Mohd Nur Asmawisham Alel

Estimation of air-overpressure produced by blasting operation through a neuro-genetic technique

Blasting operations usually produce significant environmental problems which may cause severe damage to the nearby areas. Air-overpressure (AOp) is one of the most important environmental impacts of blasting operations which needs to be predicted and subsequently controlled to minimize the potential risk of damage. In order to solve AOp problem in Hulu Langat granite quarry site, Malaysia, three non-linear methods namely empirical, artificial neural network (ANN) and a hybrid model of genetic algorithm (GA)–ANN were developed in this study. To do this, 76 blasting operations were investigated and relevant blasting parameters were measured in the site. The most influential parameters on AOp namely maximum charge per delay and the distance from the blast-face were considered as model inputs or predictors. Using the five randomly selected datasets and considering the modeling procedure of each method, 15 models were constructed for all predictive techniques. Several performance indices including coefficient of determination (R2), root mean square error and variance account for were utilized to check the performance capacity of the predictive methods. Considering these performance indices and using simple ranking method, the best models for AOp prediction were selected. It was found that the GA–ANN technique can provide higher performance capacity in predicting AOp compared to other predictive methods. This is due to the fact that the GA–ANN model can optimize the weights and biases of the network connection for training by ANN. In this study, GA–ANN is introduced as superior model for solving AOp problem in Hulu Langat site.

Optimizing Blasting’s Air Overpressure Prediction Model using Swarm Intelligence

Air overpressure (AOp) resulting from blasting can cause damage and nuisance to nearby civilians. Thus, it is important to be able to predict AOp accurately. In this study, 8 different Artificial Neural Network (ANN) were developed for the purpose of prediction of AOp. The ANN models were trained using different variants of Particle Swarm Optimization (PSO) algorithm. AOp predictions were also made using an empirical equation, as suggested by United States Bureau of Mines (USBM), to serve as a benchmark. In order to develop the models, 76 blasting operations in Hulu Langat were investigated. All the ANN models were found to outperform the USBM equation in three performance metrics; root mean square error (RMSE), mean absolute percentage error (MAPE) and coefficient of determination (R2). Using a performance ranking method, MSO-Rand-Mut was determined to be the best prediction model for AOp with a performance metric of RMSE=2.18, MAPE=1.73% and R2=0.97. The result shows that ANN models trained using PSO are capable of predicting AOp with great accuracy.

Mapping on Slope Seepage Problem using Electrical Resistivity Imaging (ERI)

The stability of slope may influenced by several factors such as its geomaterial properties, geometry and environmental factors. Problematic slope due to seepage phenomenon will influenced the slope strength thus promoting to its failure. In the past, slope seepage mapping suffer from several limitation due to cost, time and data coverage. Conventional engineering tools to detect or mapped the seepage on slope experienced those problems involving large and high elevation of slope design. As a result, this study introduced geophysical tools for slope seepage mapping based on electrical resistivity method. Two spread lines of electrical resistivity imaging were performed on the slope crest using ABEM SAS 4000 equipment. Data acquisition configuration was based on long and short arrangement, schlumberger array and 2.5 m of equal electrode spacing interval. Raw data obtained from data acquisition was analyzed using RES2DINV software. Both of the resistivity results show that the slope studied consists of three different anomalies representing top soil (200 – 1000 Ωm), perched water (10 – 100 Ωm) and hard/dry layer (> 200 Ωm). It was found that seepage problem on slope studied was derived from perched water zones with electrical resistivity value of 10 – 100 Ωm. Perched water zone has been detected at 6 m depth from the ground level with varying thickness at 5 m and over. Resistivity results have shown some good similarity output with reference to borehole data, geological map and site observation thus verified the resistivity results interpretation. Hence, this study has shown that the electrical resistivity imaging was applicable in slope seepage mapping which consider efficient in term of cost, time, data coverage and sustainability.

Estimating SPT-N Value Based on Soil Resistivity using Hybrid ANN-PSO Algorithm

Standard Penetration Resistance (N value) is used in many empirical geotechnical engineering formulas. Meanwhile, soil resistivity is a measure of soils resistance to electrical flow. For a particular site, usually, only a limited N value data are available. In contrast, resistivity data can be obtained extensively. Moreover, previous studies showed evidence of a correlation between N value and resistivity value. Yet, no existing method is able to interpret resistivity data for estimation of N value. Thus, the aim is to develop a method for estimating N-value using resistivity data. This study proposes a hybrid Artificial Neural Network-Particle Swarm Optimization (ANN-PSO) method to estimate N value using resistivity data. Five different ANN-PSO models based on five boreholes were developed and analyzed. The performance metrics used were the coefficient of determination, R2 and mean absolute error, MAE. Analysis of result found that this method can estimate N value (R2 best=0.85 and MAEbest=0.54) given that the constraint, , is satisfied. The results suggest that ANN-PSO method can be used to estimate N value with good accuracy.

Integral Analysis of Seismic Refraction and Ambient Vibration Survey for Subsurface Profile Evaluation

Geotechnical site investigation as known as subsurface profile evaluation is the process of subsurface layer characteristics determination which finally used for design and construction phase. Traditionally, site investigation was performed using drilling technique thus suffers from several limitation due to cost, time, data coverage and sustainability. In order to overcome those problems, this study adopted surface techniques using seismic refraction and ambient vibration method for subsurface profile depth evaluation. Seismic refraction data acquisition and processing was performed using ABEM Terraloc and OPTIM software respectively. Meanwhile ambient vibration data acquisition and processing was performed using CityShark II, Lennartz and GEOPSY software respectively. It was found that studied area consist of two layers representing overburden and bedrock geomaterials based on p-wave velocity value (vp = 300 – 2500 m/s and vp > 2500 m/s) and natural frequency value (Fo = 3.37 – 3.90 Hz) analyzed. Further analysis found that both methods show some good similarity in term of depth and thickness with percentage accuracy at 60 – 97%. Consequently, this study has demonstrated that the application of seismic refractin and ambient vibration method was applicable in subsurface profile depth and thickness estimation. Moreover, surface technique which consider as non-destructive method adopted in this study was able to compliment conventional drilling method in term of cost, time, data coverage and environmental sustainaibility.