Figure 3 - uploaded by Paul Ayegba
Content may be subject to copyright.
Source publication
Multilayer Perceptron (MLP) models have been developed to predict two-phase average void fraction and probability density function (PDF) of void fraction in 90 o bends. The Artificial Neural Network (ANN) methodology was reported using MLP trained with 2 algorithms. Logarithmic sigmoid transfer function was used in a single hidden layer for both al...
Context in source publication
Context 1
... order algorithms has the inherent disadvantage of being slow, but also has a unique advantage, in that the gradient term can be minimised to zero with sufficient amount of iterations. Figure 3 and 4 depict flow charts of the gradient descent and Levenberg-Marquardt algorithms respectively that were developed here. ...
Similar publications
The annual increasing use of motorized vehicles requires improvements in infrastructure to support the reliability of urban transport systems in Surakarta City. The Government of Surakarta City has attempted to improve the infrastructure of the Gilingan Viaduct by creating an underpass. The traffic performance of the model developed is analysed usi...
One model structural deficiency is that some dynamic characteristics (such as seasonal dynamics) in catchment conditions are not explicitly represented by hydrological models. This study integrates data mining techniques to develop a clustering preprocessing framework for the subannual calibration of hydrological models to simulate seasonal dynamic...
Short-term traffic estimations have a significant influence in terms of effectively controlling vehicle traffic. In this study, short-term traffic forecasting models have been developed based on different approaches. Seasonal autoregressive integrated moving average (SARIMA), artificial bee colony (ABC) and differential evolution (DE) algorithms ar...
The environment where ship operates defines its hydrodynamic characteristic. Interaction between the ship and boundaries around the regions particularly characterised the ship steer-ability and manoeuvrability. Despite the fact that ships are inherently designed to navigate in forward direction in open sea, however inevitably ships have to sometime...
Protection against jamming is more crucial than ever before as the world has become increasingly dependent on wireless communications. Moreover, jamming technology is cheap and simple as 1-click ordering; further, it has the ability to neutralize communications. This study developed a model that responds to reactive jamming in wireless ad hoc netwo...
Citations
... The developed ANN approach in this study solves the relation featuring an alternative solution between strain gauges and MTS results. Fig. 1 is a general form of the algorithm [11]. The algorithm evaluates results between gradient descent and Gauss-Newton update where the method uses J, λ, W, h as in formation in (1). ...
... Co-current flow of oil and water in pipeline is common in chemical, process and petroleum industry operations [1]. Data on flow development/redevelopment, pressure drop, flow pattern, phase mixing and interfacial characteristics are essential for optimal design and operation in the industries as well as for modelling [2][3][4]. In petroleum exploration and transport, connate water or injected water from enhanced oil recovery operations flows along with oil. ...
Pressure drop and flow pattern of oil–water flows were investigated in a 19-mm ID clear polyvinyl chloride pipe consisting of U -bend with radius of curvature of 100 mm. The range for oil and water superficial velocities tested was $$0.04 \le U_{{{\text{so}}}} \le 0.950 \;{\text{m/s}}$$ 0.04 ≤ U so ≤ 0.950 m/s and $$0.13 \le U_{{{\text{sw}}}} \le 1.10 \;{\text{m/s}}$$ 0.13 ≤ U sw ≤ 1.10 m/s , respectively. Measurements were carried out under different flow conditions in a test section that consisted of four different parts: upstream of the bend, at the bend and at two redeveloping flow locations after the bend. The result indicated that the bend had limited influence on downstream flow patterns. However, the shear forces imposed by the bend caused some shift flow pattern transition and bubble characteristics in the redeveloping flow section after the bend relative to develop flow before the bend. Generally, pressure gradient at all the test sections increased with both oil fraction and water superficial velocity and there was a sharp change of pressure gradient profile during phase inversion. The transition point where phase inversion occurred was always within the range of $$0.4 \le U_{{{\text{sw}}}} \le 0.54 \;{\text{m/s}}$$ 0.4 ≤ U sw ≤ 0.54 m/s . Pressure losses differed at the various test sections, and the difference was strongly linked to the superficial velocity of the phases and the flow pattern. At high mixture velocity, pressure losses at the redeveloping section after the bend were higher than that at the bend and that for fully developed flows. At low mixture velocity, pressure losses at the bend are higher than in the straight sections. Pressure drop generally decreased with level of flow development downstream of the bend.
