大型深水水库的水温模拟，宜采用混合k-ε湍流模型，其模拟精度高、水温分布更接近实测。 |
| 关键词: 湍流模型 李家峡水库 水温分布 水温模拟 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金项目“黄河上游龙刘段梯级水库群水温累积效应研究”（51179151） |
|
| Influence of different turbulence models on simulation of reservoir water temperature distribution |
|
TANG Wang,ZHOU Xiao-de,SONG Ce
|
| Abstract: |
| 【Objective】This paper discussed differences in simulation of reservoir water temperature distribution using different turbulence models to provide scientific basis for water temperature simulation of large deep-water reservoir.【Method】Taking Lijiaxia Reservoir in the Yellow River upstream as an example,three turbulence models including the Smagorinsky,the standard k-ε, and the mixed k-ε,were used to simulate and predict reservoir water temperature distribution during the heating period and high-temperature period.Then measured temperature data was used to compare and analyz the water temperature simulation accuracy.The reservoir water temperature distribution using the three turbulence models was also summarized.【Result】Smagorinsky model and the mixed k-ε model based dam front vertical water temperature coincided well with the measured water temperature,while standard k-ε model had certain difference.The water temperature of power diversion port predicted by all three models was higher than the actual measured water temperature with maximum error of 5.2% for mixed k-ε turbulence model,11.1% for Smagorinsky and 8.2% for standard k-ε turbulence model in May.During the heating period,longitudinal vertical water temperature from three turbulence models showed a mixed state distribution.During the high-temperature period,it was in weakly stratified state,and mixed k-ε model had obvious thermocline within 20 m under water,which was closer to the measured water temperature.【Conclusion】The mixed k-ε turbulence model had high accuracy and the predicted distribution of water was closer to the measured data.Thus,it is the optimal model for water temperature simulation of large deep-water reservoir. |
| Key words: turbulence models Lijiaxia reservoir water temperature distribution water temperature simulation |
