Introduction
Lateritic soils are commonly used for road construction in Nigeria. Lateritic soil in its natural state sometimes have low bearing capacity and low strength due to high content of clay. When lateritic soil contains a huge amount of clay materials, its strength and stability cannot be guaranteed under load in presence of moisture[1]. Also, when a weak soil is encountered on a site and sourcing for alternative soil proves economically unviable, improving the soil by way of stabilization to meet the desired objective becomes the viable option [2, 3, 4]. Popular industrial stabilizers are cement, lime, flyash and bitumen. Furthermore, the high cost of cement being used as binder, has led to the search for natural materials as either alternative or complement. Research on alternative or complement to cement has so far centered on the partial replacement of cement with different materials[5]. The improvement in the strength and durability of lateritic soil in recent times has become imperative, this has geared researchers towards using stabilizing materials that can be sourced locally at a very low cost [1]. These local materials can be grouped as either agricultural or industrial wastes [6]. In light of this, cheap agricultural waste such as rice husk ash is being studied as replacements for the more expensive cement.
Rice is one of the most cultivated and consumed cereal in the world. In rice producing countries, a traditional waste material known as ‘rice husk’ is obtained as a by-product in bulk amount from Rice mills. Globally, approximate 600 million tonnes of rice paddies are produced each year. On the average, 20% of the paddy is husk, giving an annual total production of 120 million tonnes. [7,8]. Rice husk is a by-product from agriculture produce when it is harvested, the outermost part of the paddy is the rice husk, also called the rice hull. It is separated from the brown rice in rice milling. Complete burning of rice husk results to rice husk ash (RHA), so for every 1000 kg of paddy milled, about 220 kg (22%) of husk is produced and when the husk is burnt in boilers, about 55 kg (25%) of RHA is generated, if the burning is incomplete, the carbonized rice husk (CRH) is obtained[9]. In line with the Federal Government’s drive for Nigeria to be self-sufficient in rice production and to save hundreds of billions of naira annually on rice importation, local rice production has reached 15 million metric tonnes[10], this, in turn has resulted to increased generation of rice husk as waste. Stabilization enhances the desired qualities of a soil [11], chiefly among these, is the California bearing ratio (CBR), a standard for measuring the strength of a given soil in road construction. The design of flexible pavement is much dependent on the CBR of subgrade. CBR values can be measured in the laboratory test in accordance with BS 1377[12 13,14]. The CBR test performed in the laboratory is time-consuming, a laboratory test generally takes four or more days to measure the soaked CBR value for each soil sample. The result of the tests is actually an indirect measure, which represents comparison of the strength of subgrade material to the strength of standard crushed rock referred in percentage values. Instead, it can be predicted from properties of soils determined in the laboratories. Several studies have been conducted to estimate CBR from Liquid limit, Plasticity Index, standard proctor parameters, the use of artificial neural network has so far held high promises in achieving this [15, 16]. In light of this, developing credible predictive models has been in the fore-burner, the use of artificial neural network (ANN) is fast gaining ground especially in the field of geotechnical engineering [17, 18, 19,20].
[15]. ANN is a massively parallel –distributed information processing system that has certain performance characteristics resembling biological neural networks of human brain[21]. ANNs have been developed as a generalization of mathematical models of human cognition or neural biology. The key element of ANN is the novel structure of its information processing system. An ANN is composed of a large number of highly interconnected processing elements called neurons working in unison to solve specific problems. Neurons having similar characteristics in an ANN are arranged in groups called layers. A way of classifying neural networks is by the number of layers as single, bilayer and multilayer. ANNs can also be categorized based on direction of information flow and processing The main benefits of the ANN according to Khademikia et al.,[22] in comparison to other modeling programs are the nonlinearity, adaptively, fault tolerance, uniformity and design.
In recent times, the increase in population has led to the generation of more wastes such as the rice husk, thereby necessitating the need for proper management of these wastes. It is however worthy of note that there is yet to be adequate awareness on the usefulness of these aforementioned wastes in Nigeria, in other words, little or no importance is attached to them. The practice of incinerating them to ash and adopting them as admixtures in stabilized soils due to their pozzolanic values has enhanced their economic value[23]. Also, for greater efficiency in management of time and manpower, the need for developing models in predicting California bearing ratio (CBR) an important parameter in road construction has become imperative. This is expected to address the problems of unnecessary delays at the laboratories and human errors, which can negatively impact on the project.
Materials and Methods
Materials The materials used were: rice husk ash, soil samples, ordinary Portland cement and water.
Rice Husk Ash (RHA): The rice husk ash was burnt in open atmosphere and the black ashes obtained were heated in an air tight furnace for 6 hours at 10000C to obtain a white coloured ash.
Soil Sample: Soil Sample was collected along Oye-Ekiti – Isan-Ekiti Road, Nigeria at a depth not less than 1.2m below the ground level at 5 different points of about 3m apart using the disturbed sampling technique. It was brought to the soil laboratory and marked indicating the soil description, sampling depth and date of sampling. The soil sample was air-dried for two weeks to allow for partial elimination of natural water content which may affect the analysis, then sieved with sieve no 4 (4.75mm opening) to obtain the final soil sample for the tests. After the drying period of two weeks, lumps in the sample were pulverized under minimal pressure.
Ordinary Portland Cement: This was obtained from a cement store.
Water: The water used was obtained from the running taps in the laboratory, the source was borehole. Distilled water was not used so as to obtain results that would reflect in-situ conditions.