|
| 基于GA-BP神经网络的尾煤水灰分视觉检测方法研究 |
| Research on Visual Detection Method of Ash Content in Tailing Coal Water Based on GA‐BP Neural Network |
| |
| DOI：10.16018/j.cnki.cn32-1650/n.202303005 |
| 中文关键词: 煤泥灰分 图像处理 彩色特征 遗传算法 BP神经网络 |
| 英文关键词: coal slurry ash content image processing color feature genetic algorithm BP neural network |
| 基金项目: |
|
| 摘要点击次数: 91 |
| 全文下载次数: 93 |
| 中文摘要: |
| 针对浮选回归模型精度和适应性较差的问题，提出了一种基于遗传算法优化反向传播神经网络(GA-BP)的尾煤水灰分视觉检测方法。对尾煤水图像进行预处理，在去除主要噪声干扰和保证一定彩色特征完整的前提下，提取不同颜色空间的彩色特征、灰度特征以及浓度特征值；以上述特征值为输入变量，以尾煤水灰分作为输出变量，建立基于遗传算法优化BP神经网络(GA-BP)的回归模型。该模型较好地实现了尾煤水灰分的在线检测，预测精度达97. 3%，均方误差降低至0. 23，提高了精煤产率和经济效益。 |
| 英文摘要: |
| Aiming at the poor accuracy and adaptability of flotation regression model, a visual detection method of ash content in tailings coal water based on GA-BP optimized by genetic algorithm is proposed. Firstly, the tail coal water image is preprocessed. On the premise of removing the main noise interference and ensuring the integrity of certain color features, the color features, gray features and concentration feature values of different color spaces are extracted. With the above eigenvalues as input variables and tailing ash content as output variables, a regression model based on genetic algorithm optimization of BP neural network (GA-BP) is established. The model can well realize the on-line detection of ash content in tail coal water, the prediction accuracy reaches 97. 3%, and the mean square error is reduced to 0. 23, which improves the clean coal yield and economic benefits. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |