CAJ | 학술논문

为利用最小二乘支持向量机（LS-SVM）来进行预测，首先要确定影响LS-SVM模型的两个主要参数γ和σ，针对该问题提出了采用混沌搜索算法来搜索该模型的最优参数组合。混沌搜索的运动轨迹具有遍历性，随机性，可以进行全局和局部寻优，利用该算法搜索最优参数来确定预测模型，然后将该预测模型用于预测实践。实验结果表明，该模型具有较精确的预测精度和适用性。
위이용최소이승지지향량궤（LS-SVM）래진행예측，수선요학정영향LS-SVM모형적량개주요삼수γ화σ，침대해문제제출료채용혼돈수색산법래수색해모형적최우삼수조합。혼돈수색적운동궤적구유편력성，수궤성，가이진행전국화국부심우，이용해산법수색최우삼수래학정예측모형，연후장해예측모형용우예측실천。실험결과표명，해모형구유교정학적예측정도화괄용성。
For using the least squares support vector machines (LS-SVM) to predict, the first thing to do is to determine the two main parameters which will influence the performance of the LS-SVM model. This paper proposes using the chaotic search algorithm to search the optimal parameters of the model. The characteristics of the chaotic motions such of ergodicity and randomness make it can do global and local optimization. Use this algorithm to search the optimal parameters for the prediction model, and then put the prediction model into the use of forecasting practice. The experimental results show that the proposed algorithm achieves good performance and precise predicted results.