With respect to input data having the spread of errors or the like, “a suiting method” is to obtain a suiting function, which exist within a region of this spread, and which is highly robust. A suiting region (between an upper bound function M+(xi) and a lower bound function M−(xi)) is set based on the input data with the spread of errors. An inequality system is established with respect to the suiting region and the suiting function having undetermined parameters. A distance from a boundary of a solution region of the parameter values, corresponding to a boundary of the suiting region, is set, and a suiting function f(x) which has the parameter values maximizing the distance, is found. A broken line exhibits a suiting function f′(x) having low robustness, and a solid line exhibits a most-robust suiting function f(x) obtained from an application of this invention.
展开▼
机译:关于具有误差等的散布的输入数据,“一种适合方法”为“适合方法”。为了获得合适的函数,该函数存在于该扩展的区域内,并且非常健壮。适合区域(在上限函数M + Sup>(x i Sub>)和下限函数M − Sup>(x 之间i Sub>))是根据输入数据进行设置的,其中包含误差分布。对于具有不确定参数的拟合区域和拟合函数,建立了不等式系统。设置距参数值的求解区域的边界与拟合区域的边界相对应的距离,并且找到具有使距离最大化的参数值的拟合函数f(x)。虚线显示出具有低鲁棒性的拟合函数f(x),实线显示出从本发明的应用获得的最鲁棒的拟合函数f(x)。
展开▼