... Co-current flow of oil and water in pipeline is common in chemical, process and petroleum industry operations [1]. Data on flow development/redevelopment, pressure drop, flow pattern, phase mixing and interfacial characteristics are essential for optimal design and operation in the industries as well as for modelling [2]- [4]. In petroleum exploration and transport, connate water or injected water from enhanced oil recovery operations flow along with oil. ...
Pressure drop and flow pattern of oil-water flows were investigated in a 19 mm ID clear polyvinyl chloride pipe consisting of U-bend with radius of curvature of 100 mm. The range for oil and water superficial velocities tested were and respectively. Measurements were carried out under different flow conditions in a test section that consisted of four different parts: upstream of the bend, at the bend and at two redeveloping flow locations after the bend. The result indicated that the bend had limited influence on downstream flow patterns. However, the shear forces imposed by the bend caused some shift flow pattern transition and bubble characteristics in the redeveloping flow section after the bend relative to develop flow before the bend. Generally, pressure gradient at all the test sections increased with both oil fraction and water superficial velocity and there was a sharp change of pressure gradient profile during phase inversion. The transition point where phase inversion occurred was always within the range of . Pressure losses differed at the various test sections and the difference was strongly linked to the superficial velocity of the phases and the flow pattern. At high mixture velocity, pressure losses at the redeveloping section after the bend were higher than that at the bend and that for fully developed flows. At low mixture velocity, pressure losses at the bend are higher than in the straight sections. Pressure drop generally decreased with level of flow development downstream of the bend.
... It was claimed that the ANN gave excellent prediction of pressure drop across U-bends. Ayegba et al. [25], and Ayegba et al. [26], used ANN methodology to predict void fraction and flow patterns of gas-liquid flows in a 90 bend at the top of a riser. The ANN models gave excellent predictions and also provided additional data beyond the limits of their experimental measurements. ...
... Searching for optimal-weights, while training the neural network, is aimed at minimizing a cost representative function with respect to the data used for model training. The mathematical background is well published in open literature [24,25,37,38]. It should be mentioned that a number of network topologies are available in neural networks but in this paper, the back-propagation technique with feed-forward algorithms was chosen given its performance in previous works [25,26,[39][40][41][42][43]. ...
... The mathematical background is well published in open literature [24,25,37,38]. It should be mentioned that a number of network topologies are available in neural networks but in this paper, the back-propagation technique with feed-forward algorithms was chosen given its performance in previous works [25,26,[39][40][41][42][43]. ...
This research investigates the effect of small amounts of drag-reducing polymer (DRP), (Partially hydrolysed polyacrylamide) on single phase water flow in 180° bends using both experimental and artificial neural network (ANN) modelling approach. The results show that, in both straight pipes and U-bend, drag reduction (DR) generally increased with polymer concentration until an optimum concentration is reached beyond which further increase in concentration resulted in increased drag. It was observed that using different concentrations of master solution had negligible effect on the computed DR as long as there is sufficient mixing to allow for homogeneous polymer dispersion. The results further revealed that DR in straight pipes are significantly higher than that in bends of equivalent length and the highest DR reported were 65%, 42% and 45% in straight, horizontal U-bend and down-flow U-bend pipe configurations respectively. It was also reported here that DR increased with Reynolds number for all flow configurations investigated. This study was also extended to ANN modelling of the various output parameters of interest such as friction factor and bend friction coefficient. Some test cases showed that one neuron was sufficient in predicting the output parameters. This suggests that classical models would also be adequate in modelling single-phase flow drag reduction in and around bends. Model predictions were excellent with cross correlation of nearly 100% (99.99%). In this work, the ANN model coded using Levenberg–Marquardt LM algorithm and logsigmoid transfer function gave the best Neural Network model performance as indicated by its lowest value of X² (=2.03E-04). The proposed ANN model can be used with confidence for the prediction of friction factor, bends coefficient and drag reduction DR in straight pipes and return bends.
Novel results of drag reduction (DR) in and around U-bends is presented here. Three polymer types were tested and the radii of curvature of the U-bends were 100 and 200-mm with pipe diameter of 19-mm. DR increased with polymer concentration and flow rate up to certain thresholds. Measured drag reduction was proportional to the Weissenberg number of the drag-reducing solution flow. DR differed among the different sections considered with the highest value recorded for developed flows upstream of the bend and the least values recorded in the bend. DR increased with flow development after the bend and increased with polymer molecular weight before and after the bend. The effect of bend curvature ratio on DR was predominant after the U-bend. For R = 100-mm, the highest DR recorded upstream of bend, in the bend, immediately after and further downstream of the bend were 57, 31, 36 and 33% respectively.
