基于KPCA-IPSO-LSSVM的充填管道磨损风险预测

doi:10.11872/j.issn.1005-2518.2021.02.072

Abstract

Abstract:

Filling pipeline wear system is a typical high-dimensional，nonlinear，strong coupling and multi-time-varying complex system.It is difficult to accurately predict the wear situation using traditional prediction methods.In order to overcome the poor applicability and insufficient prediction accuracy of traditional prediction models，and the defect of strong randomness in parameter selection and so on，this paper proposed a new method for predicting the wear risk of filling pipeline by combining the kernel principal component analysis （KPCA），improved particle swarm optimization （IPSO） and least squares support vector machine （LSSVM）.A comprehensive selection of 12 main influencing factors was used to establish the prediction index of wear risk of the filling pipeline.The KPCA method was used for feature extraction and dimensionality reduction of the influencing factors of the filling pipeline to eliminate redundant information between the data，so as to reduce the correlation between sample data and modeling accuracy impact.Established the corresponding LSSVM prediction model based on the dimensionality-reduced data，and used the IPSO algorithm with strong global search capability to optimize the model parameters to avoid the blindness of artificial parameter selection，thereby improving the model prediction accuracy and establishing KPCA-IPSO-LSSVM combined prediction model.Taking the filling system of Huangling County mining area as an example，combining the 80 sets of sample data measured in the field，MATLAB was used to train and predict the built model，and the prediction results were compared with the prediction results of the IPSO-LSSVM model，LSSVM model，and SVM model.For comparison，multi-error indicators were used to comprehensively analyze the prediction results of the four models.The research results show that the predicted value of the constructed model is basically consistent with the actual value curve.The KPCA method can effectively reduce the redundant information between the data.On the basis of retaining the original sample information to the maximum，five principal components containing 86.97% of the original information are extracted，it simplifies the calculation structure of the model.The prediction accuracy of the adopted IPSO-LSSVM model is 6.79%，the average relative error is 1.95%，and the judgment coefficient is 99.55%.Compared with other prediction models，the prediction model based on KPCA-IPSO-LSSVM has higher prediction accuracy and stronger generalization ability，which provides a more effective prediction method for the prediction of the wear risk of the filling pipeline，and provides a guiding basis for ensuring the smooth progress of the filling operation and the safe production of the mine.

Key words: kernel principal component analysis, improved particle swarm algorithm, least squares support vector machine, filling pipeline, wear risk, combined prediction model

CLC Number:

TD853.34

Zhengshan LUO,Renhui HUANG,Guochen SHEN. Research on Wear Risk Prediction of Filling Pipeline Based on KPCA-IPSO-LSSVM[J].Gold Science and Technology, 2021, 29(2): 245-255.

