计及风量损耗的油浸风冷外置冷却系统散热效率优化方法

Heat Dissipation Efficiency Optimization Method for ONAF External Cooling System Taking into Account Airflow Losses

为助力变压器冷却系统能效提升,“轻量化、小型化”已成为发展趋势。精确配置风扇直径既能确保高效散热,又能避免造价高、质量重及风损大等问题,与“轻”“小”理念相契合。以往常采用计算流体动力学对冷却系统进行试探性建模与改进,以找出冷却效果显著的配置结构,耗时长且优化目标单一。该文针对油浸风冷(ONAF)模式下散热器构建了一种快速迭代寻优的解析模型,获得了出口油温、油流量及风量分布,在满足既定温差且提升散热效率的同时控制风量损耗。此外,建立了流热仿真与试验相结合的平台,将模型结果与试验、仿真进行对比分析。研究结果表明:风扇直径对油温差的影响呈非线性,风量损耗与直径呈正相关。以PC2600-22/520散热器为验证对象,当风速为3.5 m/s、风扇直径是散热片宽度的1.2~1.5倍时,温差可达到期望值,此时散热效率与风量损耗协同进入最优区间。

To help improve the energy efficiency of transformer cooling systems,“lightweight and miniaturization”is a development trend.The precise configuration of the fan diameter ensures efficient heat dissipation and reasonable air volume distribution while avoiding the problems of high cost,heavy mass,and large air loss,which is in line with the“light”and“small”concepts.The main means to study the cooling performance of transformers is often based on computational fluid dynamic(CFD)tentative modeling and improvement,to explore the cooling effect of the configuration structure,but CFD numerical simulation can obtain the required high precision results,but the pre-processing such as equal scale 3D modeling,mesh drawing and multi-physics field simulation consumes a lot of computational resources and time,and the process is complicated and the optimization objective is single.Thus,a fast iterative optimization model is constructed for a radiator with natural oil circulation forced air cooling(ONAF).Firstly,the analytical model includes the momentum analysis of the overall oil circulation,the cooling air intensity analysis based on the local air loss coefficient,and the heat transfer analysis of the internal oil flow and external air of the radiator.Among them,the momentum analysis of oil circulation is the core of radiator temperature rise calculation,and the local loss coefficient of air volume is closely related to wind speed,fan diameter,cooling air distance,and duct perimeter,and directly affects the Nusselt number(Nu),which is the most important dimensionless number reflecting convective heat transfer strength.Matlab is then used to iteratively calculate the flow-heat characteristic parameters of the analytical model to obtain the radiator import and export oil temperature difference,the relationship between air loss and heat dissipation efficiency is finally integrated and controlled to achieve efficient heat dissipation in the cooling system.Secondly,to verify the accuracy of the analytical model,a combined flow-thermal simulation and test platform was also established based on the PC2600-22/520 radiator with an equal scale.The k-εmodel was selected for the simulation,and the physical parameters of mineral oil,air,and heat sink were defined.In the test,three types of fans with diameters of 455 mm,655 mm,and 855 mm were used to record the temperature difference between the inlet and outlet of the radiator,and a hot-wire anemometer was used to measure the air velocity and the temperature between the radiator fin.Finally,the research shows that the fan diameter has a non-linear effect on the oil temperature difference,and the wind loss is positively correlated with the diameter.With the PC2600-22/520 radiator as the verification object,the temperature difference can reach the desired value when the wind speed is 3.5 m/s and the fan diameter is 1.2~1.5 times the width of the radiator fin,at which time the heat dissipation efficiency and the wind loss collaborate to enter the optimal interval.In addition,the average relative error between the model calculated cooling air flow rate of the adjacent air duct at the center of the radiator fin and the test and simulation results is less than 6%,and the average relative error of the outlet oil temperature is less than 2%,which saves more than 98.56%of time cost.The work of this paper provides a new idea for the lightweight design of radiator structure under forced air cooling,cooling efficiency improvement and cooling air intensity distribution calculation.