Mathematical Problems in Engineering

Mathematical Problems in Engineering / 2014 / Article

Research Article | Open Access

Volume 2014 |Article ID 242809 | 8 pages | https://doi.org/10.1155/2014/242809

Foundation Settlement Prediction Based on a Novel NGM Model

Academic Editor: Gradimir Milovanović
Received04 Dec 2013
Accepted18 Feb 2014
Published23 Mar 2014

Abstract

Prediction of foundation or subgrade settlement is very important during engineering construction. According to the fact that there are lots of settlement-time sequences with a nonhomogeneous index trend, a novel grey forecasting model called NGM model is proposed in this paper. With an optimized whitenization differential equation, the proposed NGM model has the property of white exponential law coincidence and can predict a pure nonhomogeneous index sequence precisely. We used two case studies to verify the predictive effect of NGM model for settlement prediction. The results show that this model can achieve excellent prediction accuracy; thus, the model is quite suitable for simulation and prediction of approximate nonhomogeneous index sequence and has excellent application value in settlement prediction.

1. Introduction

Excessive settlement of foundation, subgrade, and so forth, especially differential settlement, can severely harm buildings and structures. Therefore, it is very important to observe settlement during engineering construction. Through analysis of settlement data, the development trend of current settlement is judged and the ultimate settlement is predicted so that corresponding measures can be timely taken to avoid damage to buildings and structures due to excessive settlement. At present, settlement forecasting methods based on measured data mainly include statistical prediction method, neural network, and grey theory [13]. In this paper, the work is focused on the application of grey forecasting model in settlement prediction.

The grey forecasting model has been widely applied in the field of geotechnical engineering, since it was proposed [46]. GM model, as the uppermost grey forecasting model, is the most frequently used one [7, 8]. However, simulative sequences of GM model and various improved GM models are all homogeneous index sequences, while lots of data sequences in the field of geotechnical engineering have a nonhomogeneous index trend, such as foundation settlement-time sequences [9] and pile foundation load-displacement sequences [10]. In these cases, GM model is just applicable for short-term prediction of settlement rather than medium- and long-term prediction. For medium- and long-term prediction, a grey forecasting model based on nonhomogeneous index sequence should be adopted.

Many scholars have studied grey forecasting models based on nonhomogeneous index sequence [1115]. NGM model is one of those models that substantially differ from other similar models. From the modeling method of NGM model, it can be seen that there are only two parameters in its definition, while three parameters are set in other similar models. Then, does NGM model whose parameters are one fewer than other similar models have the same prediction performance as others? Through analysis, this paper points out that parameter setting is the fatal flaw of NGM model, which severely affects its application value. According to parameter settings of other similar models, this paper proposes a novel NGM model—NGM model.

Because of the inherent deviation existed in NGM model, this paper optimizes the parameters of the whitenization differential equation of the model to make it a better match with the grey differential equation, thus realizing prediction unbiasedness of NGM model. Change in the whitenization differential equation is based on parameter reconstruction; the general solution of the whitenization differential equation is substituted into the grey differential equation to find the new parameters of the whitenization differential equation, thus establishing an optimized NGM model. Finally, the optimized NGM model is applied in settlement prediction. The results of case studies show that this model can achieve excellent prediction accuracy which is better than that of GM model and NGM model and also superior to Asaoka model and hyperbolic model commonly used for settlement prediction, and therefore it has certain application value in settlement prediction.

2. The Modeling Method of NGM Model and Its Flaw

Definition 1. Consider where , is called a grey differential equation of NGM model (Abbreviate as NGM ), which is the defining type of NGM model. The parameter in NGM model is called the development coefficient and is grey action quantity just like in GM model.

The parameters of NGM model determined by least-squares method can be written as

The equation is called a whitenization differential equation of NGM model.

The equation is the time response equation of NGM model.

The restored values of can be given by

NGM model is established to broaden the application scope of grey forecasting models so that this type of model can apply to simulation and prediction of approximate nonhomogeneous index sequence. To analyze its flaw, assume that is a nonhomogeneous index sequence of raw data; then its 1-AGO sequence is where and .

