Support vector regression (SVR) performs satisfactorily in prediction problems, especially for small sample prediction. The setting parameters (e.g., kernel type and penalty factor) profoundly impact the performance and efficiency of SVR. The adaptive adjustment of the parameters has always been a research hotspot. However, the substantial time cost and forecast accuracy of parameter adjustment are challenging to many scholars. The contradiction of big data prediction is especially prominent. In the paper, an SVR-based prediction approach is presented using the K-means clustering method (KMCM) and chaotic slime mould algorithm (CSMA). Eight high-and low-dimensional benchmark datasets are applied to obtain appropriate key parameters of KMCM and CSMA, and the forecast accuracy, stability performance and computation complexity are evaluated. The proposed approach obtains the optimal (joint best) forecast accuracy on 6 datasets and produces the most stable output on 3 datasets. It ranks first with a score of 0.024 in the overall evaluation. The outcomes reveal that the proposed approach is capable of tuning the parameters of SVR. KMCM, CSMA and SVR are skillfully integrated in this work and perform well. Although the performance is not outstanding in terms of stability, the proposed approach exhibits very strong performance with respect to prediction accuracy and computation complexity. This work validated the tremendous potential of the proposed approach in the prediction field. INDEX TERMS Machine learning, forecasting problem, K-means clustering method, chaotic slime mould algorithm, support vector regression.

No abstract

The dredger fill of Shanghai Hengsha Island Dongtan is solidified by curing agents with different fly ash content, and the shear strength index of solidified dredger fill is measured by the direct shear test. The microscopic images of solidified dredger fill are obtained by using SEM. The microscopic images are processed and analyzed by using IPP, and the fractal dimension including particle size fractal dimension Dps, aperture fractal dimension Dbs and particle surface fractal dimension Dpr is calculated by fractal theory. The quantitative analysis of the relationship between shear strength index and fractal dimension of solidified dredger fill is done. The research results show that the internal friction angle and the cohesion are closely related to the fly ash content λ and the curing period T, and the addition of fly ash can improve the effect of curing agent; There is no obvious linear relationship between the internal friction angle and the three fractal dimensions; The smaller particle surface fractal dimension Dpr and particle size fractal dimension Dps, the larger aperture fractal dimension Dbs, the greater the cohesion, and the cohesion has a good linear relationship with three fractal dimensions, and the correlation coefficient R2 is above 0.91.

In the South China Sea, calcareous sand, as a natural foundation, has the features of low mechanical properties, including its compressive strength. With the development of South China Sea islands, the problems of calcareous sand foundation are encountered in the process. However, the experience of traditional pile foundation engineering could not be applied to calcareous sand. In this study, different proportions of curing agents were added to calcareous sand to improve the compressive strength. The quantitative analysis of the relationship between the unconfined compressive strength and microstructure of solidified calcareous sand is discussed. The unconfined compressive strength was gauged from unconfined compressive strength tests. Microscopic images, acquired using a scanning electron microscope (SEM), were processed using the Image-Pro Plus (IPP) image processing software. The microscopic parameters, obtained using IPP, include the average equivalent particle size (Dp), the average equivalent aperture size (Db), and the plane pore ratio (e). This research demonstrates that the curing agent could improve the compressive strength, which has a relation with the three microstructure parameters. The curing agent, through hydration reaction, generates hydration products, i.e., calcium silicate hydrate, calcium hydroxide, and calcite crystals. They adhere to the surface of the particles or fill the space between the particles, which helps increase the compressive strength. In addition, there is a good linear relationship between the macroscopic mechanics and the microscopic parameters. Using the mathematical relation between the macroscopic and microscopic parameters, the correlation can be built for macro-microscopic research.

Adding a curing agent can enhance the mechanical properties of soil including its compressive strength. However, few studies have quantitatively analyzed the compressive strength and microstructure of soils to explore the impact of changes in the microstructure on compressive strength. In addition, the cost of curing agents is too high to be widely used. In this study, curing agents with different proportions of fly ash were added to dredger fill to reduce the amount of curing agents needed. The quantitative analysis of the relationships between the modulus of compression Es and microstructures of stabilized soil samples is presented. The modulus of compression Es was gauged from compression tests. Microscopic images acquired using a scanning electron microscope were processed using the Image-Pro Plus (IPP) image processing software. The microscopic parameters, obtained using IPP, included the average equivalent particle size Dp, the average equivalent aperture size Db, and the plane pore ratio e. This research demonstrated that the fly ash added to the curing agent achieved the same effect as the curing agent, and the amount of curing agent required was reduced. Therefore, the modulus of compression for stabilized soil can be improved. This is due to the hydration products (i.e., calcium silicate hydrate, calcium hydroxide, and ettringite), produced by the hydration reaction, and which adhere to the surface of the particles and fill the spaces among them. Thus, the change in the pore structure and the compactness of the particles helps to increase the modulus of compression. In addition, there was a good linear relationship between the modulus of compression and the microscopic parameters. Using the mathematical relationships between the macroscopic and microscopic parameters, correlations can be built for macro–microscopic research.