基于起源信息的水位衰老数据的筛选算法

为提高数据质量和决策依据的权威性,需要筛选和处理因过时而无法准确描述事物、从而影响决策的衰老数据。基于最小二乘法多项式曲线模拟数据序列,结合水位数据起源信息特征,加入时间影响因子,提出了一种带权值的多项式筛选算法,并对其进行了数学推导。为了对该算法进行正确性验证和效用分析,采用滁河流域水库水位数据进行了相关试验,并与最小二乘法多项式筛选方法进行对比分析。试验得出,该算法较最小二乘法多项式筛选方法,误筛、漏筛率降低约40%。衰老数据处理之后,系统综合处理时间减少约20%,而分析结果的准确率提升约57%。结果表明:该筛选算法能有效筛选衰老数据,提高数据分析结果的准确性和效率,从而辅助提高决策能力。

Provenance information-based screening algorithm for aged data of water level

In order to improve the quality of data and the authority of decision-making，it is necessary to screen and process the aged data that cannot accurately describe things and then affect decisions. Based on the least squares polynomial curve fitting data series and combined with the provenance information of water level data，a polynomial screening algorithm with weights is proposed herein by adding time influence factor，which is，moreover，mathematically deduced as well. A relevant experiment is made with the data of the water level from the reservoirs with Chuhe River Bsin，from which the result is compared with that from the least square polynomial screening method，so as to make the correctness verification and the availability analysis for this algorithm. The experiment indicates that the rates of mistaken screening and missed screening are lowered by about 40% ，if compared with those from the least square polynomial screening method. After processing the aged data，the overall processing time of the system is reduced by about 20% ，while the precision of the analysis result is increased by about 57% . The result shows that this screening algorithm can effectively screen the aged data and enhance the precision and efficiency of the data analysis result， thus assist to enhance the decision-making capacity concerned.