Compare (4) with (6), if then

In the above equation, certain correlation exists among ,  , and , or there are only two independent variables; is an arbitrary function with three parameters and no direct correlation exists among ,  , and . Therefore, for NGM model, its simulative sequence cannot express an arbitrary nonhomogeneous index sequence, and its application scope is limited. To verify the flaw of NGM model, a NGM model is established based on nonhomogeneous index sequence . See Table 1 for the raw data and prediction results.


Original valueNGM ( ) modelGM ( ) model
Simulative value Error ( )Simulative value Error ( )

(1)8.600 8.600 0.00%8.600 0.00%
(2)9.320 5.501 40.97%9.211 1.17%
(3)10.184 9.077 10.87% 10.201 0.17%
(4)11.221 11.065 1.39% 11.297 0.68%
(5)12.465 12.171 2.36% 12.511 0.37%
(6)13.958 12.786 8.39% 13.855 0.74%
(avg)10.66%0.52%

From the results in Tables 1 and 2, it can be seen that the prediction accuracy of NGM model which is intended to be applicable for modeling of approximate nonhomogeneous index sequence is not as good as that of GM model, let alone nonhomogeneous index sequence which is selected in the case study of this paper, while GM model is only applicable for modeling of approximate homogeneous index sequence. Either simulation accuracy or prediction accuracy of NGM model is unsatisfactory or the relative error is more than 10%. As to GM model, its simulation accuracy and one-step prediction accuracy are both higher, indicating that the applicability of GM model is stronger than that of NGM model.


Original valueNGM ( ) modelGM ( ) model
Predictive value Error ( )Predictive value Error ( )

(7)15.750 13.129 16.642 15.344 2.576
(8)17.89913.31925.59116.9925.067
(9)20.47913.42534.44818.8188.111

3. A Novel NGM Model—NGM Model and Its Optimization

3.1. The Modeling Method of NGM Model

The above theoretical analyses and case study suggest that NGM model has its own flaw; that is, its simulative sequence cannot express an arbitrary nonhomogeneous index sequence. Compare it with other similar grey forecasting models.

The whitenization differential equation of the grey model proposed by Yu and Wei [13] is .

The whitenization differential equation of the grey model proposed by Zhang and Gu [14] is .

The above models both have three parameters and the parameters of solutions of their whitenization differential equations are all independent from each other. Thus, these models are both applicable for simulation and prediction of approximate nonhomogeneous index sequence, which are relatively effective models. According to similar models, a parameter must be added to NGM model.

Definition 2. The equation is called a grey differential equation of NGM model (abbreviated as NGM ), which is the defining type of NGM model. The parameter in the NGM model is called the development coefficient and is grey action quantity just like in GM model.

The parameters of NGM model calculated by least-square method can be written as where

The first-order differential equation is called a whitenization differential equation of NGM model.

The equation is called a time response equation of NGM model.

The restored values are where , , and are independent from each other, satisfying the requirement. Therefore, NGM model can be taken for simulation and prediction of approximate nonhomogeneous index sequence.

3.2. The Flaw of NGM Model

According to the above modeling method of NGM model, a problem similar to that in GM model can be found; that is, its grey differential equation and white differential equation do not match strictly, which results in inherent deviation of the model. This problem will be analyzed below.

On the interval , the integral form of whitenization differential equation (12) is then

Comparison between (16) and grey differential equation (15) shows that they are differences in and as well as background value and . Therefore, the grey differential equation and the white differential equation in NGM model do not match strictly.

3.3. Optimization of NGM Model

An optimization method similar to that for GM model, such as improved grey derivative or background value equation [1623], would disobey the definition of NGM model. Reference [7] mentioned that “GM white model itself and all results derived out from the white model just establish only when they are not contradictory with the defining type, otherwise, invalid.” Based on this idea, this paper improves the whitenization differential equation to make it a better match with the grey differential equation, so as to realize unbiasedness of the prediction model under the condition that the definition of NGM model is obeyed and to establish an optimized NGM model (abbreviated as ONGM ) as well.

