论文部分内容阅读
针对小麦加工HACCP(Hazard Analysis and Critical Control Point)关键控制点的选择问题,采用一种基于支持向量机模型来实现关键控制点智能发现的方法。为了提高小麦加工关键控制点发现试验的识别稳定性和精确性,提出一种自适应动态搜索粒子群算法优化SVM模型的核函数参数方法,该算法引入进化因子和进化阈值估计进化状态,动态调整搜索策略。当进化因子大于进化阈值时,采用基本粒子群搜索策略;反之,采用反向搜索策略,以扩大种群的多样性。算法基于进化因子为速度定义了速度惯性参数。仿真结果表明,基于ADS-PSO的SVM模型能够很好的实现关键控制点的智能发现,并取得了较高的识别率和稳定性。
Aiming at the selection of key control points of Hazard Analysis and Critical Control Point (HACCP) in wheat processing, a method based on support vector machine (SVM) model was proposed to realize the key control point intelligent discovery. In order to improve the identification stability and accuracy of the key control point discovery test of wheat processing, an adaptive dynamic search particle swarm optimization algorithm is proposed to optimize the kernel function parameters of the SVM model. The algorithm introduces the evolutionary factor and evolutionary threshold to estimate the evolution state and dynamically adjust Search strategy When the evolutionary factor is greater than the evolutionary threshold, the basic PSO strategy is adopted; on the contrary, reverse search strategy is used to expand the diversity of the population. The algorithm defines the speed inertia parameter for speed based on evolutionary factor. Simulation results show that SVM model based on ADS-PSO can realize the intelligent detection of key control points well and achieve high recognition rate and stability.