基于PCA-PSO-SVM的球磨机负荷预测研究

doi:10.16180/j.cnki.issn1007-7820.2022.01.005

摘要/Abstract

摘要：

球磨机是磨矿生产中的主要设备,但其运行时的内部负荷无法被直接检测,因此难以对负荷进行实时有效地控制,导致磨矿效率受到影响。针对此问题,文中通过磨矿实验采集球磨机磨音信号,对信号进行Welch功率谱分析,研究了磨音频谱信息与球磨机负荷之间的关系。利用PCA对功率谱进行特征提取,为球磨机负荷预测提供外部特征信息。然后,采用PSO对SVM相关参数进行寻优并建立PCA-PSO-SVM球磨机负荷预测模型。研究结果表明,该球磨机预测模型的预测均方根误差为1.144 3,平均绝对误差为0.912 5,平均百分比误差为2.797 9%,证明了该模型对球磨机负荷预测的有效性和稳定性。

关键词: 球磨机, 磨音信号, 信号处理, Welch功率谱分析, 主元分析法, 粒子群算法, 支持向量机, 负荷预测

Abstract:

As the main equipment in the grinding production, the internal load of the ball mill cannot be directly detected during operation. Therefore, it is difficult to effectively control the load in real time, which leads to the reduction of grinding efficiency. To solve this problem, the grinding sound signal of the ball mill is collected through the grinding experiment, and the signal is analyzed by the Welch power spectrum, and the relationship between the grinding sound spectrum information and the load of the ball mill is investigated in the proposed study. PCA is used to extract the features of the power spectrum to provide external feature information for load prediction of the ball mill. PSO is adopted to optimize the relevant parameters of SVM and the load prediction model of PA-PSO-SVM ball mill is established. Research shows that the prediction root mean square error of the ball mill prediction model is 1.144 3, the average absolute error is 0.912 5, and the average percentage error is 2.797 9%, which proves the effectiveness and stability of the model for predicting the load of the ball mill.

Key words: ball mill, grinding sound signal, signal processing, Welch power spectrum analysis, PCA, PSO, SVM, load forecasting

中图分类号:

TP312

冯先丁,魏镜弢,吴张永,钱杰,浦友尚. 基于PCA-PSO-SVM的球磨机负荷预测研究[J]. 电子科技, 2022, 35(1): 29-34.

FENG Xianding,WEI Jingtao,WU Zhangyong,QIAN Jie,PU Youshang. Research on Load Forecast of Ball Mill Based on PCA-PSO-SVM[J]. Electronic Science and Technology, 2022, 35(1): 29-34.