Parameters of the whitenization differential equation are changed to match it with the grey differential equation; the whitenization differential equation with redefined parameters is written as

The general solution of the above equation is

1-IAGO sequence of is

Apparently, if (18) and (19) satisfy grey differential equation (9), the whitenization differential equation can match with the grey differential equation at this point. Substituting (18) and (19) into grey differential equation (9) gives

To make (20) hold true, we need then

The above are the parameters after the whitenization differential equation of NGM model is optimized. The time response equation of the model is The restored values are

4. Validation of White Exponential Law Coincidence

Similar to the flaw of GM model, the whitenization differential equation and the grey differential equation does not strictly match, which results in that NGM model cannot completely fit a pure nonhomogeneous index sequence.

A pure nonhomogeneous index sequence is used to verify the unbiasedness of prediction of ONGM model.

Accumulating the generation of , we get

Substituting all data values into the calculation formula (10) for parameters gives

Now we can obtain the parameters of whitenization differential equation (23) as follows:

Substitute (28) into the time response equation (23), then we can get

The restored values are

is exactly equal to , while , , and can take any value. Therefore, as long as the sequence of raw data has an approximate nonhomogeneous index trend, the optimized NGM model can be used for simulation and prediction.

5. Application of ONGM Model in Prediction of Foundation Settlement

5.1. Case Study 1

Close neighboring the Yangtze River, the northwest stockyard of a dock is soft subsoil; the bearing capacity of which cannot satisfy the stacking requirement of the stockyard. Gravel compaction piles with the diameter of 0.5 m which are arranged in a regular triangle and the drainage consolidation method of staged loading are adopted for reinforcement. The reinforced area covers the area within 10 m away from the stockyard [24]. 11 sets of observed values of postconstruction settlement (observation interval: 10 d) at Point H16 in Area B of the stockyard are chosen as the samples, and  d is denoted by ; ONGM model, NGM model, NGM model, and GM model are established for prediction of the ultimate settlement, so as to provide basis for engineering construction. Meanwhile, Asaoka model (graphic method) [9] and hyperbolic model [25] commonly used for settlement prediction are established for comparison with ONGM model presented in this paper. See Table 3 for the raw data and simulation results, and see Table 4 for the prediction results of ultimate settlement.


Number Time (day) Observed value (cm)Predictive value (cm)
ONGM (1, 1, , ) modelNGM (1, 1, , ) modelNGM (1, 1, ) modelGM (1, 1) modelAsaoka modelHyperbolic model

11023.3623.3623.3623.3623.3623.3625.55
22043.1942.1829.7031.9760.7143.4444.85
33058.7359.2549.2549.9765.8059.5859.94
44070.8772.8464.8264.6371.3172.5572.06
55083.7183.6477.2276.5877.2882.9982.01
66092.9192.2387.0886.3083.7591.3890.33
77099.7399.0794.9494.2390.7698.1297.39
880105.08104.50101.20100.6898.36103.54103.45
990109.73108.83106.18105.94106.60107.90108.71
10100112.19112.27110.14110.22115.52111.41113.32
11110113.45115.00113.30113.71125.20114.23117.39


ONGM (1, 1, , ) modelNGM (1, 1, , ) modelNGM (1, 1, ) modelGM (1, 1) modelAsaoka modelHyperbolic model

Mean absolute error 0.73 5.07 5.08 6.77 1.05 1.87
Mean absolute relative error0.94% 7.60% 7.31% 9.30% 1.18% 2.82%
Mean squared error0.88 39.95 36.71 69.73 1.45 4.18
Predictive value of ultimate settlement (cm) 125.63125.63129.03+∞125.79 183.28

The prediction performance of these models are compared by utilizing three indexes, namely, mean absolute error, mean absolute relative error, and mean squared error. See Table 4 for the results.

First of all, four grey forecasting models are compared. From the results in Table 4, it can be seen that the values of the indexes of GM model are the biggest, and the prediction performance of this model is the poorest; the values of indexes of ONGM model are much smaller than those of other models, and the prediction performance of this model is the best; NGM model and NGM model both have inherent deviation, and their prediction performance is quite close to each other in this case. This indicates that for approximate nonhomogeneous index sequence, ONGM model has a good prediction performance.

The results in Table 4 show that the predictive value of ultimate settlement of GM model goes to infinity, which goes against the actual situation; the predictive value of ultimate settlement of ONGM model is equal to that of NGM model, with the only difference in settlement convergence trend; the predictive value of ultimate settlement of NGM is greater than that of the former two models.

From the comparison among ONGM model, Asaoka model and hyperbolic model, it can be seen that all indexes of ONGM model are smaller than those of the other two models, suggesting that ONGM model has the best prediction performance among these three models. The prediction result of Asaoka model is closer to that of ONGM model, because these two models are essentially exponential models which are different only in modeling mechanism. Compared with the other two models, the prediction result of ultimate settlement from hyperbolic model is significantly greater.

5.2. Case Study 2

To construct a high-grade highway on soft silt, postconstruction settlement and embankment stability are two critical problems that must be solved. Prediction of ultimate settlement through analysis of postconstruction settlement is of great significance for the evaluation of embankment stability. The observed data of settlement of a cross-section after loading on the test section of a highway in the Pearl River Delta Region are taken as samples [26], and  d is denoted by ; ONGM model, NGM model, NGM model, GM model, Asaoka model (graphic method), and hyperbolic model are established for prediction of the ultimate settlement. See Table 5 for the raw data and simulation results, and see Table 6 for the prediction results of ultimate settlement.


Number Time (day) Observed value (cm)Predictive value (cm)
ONGM (1, 1, , ) modelNGM (1, 1, , ) modelNGM (1, 1, ) modelGM (1, 1) modelAsaoka modelHyperbolic model

185646464646464 57.47
2957373.12 64.58 43.53 76.19 72.93 76.24
31058180.69 73.27 70.51 80.78 80.63 85.56
41158787.26 80.81 84.94 85.64 87.27 91.13
51259392.96 87.36 92.66 90.80 92.99 94.83
61359897.90 93.03 96.79 96.27 97.92 97.47
7145102102.18 97.96 99.00 102.07 102.17 99.45
8155106105.90 102.23 100.18 108.22 105.83 100.99


ONGM (1, 1, , ) modelNGM (1, 1, , ) modelNGM (1, 1, ) modelGM (1, 1) modelAsaoka modelHyperbolic model

Mean absolute error 0.1395.0946.5481.3730.142 3.548
Mean absolute relative error0.158%5.855%8.214%1.561%0.160% 5.544%
Mean squared error0.02931.987128.4293.1000.035 15.786
Predictive value of ultimate settlement (cm) 130.21130.21101.54+∞128.72 113.24

First of all, four grey forecasting models are compared. From the results in Table 6, it can be seen that the values of the indexes of NGM model are the biggest, and the prediction performance of this model is the poorest; the values of indexes of ONGM model are much smaller than those of other models, and the prediction performance of this model is the best; the prediction performance of NGM model is slightly better than that of NGM model; the prediction performance of GM model is only second to ONGM model but cannot forecast the ultimate settlement. The results in Table 6 also indicate that for approximate nonhomogeneous index sequence, ONGM model has a good prediction performance.

The results in Table 6 show that the predictive value of ultimate settlement of NGM model is identical to that of NGM model, which is 130.21 cm bigger than the last observed value of settlement which is 106 cm; this implies that settlement is still unstable when observation is finished. The predictive value of ultimate settlement of NGM model is smaller than the last observed value of settlement, which goes against the actual situation.

From the comparison among ONGM model, Asaoka model, and hyperbolic model, it can be seen that all indexes of ONGM model are almost equal to but still smaller than those of Asaoka model and also smaller than hyperbolic model, suggesting that ONGM model has the best prediction performance among these three models. Compared with the other two models, the prediction result of ultimate settlement from the hyperbolic model is significantly smaller.

6. Conclusions

GM model has been successfully applied in many fields, but its prediction performance is sometimes unsatisfactory since its simulative sequence is homogeneous index sequence. For a lot of settlement-time sequences with a nonhomogeneous index trend, GM model is not appropriate for medium- and long-term prediction of foundation settlement.

