CAJ | 학술논문

在处理矿井通风机实测数据时，传统的拟合算法很难得到最佳的拟合曲线，曲线精度不高，难以反映实际情况。针对这一问题，提出了最小一乘法求解矿井通风机性能曲线，以零偏差点个数、零偏差点比率、总偏差、平均偏差和偏差率作为评估拟合通风机性能曲线的精度指标。通过对现场实测通风机数据进行三次多项式拟合，结果表明：最小一乘法拟合的曲线方程出现了多个零偏差点，减少了异常点对曲线方程的影响，具有更高的精度。
재처리광정통풍궤실측수거시，전통적의합산법흔난득도최가적의합곡선，곡선정도불고，난이반영실제정황。침대저일문제，제출료최소일승법구해광정통풍궤성능곡선，이령편차점개수、령편차점비솔、총편차、평균편차화편차솔작위평고의합통풍궤성능곡선적정도지표。통과대현장실측통풍궤수거진행삼차다항식의합，결과표명：최소일승법의합적곡선방정출현료다개령편차점，감소료이상점대곡선방정적영향，구유경고적정도。
The conventional fitting algorithm is hard to get the best fitting curve when processing test data of mine ventilator , and the precision is low , so it is difficult to reflect the actual situation .Aiming at this problem , the method of applying least absolute deviation to solve performance curves of mine ventilator was proposed , and some precision indexes , such as zero bias deviation number , zero deviation point ratio , total deviation , average deviation and deviation rate , were selected to evaluate the fitting performance curves of ventilator .Through on-site ventilation measurement data , a cubic polynomial fitting was conducted .The result showed that the curve equation fitted by least squares fitting appears many zero bias point .It can reduce the influence of curvilinear equation result from ab-normal point , and has a higher accuracy .