CAJ | 학술논문

根据实测冰情数据分析发现，可将冰盖厚度演变过程作为预测头道拐站的开河日期的主要依据，同时还应考虑封冻期气温、流量等对冰盖厚度的持续性和累积性影响。据此提出了一种应用数据挖掘技术和LSSVM 进行头道拐站开河日期预测的新方法。应用LSSVM 模型对头道拐站2010年、2011年和2012年开河日期的预测结果表明，可在封冻期内任一冰盖厚度测量日期利用上述方法对该站的开河日期进行预测，有效延长了预见期，且在3月6日前的预测值均满足许可误差合格率的要求。根据 LSSVM 模型预测误差呈波动性变化的特点，提出了预测开河日期的均值法，可使开河日期预测精度得到显著提高。
근거실측빙정수거분석발현，가장빙개후도연변과정작위예측두도괴참적개하일기적주요의거，동시환응고필봉동기기온、류량등대빙개후도적지속성화루적성영향。거차제출료일충응용수거알굴기술화LSSVM 진행두도괴참개하일기예측적신방법。응용LSSVM 모형대두도괴참2010년、2011년화2012년개하일기적예측결과표명，가재봉동기내임일빙개후도측량일기이용상술방법대해참적개하일기진행예측，유효연장료예견기，차재3월6일전적예측치균만족허가오차합격솔적요구。근거 LSSVM 모형예측오차정파동성변화적특점，제출료예측개하일기적균치법，가사개하일기예측정도득도현저제고。
According to the analysis of measured ice data, the paper pointed outthatthe evolution of the ice sheetthickness can be used as the main basis for forecasting the break-up date atthe TouDaoguai station along the YellowRiver, and the persistentand cumulative effects of temperature and flowduring the freezing period are also importantfactors which affectthe timing of break-up date. A newmethod to forecastthe break-up date using the data mining technology( DM) and leastsquare supportma-chine( LSSVM) was presented in this paper. The LSSVM model was used to predictthe break-up dates atthe TouDaoguai sta-tion in 2010, 2011, and 2012, and the results showed thatthe method can predictthe break-up date on any date of ice sheetthickness observation during the frozen period, which can expand the forecastperiod effectively. Furthermore, the predicted val-ues of break-up date before March 6th metthe requirements of the al owable prediction error rate. An average value method of forecasting the break-up date was proposed according to the characteristics of volatility change in the prediction error produced by the LSSVM model, which can improve the prediction accuracy significantly.