CAJ | 학술논문

提出一种基于自适应混沌粒子群优化和支持向量机结合的非线性预测建模算法（ACPSO—SVR），引入ACPSO启发式寻优机制对SVR模型的超参数进行自动选取，在超参数取值范围变化较大的情况下，效果明显优于网格式搜索算法。选取UCI机器学习数据库中的Forestfires标准数据集进行测试，实验结果表明该方法具有较高的精度和良好的泛化能力．对于解决多变量的回归预测问题是一种有效的方法。最后给出了混合算法在碳一多相催化领域的两种典型应用．在反应动力学模型未知的情况下建立催化剂组份模型和操作条件模型，以及基于混合算法的最优催化剂设计框架。
제출일충기우자괄응혼돈입자군우화화지지향량궤결합적비선성예측건모산법（ACPSO—SVR），인입ACPSO계발식심우궤제대SVR모형적초삼수진행자동선취，재초삼수취치범위변화교대적정황하，효과명현우우망격식수색산법。선취UCI궤기학습수거고중적Forestfires표준수거집진행측시，실험결과표명해방법구유교고적정도화량호적범화능력．대우해결다변량적회귀예측문제시일충유효적방법。최후급출료혼합산법재탄일다상최화영역적량충전형응용．재반응동역학모형미지적정황하건립최화제조빈모형화조작조건모형，이급기우혼합산법적최우최화제설계광가。
An effective relevance prediction algorithm is presented for nonlinear system forecasting and modeling based on adaptive chaos particle swarm optimization and support vector regression, namely ACPSO-SVR method. A heuristic optimization method ACPSO was introduced to automatic selection of hyper-parameters in SVR. The forest fires standard data set of UCI machine learning database was selected to test. The experimental results showed that the new method has high relatively precision and good generalization ability with a wide range of parameter values, better than that of mesh searching algorithm. It could be used as an effective method to solve the problems of multivariate regression predication. Moreover, two main applications were introduced in C1 heterogeneous catalysts yield, obtaining the catalyst composition model and the kinetic model, and building the optimization framework of catalyst development in laboratory.