改进BP神经网络的海参原料品质检测

Improvement of BP Neural Network for Sea Cucumber Raw Material Quality Detection

摘要: 为解决海参原料分选缺乏客观科学评价标准的难题，本文提出基于BP神经网络的品质检测分选方法，可实现其品质快速检测与分选，为后续加工营养保留提供支撑。研究首先通过单轴拉伸试验获取不同品质海参本构曲线，明确品质边界以支撑后续变形有限元分析，搭建有限元模型及本构–载荷–位移数据库，建立BP神经网络本构识别模型。最后结合感官评价明确本构与品质的对应关系，通过试验验证方法准确误差低于3%，证明方法可行。本研究为海参原料科学化分选提供新路径，对海参原料加工产业提质增效具有重要意义。

Abstract: To address the problem of the lack of objective and scientific evaluation criteria for the selection of sea cucumber raw materials, this paper proposes a quality detection and sorting method based on BP neural network, which can achieve rapid quality detection and sorting of sea cucumbers, providing support for the subsequent processing of nutrient retention. The research first obtained the constitutive curves of different quality sea cucumbers through uniaxial tensile tests, clarified the quality boundaries to support the subsequent finite element analysis of deformation, established a finite element model and constitutive-load-displacement database, and established a BP neural network constitutive identification model. Finally, by combining sensory evaluation, the corresponding relationship between the constitutive and quality was clarified, and the method was verified through experiments, with an accuracy error of less than 3% proving the feasibility of the method. This research provides a new path for the scientific sorting of sea cucumber raw materials and is of great significance for improving the quality and efficiency of the sea cucumber raw material processing industry.