• Korean Chemical Engineering Research
• /
• v.61 no.1
• /
• pp.123-141
• /
• 2023

A sufficient amount of data with quality is needed for training artificial neural networks (ANNs). However, developing ANN models with a small amount of data often appears in engineering fields. This paper presented an ANN model to improve prediction performance of the ammonia emission amount with 83 data. The ammonia emission rate included eleven inputs and two outputs (maximum ammonia loss, N_max and time to reach half of N_max, K_m). Categorical input variables were transformed into multi-dimensional equal-distance variables, and 13 data were added into 66 training data using a generative adversarial network. Hyperparameters (number of layers, number of neurons, and activation function) of ANN were optimized using Gaussian process. Using 17 test data, the previous ANN model (Lim et al., 2007) showed the mean absolute error (MAE) of K_m and N_max to 0.0668 and 0.1860, respectively. The present ANN outperformed the previous model, reducing MAE by 38% and 56%.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.10 no.2
• /
• pp.133-144
• /
• 2006

In order to catch out such Bond Strength, the preceding researchers had ever examined the Bond Strength of FRP Plate through their experimentations by setting up of various fluent. However, since the experiment for research on such Bond Strength takes much of expenditure for equipment structure and time-consuming, also difficult to carry out, it is conducting limitedly. This Study purposes to develop the most suitable Artificial Neural Network Model by application of various Neural Network Model and Algorithm to the adhering experiment data of the preceding researchers. Output Layer of Artificial Neural Network Model, and Input Layer of Bond Strength were performed the learning by selection as the variable of the thickness, width, adhered length, the modulus of elasticity, tensile strength, and the compressive strength of concrete, tensile strength, width, respectively. The developed Artificial Neural Network Model has applied Back-Propagation, and its error was learnt to be converged within the range of 0.001. Besides, the process for generalization has dissolved the problem of Over-Fitting in the way of more generalized method by introduction of Bayesian Technique. The verification on the developed Model was executed by comparison with the resulted value of Bond Strength made by the other preceding researchers which was never been utilized to the learning as yet.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.20 no.3
• /
• pp.543-560
• /
• 2018

With the increased development in downtown underground space facilities that vertically cross under a railway at a shallow depth, the demand for non-open cut method is increasing. However, most construction sites still adopt the pipe roof method, where medium and large diameter steel pipes are pressed in to form a roof, enabling excavation of the inside space. Among the many factors that influence the loosening region and loads that occur while pressing in steel pipes, the size of the pipe has the largest impact, and this factor may correspond to the magnitude of load applied to the underground structure inside the steel pipe roof. The super equilibrium method (SEM) has been developed to minimize ground disturbance and loosening load, and uses small diameter pipes of approximately 114 mm instead of conventional medium and large diameter pipes. This small diameter steel pipe is called an SEM pile. After SEM piles are pressed in and the grouting reinforcement is constructed, a crossing structure is pressed in by using a hydraulic jack without ground subsidence or heaving. The SEM pile, which plays the role of timbering, is a fore-poling pile of approximately 5 m length that prevents ground collapse and supports surface load during excavation of toe part. The loosening region should be adequately calculated to estimate the spacing and construction length of the piles and stiffness of members. In this paper, we conducted a comparative analysis of calculations of loosening load that occurs during the press-in of SEM pile to obtain an optimal design of SEM. We analyzed the influence of factors in main theoretical and empirical formulas applied for calculating loosening regions, and carried out FEM analysis to see an appropriate loosening load to the SEM pile. In order to estimate the soil loosening caused by actual SEM-pile indentation and excavation, a steel pipe indentation reduction model test was conducted. Soil subsidence and soil loosening were investigated quantitatively according to soil/steel pipe (H/D).

• Composites Research
• /
• v.14 no.3
• /
• pp.18-31
• /
• 2001

Interfacial and microfailure properties of the modified steel, carbon and glass fibers/cement composites were investigated using electro-pullout test under tensile and compressive tests with acoustic emission (AE). The hand-sanded steel composite exhibited higher interfacial shear strength (IFSS) than the untreated and even neoalkoxy zirconate (Zr) treated steel fiber composites. This might be due to the enhanced mechanical interlocking, compared to possible hydrogen or covalent bonds. During curing process, the contact resistivity decreased rapidly at the initial stage and then showed a level-off. Comparing to the untreated case, the contact resistivity of either Zr-treated or hand-sanded steel fiber composites increased to the infinity at latter stage. The number of AE signals of hand-sanded steel fiber composite was much more than those of the untreated and Zr-treated cases due to many interlayer failure signals. AE waveforms for pullout and frictional signals of the hand-sanded composite are larger than those of the untreated case. For dual matrix composite (DMC), AE energy and waveform under compressive loading were much higher and larger than those under tensile loading, due to brittle but well-enduring ceramic nature against compressive stress. Vertical multicrack exhibits fur glass fiber composite under tensile test, whereas buckling failure appeared under compressive loading. Electro-micromechanical technique with AE can be used as an efficient nondestructive (NDT) method to evaluate the interfacial and microfailure mechanisms for conductive fibers/brittle and nontransparent cement composites.

• Journal of Animal Science and Technology
• /
• v.46 no.4
• /
• pp.593-602
• /
• 2004

A feeding trial was carried out to investigate the effect of supplemental complex probiotics on performances, physio-chemica1 properties of meat and inetestinal microflora in broiler chicks. Four hundred eighty broiler chickens, one days old with mixed sexes were fed one of four diets containing 0, 0.1, 0.2 and 0.4% complex probiotics for 7 weeks. There were four replicates with thirty chicks per pen. Diet contained ME 3,100, 3,l00kcal/kg, and CP 22.0, 20.0% for starting and finishing period, respectively. Body Weight gain of chicks fed the complex probiotics tended to increase from the frist week and all complex probiotics higher than control from the 4th week. Chickens fed the diets containing 0.2% probiotics had higher(P<0.05) than those fed the other levels from the 4th week to 5th week. Feed conversion also improved significantly(P<0.05) in the supplemental 0.2% probiotics from the 4th week to 5th week. In physio-chemica1 properties of meat, carcass rate increased significantly(P<0.05) in the supplemental 0.4% probiotics compared to that of control at 7 weeks overall means and abdominal fat pad rate increased significantly(P< 0.05) in the supplemental 0.2% probiotics compared to that of control. Cooking loss decreased significantly(P<0.05) in the supplemental all probiotics. But shear force increased significantly(P<0.05) in the supplemental 0.4% probiotics. The number of ileum and cecum Lactobacillus spp. tended to increase in the supplemental complex probiotics at 7 week of age, but was not significantly different. As the result, supplemental complex probiotics increased performance and physio-chemica1 properties of meat and the number of intestinal Lactobacillus of broiler chicks.