Slope displacement prediction based on morphological filtering

Abstract

Combining mathematical morphology (MM), nonparametric and nonlinear model, a novel approach for predicting slope displacement was developed to improve the prediction accuracy. A parallel-composed morphological filter with multiple structure elements was designed to process measured displacement time series with adaptive multi-scale decoupling. Whereafter, functional-coefficient auto regressive (FAR) models were established for the random subsequences. Meanwhile, the trend subsequence was processed by least squares support vector machine (LSSVM) algorithm. Finally, extrapolation results obtained were superposed to get the ultimate prediction result. Case study and comparative analysis demonstrate that the presented method can optimize training samples and show a good nonlinear predicting performance with low risk of choosing wrong algorithms. Mean absolute percentage error (MAPE) and root mean square error (RMSE) of the MM-FAR&LSSVM predicting results are as low as 1.670% and 0.172 mm, respectively, which means that the prediction accuracy are improved significantly.

Keywords

slope displacement prediction / parallel-composed morphological filter / functional-coefficient auto regressive / prediction accuracy

References

[1] KodamaJ, NishiyamaE, KanekoK. Measurement and interpretation of long-term deformation of a rock slope at the Ikura limestone quarry, Japan [J]. International Journal of Rock Mechanics & Mining Sciences, 2009, 46(1): 148-158

[2] LinH, CaoP, GongF-q, LiJ-t, GuiY-lin. Directly searching method for slip plane and its influential factors based on critical state of slope [J]. Journal of Central South University of Technology, 2009, 16(1): 131-135

[3] ZhangK, CaoPing. Modified electromagnetism-like algorithm and its application to slope stability analysis [J]. Journal of Central South University of Technology, 2011, 18(6): 2100-2107

[4] MaW-tao. Prediction of slope displacement based on wavelet transform and genetic algorithm-least square support vector machine [J]. Rock and Soil Mechanics, 2009, 30z2: 394-405

[5] GuoZ-h, ZhaoJ, ZhangW-y, WangJ-zhou. A corrected hybrid approach for wind speed prediction in Hexi Corridor of China [J]. Energy, 2011, 36(3): 1668-1679

[6] AngX-l, LiHui. One-month ahead prediction of wind speed and output power based on EMD and LSSVM [C]. International Conference on Energy and Environment Technology, 2009PiscatawayIEEE Computer Society439-442

[7] SchavemakerJ G M, ReindersM J T, GerbrandsJ J, BackerE. Image sharpening by morphological filtering [J]. Pattern Recognition, 2000, 33(6): 997-1012

[8] LiB, ZhangP-l, WangZ-j, MiS-s, ZhangY-tang. Gear fault detection using multi-scale morphological filters [J]. Measurement: Journal of the International Measurement Confederation, 2011, 44(10): 2078-2089

[9] JiangY-chong. X-ray bone fracture subtraction using geodesic active contour and mathematical morphology operations [J]. International Congress Series, 2003, 1256: 226-231

[10] MariuszM. Description and classification of rock surfaces by means of laser profilometry and mathematical morphology [J]. International Journal of Rock Mechanics and Mining Sciences, 2010, 47(1): 138-149

[11] FengX-t, ZhaoH-b, LiS-jun. Modeling non-linear displacement time series of geo-materials using evolutionary support vector machines [J]. International Journal of Rock Mechanics and Mining Sciences, 2004, 41(7): 1087-1107

[12] YanW-w, ShaoH-he. Application of support vector machines and least squares support vector machines to heart disease diagnoses [J]. Control and Decision, 2003, 18(3): 356-360

[13] GuoJ-r, HeY-g, LiuC-qing. Nonlinear correction of photoelectric displacement sensor based on least square support vector machine [J]. Journal of Central South University of Technology, 2011, 18(5): 1614-1618

[14] IsmailS, ShabriA, SamsudinR. A hybrid model of self-organizing maps (SOM) and least square support vector machine (LSSVM) for time-series forecasting [J]. Expert Systems with Applications, 2011, 38(8): 10574-10578

[15] SamuiP, KothariD P. Utilization of a least square support vector machine (LSSVM) for slope stability analysis [J]. Scientia Iranica, 2011, 18(1): 53-58

[16] HarvillJ L. Functional coefficient autoregressive models for vector time series [J]. Computational Statistics & Data Analysis, 2006, 50(12): 3547-3566

[17] SerraJImage analysis and mathematical morphology [M], 1982LondonAcademic Press23-35

[18] KumarJ, ShunmugamM S. A new approach for filtering of surface profiles using morphological operations [J]. International Journal of Machine Tools and Manufacture, 2006, 46(3): 260-270

[19] JonesR, SvalbeI. The design of morphological filters using multiple structuring elements: Part II. Open(close) and close(open) [J]. Pattern Recognition Letters, 1992, 13(3): 175-181

[20] CaiZ-w, LiQ, ParkJ Y. Functional-coefficient models for nonstationary time series data [J]. Journal of Econometrics, 2009, 148(2): 101-113

[21] HarvillJ L, RayB K. A note on multi-step forecasting with functional coefficient autoregressive models [J]. International Journal of Forecasting, 2005, 21(4): 717-727

[22] SuykensJ A K, VanG T, DeB J, et al. Least square support vector machines [M], 2002SingaporeWorld Scientific15-61

[23] DongH, FuH-l, LengW-ming. Kernel design for displacement time series of landslide [J]. Rock and Soil Mechanics, 2008, 29(4): 1087-1906