图/表 7

表1

图1

图2

图3

图4

图5

表2

参考文献 19

[1]	石立, 张国旺, 肖骁. 金属矿山选矿厂磨矿分级自动控制研究现状[J]. 有色金属(选矿部分), 2013(S1):44-49.
	Shi Li, Zhang Guowang, Xiao Xiao. Research status of automatic control of grinding and classification in metal mine concentrato[J]. Nonferrous Metals(Mineral Processing Section), 2013(S1):44-49.
[2]	钱杰, 魏镜弢, 王家涛. 球磨机给矿量的智能控制[J]. 化工自动化及仪表, 2019, 46(7):523-527.
	Qian Jie, Wei Jingtao, Wang Jiatao. Intelligent control of ball mill feeding[J]. Control and Instruments in Chemical Industry, 2019, 46(7):523-527.
[3]	汤健, 郑秀萍, 赵立杰, 等. 基于频域特征提取与信息融合的磨机负荷软测量[J]. 仪器仪表学报, 2010, 31(10):2161-2167.
	Tang Jian, Zheng Xiuping, Zhao Lijie, et al. Soft sensing of mill load based on frequency domain feature extraction and information fusion[J]. Chinese Journal of Scientific Instrument, 2010, 31(10):2161-2167.
[4]	罗小燕, 卢小江, 熊洋, 等. 小波分析球磨机轴承振动信号特征提取方法[J]. 噪声与振动控制, 2016, 36(1):148-152.
	Luo Xiaoyan, Lu Xiaojiang, Xiong Yang, et al. Feature extraction method for ball-mill bearing's vibration signals using wavelet analysis[J]. Noise and Vibration Control, 2016, 36(1):148-152.
[5]	方涛, 陈乐庚. 音频信号用于球磨机料位检测[J]. 噪声与振动控制, 2012, 32(2):143-146.
	Fang Tao, Chen Legeng. Audio signal analysis for detection of the raw material level in ball mill[J]. Noise and Vibration Control, 2012, 32(2):143-146.
[6]	阎高伟, 龚杏雄, 李国勇. 基于振动信号和云推理的球磨机负荷软测量[J]. 控制与决策, 2014, 29(6):1109-1114.
	Yan Gaowei, Gong Xingxiong, Li Guoyong. Soft sensor for ball mill fill level based on vibration signal and cloud model reasoning[J]. Control and Decision, 2014, 29(6):1109-1114.
[7]	惠瑜. 基于磨音的球磨机负荷软测量方法研究[D]. 西安:西安理工大学, 2014.
	Hui Yu. Research on the soft-sensing method of ball mill load by grinding sound[D]. Xi'an:Xi'an University ofTechnology, 2014.
[8]	刘志刚, 蔡改贫, 林龙飞, 等. 主元分析的振动频域特征识别与磨机负荷建模研究[J]. 中国钨业, 2016, 31(3):68-73.
	Liu Zhigang, Cai Gaipin, Lin Longfei, et al. Feature recognition for principal component analysis'vibration frequency domain and mill load modeling[J]. China Tungsten Industry, 2016, 31(3):68-73.
[9]	湛克金, 石建武. 球磨机噪声控制技术综述[J]. 中国电力, 1995(4):52-53.
	Zhan Kejin, Shi Jianwu. Summary of noise control technology of ball mill[J]. Electric Power, 1995(4):52-53.
[10]	姜德轩, 申涛, 王孝红. 基于小波分析的磨音信号处理[J]. 济南大学学报(自然科学版), 2011, 25(3):305-309.
	Jiang Dexuan, Shen Tao, Wang Xiaohong. Mill noise processing based on wavelet analysis[J]. Journal of University of Jinan(Science and Technology), 2011, 25(3):305-309.
[11]	沈志远, 王黎明, 陈方林. 基于有限长序列分析的Welch法谱估计研究[J]. 计算机仿真, 2010, 27(12):391-395.
	Shen Zhiyuan, Wang Liming, Chen Fanglin. Research on spectral estimation using Welch method based on analysis of finite-length sequence[J]. Computer Simulation, 2010, 27(12):391-395.
[12]	张杰, 肖玲玲, 张攀登, 等. 基于组合窗Welch法的心率变异性分析系统设计[J]. 中国医学物理学杂志, 2017, 34(2):194-198.
	Zhang Jie, Xiao Lingling, Zhang Pandeng, et al. Design of heart rate variability analysis system based on Welch method using combination window[J]. Chinese Journal of Medical Physics, 2017, 34(2):194-198.
[13]	邓泽怀, 刘波波, 李彦良. 常见的功率谱估计方法及其Matlab仿真[J]. 电子科技, 2014, 27(2):50-52.
	Deng Zehuai, Li Bobo, Li Yanliang. Common power spectrum estimation methods and Matlab simulation[J]. Electronic Science and Technology, 2014, 27(2):50-52.
[14]	周江嫚. 基于PCA-ELM的模拟电路故障诊断[J]. 电子科技, 2017, 30(5):72-75.
	Zhou Jiangman. A fault diagnosis method in analog circuits based on PCA-ELM[J]. Electronic Science and Technology, 2017, 30(5):72-75.
[15]	龚志飞, 郭迎清. 基于主元分析法的航空发动机传感器故障诊断研究[J]. 计算机测量与控制, 2012, 20(8):2017-2019.
	Gong Zhifei, Guo Yingqing. Investigation of aero-engine fault diagnosis based on principal component analysis[J]. Computer Measurement & Control, 2012, 20(8):2017-2019.
[16]	陈昌华, 谭俊, 尹健康, 等. 基于PCA-RBF神经网络的烟田土壤水分预测[J]. 农业工程学报, 2010, 26(8):85-90.
	Chen Changhua, Tan Jun, Yin Jiankang, et al. Prediction for soil moisture in tobacco fields based on PCA-RBF neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2010, 26(8):85-90.
[17]	李文锋, 曹玉莲, 张汉. 简单高效耦合策略的粒子群混合算法[J]. 控制理论与应用, 2018, 35(1):13-23.
	Li Wenfeng, Cao Yulian, Zhang Han. Hybrid particle swarm optimization algorithm with simple and efficient coupling strategy[J]. Control Theory & Applications, 2018, 35(1):13-23.
[18]	赵东波, 姚灵灵, 袁坤坤, 等. 基于粒子群算法多目标优化液压支架四连杆[J]. 工程设计学报, 2017, 24(4):433-439.
	Zhao Dongbo, Yao Lingling, Yuan Kunkun, et al. Multi-objective optimization for four-bar mechanism of hydraulic support based on particle swarm optimization algorithm[J]. Chinese Journal of Engineering Design, 2017, 24(4):433-439.
[19]	宋小杉, 蒋晓瑜, 罗建华, 等. 基于类间距的径向基函数-支持向量机核参数评价方法分析[J]. 兵工学报, 2012, 33(2):203-208.
	Song Xiaoshan, Jiang Xiaoyu, Luo Jianhua, et al. Analysis of the inter-class distance-based kernel parameter evaluating method for RBF-SVM[J]. Acta Armamentarii, 2012, 33(2):203-208.

负荷/%	R₁	R₃	R₄	R₅	R₆	R₇	R₈	R₉	…
10	0.469	-0.723	0.728	-0.217	0.487	-0.701	0.195	0.043	…
10	0.453	-0.759	0.659	-0.232	0.486	-0.786	0.195	0.158	…
20	0.854	-0.384	-0.079	-0.703	0.071	0.031	0.375	0.328	…
20	0.836	-0.313	-0.063	-0.67	0.046	0.040	0.050	0.371	…
30	0.020	0.196	-0.337	0.432	0.419	0.369	-0.296	0.361	…
30	0.021	0.344	-0.152	0.239	0.283	0.336	-0.406	0.517	…
40	-0.103	0.857	0.163	-0.766	-0.008	-0.191	0.319	-0.325	…
40	-0.129	0.793	0.144	-0.701	-0.198	-0.293	0.366	-0.071	…
50	-0.304	-0.397	0.564	-0.374	-0.062	0.002	-0.891	0.256	…
50	-0.332	-0.444	0.505	-0.275	0.037	0.201	-0.820	0.153	…

模型	RMSE	MAE	MAPE/%	时间/s
SVM	3.919 4	3.138 6	7.471 0	0.824 9
PCA-SVM	2.048 9	1.583 4	5.003 9	0.567 9
PCA-PSO-SVM	1.143 3	0.912 5	2.797 9	0.531 0