风场实测数据缺失的插补方法研究

Research on interpolation method for deficient wind field data

现场实测是风工程的研究方法之一,可以获得比较全面、可靠的风速、风向实测数据。实测数据是非常宝贵的,并且是进一步研究的基础。但在实测时如出现天气恶劣、仪器故障等因素,就可能使得风场实测数据的部分时段出现一些缺失和异常,影响实测数据的完整性和合理性,及其在实践中的应用。为了充分、有效地利用实测数据,就需要对风场时程数据中出现缺失和异常的部分时段进行预测和插补。为了保证实测数据的合理、完整及实用性,提出了一种基于时间序列模型,充分利用缺失段前、后的有效实测数据,综合采用顺插和逆插方式插补出缺失段的风场数据的方法。通过具体算例对所提方法与以往的方法进行了比较,发现所提方法更加有效、准确。

Field measurement is an important research method of wind engineering.It can get reliable,comprehensive wind speed and wind direction data,which are very valuable,and are the basis for further research.However,parts of abnormal or missing wind speed or pressure data may appear due to bad weathers,instrument failure and other factors,which affect the integrity,rationality and application of the measured data.In order to fully use the measured data,it is necessary to predict and interpolate the abnormal and missing data.To ensure the rationality,completeness and practicality of the measured data,this article proposed a method based on time series model to make full use of the effective data before and after the missing segment.It combined the direct and inverse interpolation method to interpolate the missing segment.Comparison between proposed method and existing methods by specific examples showed that the proposed method is more efficient and accurate than existing ones.