Figures/Tables 15

Fig.1

Fig.2

Fig.3

Fig.4

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

Fig.5

Fig.6

Fig.7

Fig.8

Table 7

References 0

	Batista V R，Zampirolli F A，2019.Optimising robotic pool-cleaning with a genetic algorithm［J］.Journal of Intelligent & Robotic Systems，95（2）：443-458.
	Chen Yong，Li Peng，Zhang Zhongjun，al et，2019.Online prediction model for power transmission line icing load based on PCA-GA-LSSVM［J］.Power System Protection and Control，47（10）：110-119.
	Feng Ju’en，Wu Chao，2005.Fuzzy comprehensive evaluation on probability failure criteria of the deep level pipelines filling system［J］.Journal of Central South University（Natural Science Edition），（6）：1079-1083.
	Feng Yunfei，Meng Fanbo，Zhu Fubin，al et，2013.Risk assessment of long-distance pipeline based on gray analytic hierarchy method［J］.Gas Storage and Transportation，32（12）：1289-1294.
	Guo Jiang，Zhang Bixiao，2015.Invalidation risk evaluation of backfill pipe based on PCA and BP neural network［J］.Gold Science and Technology，23（5）：66-71.
	Huang Xiaolu，Zhou Xiangzhen，2019.A fault diagnosis method based on support vector machine optimized by improved fruit fly optimization algorithm［J］.Journal of Mechanical Strength，41（3）：568-574.
	Kennedy J，2011.Particle swarm optimization［C］//Encyclopedia of Machine Learning.Boston：Springer..
	Lahdhiri H， Taouali O，2021.Reduced Rank KPCA based on GLRT chart for sensor fault detection in nonlinear chemical process［J］.Measurement，169：108342. doi.org/10.1016/j.measurement.2020.108342.
	Li Xiaochen，Nie Xingxin，2019.KPCA-PSO-GRNN evaluation model and application of filling pipeline wear risk［J］.Nonferrous Metal Engineering，9（2）：84-92.
	Liu Mingqiang，Li Yingpan，Chen Xiao，al et，2018.Prediction of safety-civilized measure cost for fabricated building project based on RA-LSSSVM［J］.China Safety Science Journal，28（1）：149-154.
	Luo Zhengshan，Wang Wenhui，Zhang Xinsheng，2019.Study on failure risk of backfill pipeline based on RS-GWO-GRNN［J］.Nonferrous Metal Engineering，9（6）：76-83.
	Pengfei Lü，Chen Xuehua，2017.PSO optimized LSSVM model by predicting the convergence between the roof and the floor in the gateways［J］.Journal of Safety and Environment，17（6）：2045-2049.
	Mao Zhiyong，Huang Chunjuan，Lu Shichang，al et，2018.KPCA-MPSO-ELM based model for discrimination of mine water inrush source［J］.China Safety Science Journal，28（8）：111-116.
	Nabipour N，Qasem S N， Salwana E，al et，2020.Evolving LSSVM and ELM models to predict solubility of non-hydrocarbon gases in aqueous electrolyte systems［J］.Measurement，164：107999. doi.org/10.1016/j.measurement.2020.107999.
	Shen G C，Jia W Y，2014.The prediction model of financial crisis based on the combination of principle component analysis and support vector machine［J］.Open Journal of Social Sciences，2：204-212.
	Suykens J A K，Vandewalle J，1999. Least squares support vector machine classifiers［J］.Neural Processing Letters，9（3）： 293-300．
	Wang Enjie，Zhao Guoyan，Wu Hao，al et，2018.Weight-variation-fuzzy model for assessing wear risk of backfilling pipeline and its application［J］.China Safety Science Journal，28（3）：149-154.
	Wang Sheng，Yang Xinfeng，2019.Short-term passenger flow forecasting of public transport based on EEMD-GWO-LSSVM ［J］.Computer Engineering and Applications，55（20）：216-221，239.
	Wang Xinmin，Wang Shi，Yan Debo，al et，2012.Risk assessment on blocking of filling pipeline based on uncertainty measure theory［J］.China Safety Science Journal，22（4）：151-156.
	Yang Lei，Zhang Baofeng，Zhu Junchao，al et，2018.Temperature prediction method of greenhouse based on PCA-PSO-LSSVM［J］.Transducer and Microsystem Technologies，37（7）：52-55.
	Zhang Deming，2012.A Study on Wear Mechanism and the Reliability Evaluation System for Backfilling Pipelines in Deep Mine［D］.Changsha：Central South University.
	Zhang Qinli，Wang Jing，Wang Xinmin，2017.Risk assessment model for filling pipeline based on KPCA and PSO-SVM［J］.Gold Science and Technology，25（3）：70-76.
	Zhang Wansheng，2018.Research of railway monthly passenger volume forecast model based on PCA-IPSO-GNN model［J］.China Public Safety（Academic Edition），（2）：122-127.
	陈勇，李鹏，张忠军，等，2019.基于PCA-GA-LSSVM的输电线路覆冰负荷在线预测模型［J］.电力系统保护与控制，47（10）：110-119.
	冯巨恩，吴超，2005.深井充填管道失效概率准则的模糊综合评判［J］.中南大学学报（自然科学版），（6）：1079-1083.
	冯云飞，孟繁博，朱富斌，等，2013.基于灰色层次分析法的长输管道风险评价［J］.油气储运，32（12）：1289-1294.
	过江，张碧肖，2015.基于PCA与BP神经网络的充填管道失效风险评估［J］.黄金科学技术，23（5）：66-71.
	黄晓璐，周湘贞，2019.基于改进果蝇优化算法优化支持向量机的故障诊断［J］.机械强度，41（3）：568-574.
	李晓晨，聂兴信，2019.充填管道磨损风险的KPCA-PSO-GRNN评估模型及应用［J］.有色金属工程，9（2）：84-92.
	刘名强，李英攀，陈晓，等，2018.装配式建筑安全文明施工费RS-LSSVM预测方法［J］.中国安全科学学报，28（1）：149-154.
	骆正山，王文辉，张新生，2019.基于RS-GWO-GRNN的充填管道失效风险研究［J］.有色金属工程，9（6）：76-83.
	吕鹏飞，陈学华，2017.基于PSO优化LSSVM模型的回采巷道顶底板移近量预测［J］.安全与环境学报，17（6）：2045-2049
	毛志勇，黄春娟，路世昌，等，2018.基于KPCA-MPSO-ELM的矿井突水水源判别模型［J］.中国安全科学学报，28（8）：111-116.
	王恩杰，赵国彦，吴浩，等，2018.充填管道磨损变权—模糊风险评估模型［J］.中国安全科学学报，28（3）：149-154.
	王盛，杨信丰，2019.基于EEMD-GWO-LSSVM的公共交通短期客流预测［J］.计算机工程与应用，55（20）：216-221，239.
	王新民，王石，鄢德波，等，2012.基于未确知测度理论的充填管道堵塞风险性评价［J］.中国安全科学学报，22（4）：151-156.
	杨雷，张宝峰，朱均超，等，2018.基于PCA-PSO-LSSVM的温室大棚温度预测方法［J］.传感器与微系统，37（7）：52-55.
	张德明，2012.深井充填管道磨损机理及可靠性评价体系研究［D］.长沙：中南大学.
	张钦礼，王兢，王新民，2017.基于核主成分分析与PSO-SVM的充填管道失效风险性分级评价模型［J］.黄金科学技术，25（3）：70-76.
	张万胜，2018.基于PCA-IPSO-GNN模型的铁路月度客运量预测模型研究［J］.中国公共安全（学术版），（2）：122-127.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Comments