The simulative sequence of NGM model cannot express an arbitrary nonhomogeneous index sequence and sometimes it cannot achieve a satisfactory prediction performance since this model only has two undetermined parameters. Using other similar models for reference, this paper adds a parameter to NGM model and establishes NGM model which is a novel NGM model.

Like GM model, NGM model also has the flaw of mismatching between its grey differential equation and whitenization differential equation. Based on the idea that “GM white model itself and all results derived out from the white model just establish only when they are not contradictory with the defining type, otherwise invalid”; this paper realizes prediction unbiasedness of NGM model through optimizing the parameters of the whitenization differential equation of this model.

Finally, the optimized NGM model is applied in settlement prediction. The results of case studies show that the optimized NGM model has good prediction performance which is better than that of GM model, NGM model, and NGM model and also superior to Asaoka model and hyperbolic model commonly used for settlement prediction; it has an excellent application value for approximate nonhomogeneous index sequence; thus, it is suitable for settlement prediction in geotechnical engineering.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgment

This research was supported by the National Natural Science Foundation of China (41272377).

References

  1. G. H. Li, H. B. Liu, and X. X. Qin, “Settlement prediction of roadbed based on mixture model with exponential curve and ANN,” Advanced Materials Research, vol. 663, pp. 76–79, 2013. View at: Publisher Site | Google Scholar
  2. G. Zhang, X. Xiang, and H. Tang, “Time series prediction of chimney foundation settlement by neural networks,” International Journal of Geomechanics, vol. 11, no. 3, pp. 154–158, 2011. View at: Publisher Site | Google Scholar
  3. H. B. Liu, Y. M. Xiang, and Y. X. Ruan, “A multivariable grey model based on background value optimization and its application to subgrade settlement prediction,” Rock and Soil Mechanics, vol. 34, no. 1, pp. 173–181, 2013. View at: Google Scholar
  4. F. Xie, L. M. Zhu, and L. Z. Wang, “Modified grey system forecasting model and its application for analyzing information of landslide monitory,” Chinese Journal of Rock Mechanics and Engineering, vol. 24, no. 22, pp. 4099–4105, 2005. View at: Google Scholar
  5. Z. Feng, Z. Li, and Y. Li, “Application of a multi-point grey model to deformation predicting of supporting structure for deep pit,” Chinese Journal of Rock Mechanics and Engineering, vol. 26, supplement 2, pp. 4319–4324, 2007. View at: Google Scholar
  6. X.-P. He, X.-S. Hua, and X.-F. He, “Weighted multi-point grey model and its application to high rock slope deformation forecast,” Rock and Soil Mechanics, vol. 28, no. 6, pp. 1187–1191, 2007. View at: Google Scholar
  7. J. L. Deng, Grey Prediction and Decision, Press of Huazhong University of Science & Technology, Wuhan, China, 2002.
  8. S. F. Liu and N. M. Xie, Grey System Theory and Its Application, Science Press, Beijing, China, 4th edition, 2008.
  9. A. Asaoka, “Observational procedure of settlement prediction,” Soils and Foundations, vol. 18, no. 4, pp. 87–101, 1978. View at: Publisher Site | Google Scholar
  10. C. Zhao, S. Li, J. Lu, and A. Gan, “Analysis of fitting loading settlement curves of single piles by integrated exponential function,” Journal of Tongji University, vol. 38, no. 4, pp. 486–492, 2010. View at: Publisher Site | Google Scholar
  11. J. Cui, Y. G. Dang, and S. F. Liu, “Novel grey forecasting model and its modeling mechanism,” Control and Decision, vol. 24, no. 11, pp. 1702–1706, 2009. View at: Google Scholar | Zentralblatt MATH
