基于GA-LSSVM的多信息融合算法

An algorithm of multiple information fusion based on GA-LSSVM

摘要:

由于在多传感器测控系统中被测系数与相关参数之间存在着非线性关系，提出了一种基于遗传和最小二乘支持向量机(GA-LSSVM)的多信息融合模型及算法，借助其优越的全局最优搜索能力进行参数的优化。这种方法为小样本、非线性、高维数的多传感器信息融合问题的建模提供了一种有效途径。通过对一个简单的低压负荷电路系统进行算例分析表明，基于最小二乘支持向量机的多传感器信息融合模型及算法在测量准确度和推广性能上都具有一定的优越性。

Abstract:

In multiple sensor measurement and control system，there exists nonlinearity relationship between the tested coefficients and relevant parameters. A kind of model and algorithm of multiple sensor information fusion based on Genetic Algorithm and Least Squares Support Vector Machine(GA-LSSVM) are proposed in this paper. For improving accuracy of the model，genetic algorithm is recommended for parameter optimization because of its excellent global optimization ability. It is an effective way for modeling multiple sensor system of small sample, nonlinearity and high dimension. To verify the effectiveness of this method，a simple low-voltage load circuit system is adopted as an example，and the result shows that the model and algorithm have certain superiority in measuring precision and are easy to be promoted.