Recommended 0

No Suggested Reading articles found!

赋值	浆料腐蚀性	充填骨料形状	管道的耐磨性	管线变化程度	管道安装质量
[0，2）	偏酸或偏碱性，含易腐蚀管道的物质	表面锋利的极不规则形	极差	布置极复杂、弯管极多	极差
[2，4）	弱酸或弱碱性，含轻微腐蚀管道的物质	表面钝化的多菱角形	差	布置复杂、弯管多	差
[4，6）	pH值偶尔变化引起的管道腐蚀	基本光滑的多边形、椭球型	好	布置简单、弯管少	好
[6，8）	中性且基本不会发生化学腐蚀	表面较光滑的球形	极好	布置极简单、弯管极少	极好

样本	C₁/（t·m^-3）	C₂/mm	C₃	C₄	C₅/mm	C₆	C₇/mm	C₈	C₉	C₁₀	C₁₁	C₁₂/%	等级
1	1.97	0.62	1.2	4.5	14	3.1	152	1.83	2.9	4.1	3.2	4.6	Ⅱ
2	1.76	0.08	6.8	3.8	20	4.8	179	2.52	4.8	5.3	2.3	1.25	Ⅲ
?	?	?	?	?	?	?	?	?	?	?	?	?	?
73	1.78	0.05	4.4	5	20	5.6	148	1.66	4.7	5.2	5.2	1.58	Ⅳ
74	1.78	0.62	2.1	3.2	18	3.1	168	1.8	4.2	2.8	3.2	1.5	Ⅱ
75	1.69	0.08	3.5	4.8	22	6.2	145	3.2	9.6	5.1	4.2	0.91	Ⅲ
76	1.83	0.52	7.4	6.3	28	7.7	69	1.5	6.7	4.9	7.5	1.65	Ⅱ
77	1.92	0.11	4.8	4.4	24	4.8	98	3.5	5.8	5.4	5.5	1.19	Ⅲ
78	1.92	0.06	6.9	4.2	26	5.6	104	3.3	3.8	5.6	4.7	1.01	Ⅳ
79	1.71	0.05	4.4	6.1	28	7.2	72	1.7	3.5	7.5	6.8	2.67	Ⅲ
80	1.68	0.28	4.2	4.6	20	5.5	78	1.6	5.2	5.9	5.5	1.18	Ⅱ