  12. J. Cui, S. F. Liu, B. Zeng, and N. M. Xie, “A novel grey forecasting model and its optimization,” Applied Mathematical Modelling, vol. 33, no. 4, pp. 1894–1903, 2009. View at: Publisher Site | Google Scholar
  13. D. Yu and Y. Wei, “Non-homogeneous index sequence of GM model,” in Proceedings of the 16th Workshop on Grey System Theory and Its Applications, pp. 355–366, China Center of Advanced Science and Technology, 2008. View at: Google Scholar
  14. W. W. Zhang and G. D. Gu, Non-Homogeneous Exponential Model of National Power Generating Capacity and Its Recursive Algorithm, Progress of Management Science and System Science, Systems Engineering Society of China, 1995.
  15. N. M. Xie, S. F. Liu, Y. J. Yang, and C. Q. Yuan, “On novel grey forecasting model based on non-homogeneous index sequence,” Applied Mathematical Modelling, vol. 37, no. 7, pp. 5059–5068, 2013. View at: Publisher Site | Google Scholar
  16. Y. Zhang and Y. Wei, “The improved approach of grey derivative in GM (1, 1) model,” Journal of Grey System, vol. 18, no. 4, pp. 375–380, 2006. View at: Google Scholar
  17. Y. Zhang and Y. Wei, “An approach of GM (1, 1) based on optimum grey derivative,” Journal of Grey System, vol. 19, no. 4, pp. 397–404, 2007. View at: Google Scholar
  18. Y. N. Sun and Y. Wei, “Optimization of grey derivative in GM (1, 1) based on the discrete exponential sequence,” in Proceedings of the International Symposium on Information Processing, pp. 313–315, 2009. View at: Google Scholar
  19. B. LI and Y. Wei, “Optimizes grey derivative of GM (1, 1),” System Engineering Theory and Practice, vol. 29, no. 2, pp. 101–105, 2009. View at: Google Scholar
  20. Y. Wang, Y. Dang, Y. Li, and S. Liu, “A new method for improving prediction precision of GM(1,1) model,” Journal of Grey System, vol. 21, no. 3, pp. 301–308, 2009. View at: Google Scholar
  21. L. Bao and W. Yong, “Optimized GM(1,1) grey model based on romberg algorithm,” Journal of Grey System, vol. 20, no. 2, pp. 125–134, 2008. View at: Google Scholar
  22. D. Q. Truong and K. K. Ahn, “An accurate signal estimator using a novel smart adaptive grey model SAGM(1,1),” Expert Systems with Applications, vol. 39, no. 9, pp. 7611–7620, 2012. View at: Publisher Site | Google Scholar
  23. Z. X. Wang, Y. G. Dang, and S. F. Liu, “An Optimal GM(1, 1) based on the discrete function with exponential law,” System Engineering Theory and Practice, vol. 28, no. 2, pp. 61–67, 2008. View at: Publisher Site | Google Scholar
  24. T. Ling, X. L. Ouyang, and C. Q. Hu, “Combined prediction of settlement of composite foundation with crushed gravel piles,” Journal of Lanzhou University of Technology, vol. 36, no. 5, pp. 113–116, 2010. View at: Google Scholar
  25. L. Gao, Q. Y. Zhou, X. J. Yu, and Z. H. Chen, “Analysis and model prediction of subgrade settlement for Linhai highway in China,” Electronic Journal of Geotechnical Engineering, vol. 19, no. A, pp. 11–21, 2014. View at: Google Scholar
  26. J. Y. Shi, C. E. Zhou, Z. X. Gao, and M. K. Lv, “Local test analysis of auto-way,” in Proceedings of the 7th Conference of Soil Mechanics and Foundation Engineering, pp. 464–469, China Architecture and Building Press, 1994. View at: Google Scholar

Copyright © 2014 Peng-Yu Chen and Hong-Ming Yu. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

853 Views | 640 Downloads | 13 Citations
 PDF  Download Citation  Citation
 Download other formatsMore
 Order printed copiesOrder

We are committed to sharing findings related to COVID-19 as quickly and safely as possible. Any author submitting a COVID-19 paper should notify us at help@hindawi.com to ensure their research is fast-tracked and made available on a preprint server as soon as possible. We will be providing unlimited waivers of publication charges for accepted articles related to COVID-19. Sign up here as a reviewer to help fast-track new submissions.