样本	C₁/（t·m^-3）	C₂/mm	C₃	C₄	C₅/mm	C₆	C₇/mm	C₈	C₉	C₁₀	C₁₁	C₁₂/%	等级
1	1.57	1.48	-1.82	-0.47	-1.57	-1.60	0.90	-0.49	-1.05	-0.93	-1.30	2.41	Ⅱ
2	-0.44	-0.73	1.00	-1.04	-0.35	-0.55	1.54	0.37	-0.09	0.03	-1.88	-0.42	Ⅲ
?	?	?	?	?	?	?	?	?	?	?	?	?	?
73	-0.25	-0.85	-0.21	-0.07	-0.35	-0.06	0.80	-0.70	-0.14	-0.05	0.01	-0.14	Ⅳ
74	-0.25	1.48	-1.37	-1.52	-0.75	-1.60	1.28	-0.52	-0.39	-1.98	-1.30	-0.21	Ⅱ
75	-1.11	-0.73	-0.66	-0.23	0.06	0.32	0.73	1.22	2.32	-0.13	-0.64	-0.70	Ⅲ
76	0.23	1.07	1.30	0.99	1.28	1.24	-1.09	-0.90	0.86	-0.29	1.51	-0.08	Ⅱ
77	1.09	-0.61	-0.01	-0.55	0.47	-0.55	-0.39	1.59	0.41	0.11	0.21	-0.47	Ⅲ
78	1.09	-0.81	1.05	-0.71	0.88	-0.06	-0.25	1.34	-0.60	0.27	-0.32	-0.62	Ⅳ
79	-0.92	-0.85	-0.21	0.83	1.28	0.94	-1.01	-0.65	-0.75	1.80	1.05	0.78	Ⅲ
80	-1.20	0.09	-0.31	-0.39	-0.35	-0.12	-0.87	-0.77	0.11	0.51	0.21	-0.48	Ⅱ

主成分	特征值	贡献率/%	累计贡献率/%
1	1.775	40.22	40.22
2	0.854	19.35	59.56
3	0.503	11.39	70.96
4	0.367	8.31	79.27
5	0.340	7.71	86.97
6	0.202	4.57	91.55

影响因素	F₁	F₂	F₃	F₄	F₅
C₁	-0.870	0.215	0.052	0.005	0.031
C₂	-0.155	-0.378	0.052	-0.029	-0.084
C₃	-0.583	0.342	-0.341	-0.003	-0.089
C₄	-0.871	0.034	0.273	-0.047	0.058
C₅	0.198	-0.282	-0.113	-0.266	0.024
C₆	0.661	0.184	0.060	-0.055	-0.038
C₇	-0.008	-0.261	-0.134	0.082	0.159
C₈	0.140	-0.207	-0.084	0.167	0.107
C₉	0.570	0.226	0.070	0.096	-0.026
C₁₀	0.295	-0.035	0.027	0.022	-0.151
C₁₁	0.020	-0.177	0.079	0.094	-0.126
C₁₂	0.603	0.337	0.058	-0.066	0.136

Research on Wear Risk Prediction of Filling Pipeline Based on KPCA-IPSO-LSSVM

RichHTML

PDF (PC)

Abstract

Cite this article

share this article

Figures/Tables 15

References 0

Related Articles 2

Metrics

Comments

Recommended 0

样本编号	F₁	F₂	F₃	F₄	F₅	等级
1	-1.041	0.569	0.396	-0.067	-0.578	Ⅱ
2	-1.822	-0.024	-0.995	-0.335	0.043	Ⅲ
?	?	?	?	?	?	?
73	-0.748	0.300	-0.32	0.210	-0.111	Ⅳ
74	-0.517	1.644	-0.352	-0.421	-0.160	Ⅱ
75	1.146	0.614	-0.574	-0.619	0.213	Ⅲ
76	1.076	0.005	0.614	0.489	0.424	Ⅱ
77	-0.818	0.031	0.262	-0.563	0.253	Ⅲ
78	-2.406	-0.665	0.045	-0.433	0.236	Ⅳ
79	0.253	-0.760	0.333	0.630	-0.430	Ⅲ
80	2.053	0.175	0.108	0.150	-0.115	Ⅱ

预测模型	RMSE	MRE/%	R²/%
SVM	0.4541	16.15	90.46
LSSVM	0.2483	9.10	95.54
IPSO-LSSVM	0.1816	6.76	97.26
KPCA-IPSO-LSSVM	0.0679	1.95	99.55

[1]	Rui XU, Kuikui HOU, Xi WANG, Xingquan LIU, Xibing LI. Combined Prediction Model of Rockburst Intensity Based on Kernel Principal Component Analysis and SVM [J]. Gold Science and Technology, 2020, 28(4): 575-584.
[2]	Shi WANG,Yi TANG,Xiao FENG. Risk Assessment of Filling Pipeline Wearing Based on Improved PCA and Ordered Multi-class Logistic [J]. Gold Science and Technology, 2019, 27(5): 740-746.