Detection of fouling in a cross-flow heat exchanger using a neural network based technique

Lalot, Sylvain; Pálsson, Halldór

doi:10.1016/j.ijthermalsci.2009.10.011

Cited by 48 publications

(13 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although this approach has also been applied to water filtration and thermal investigations of fouling, for example, to predict fouling in the filtration of drinking water [18], heat exchangers [19], and heaters [20], transfer learning and deep CNN have not been applied to biofouling of marine structures such as ships' hulls. Notably, the above-mentioned classification results yielded more than 70% accuracy after using the transfer learning approach on CNN.…”

Section: Introductionmentioning

confidence: 99%

Intelligent Image Recognition System for Marine Fouling Using Softmax Transfer Learning and Deep Convolutional Neural Networks

Chin

Clare

2017

Complexity

View full text Add to dashboard Cite

The control of biofouling on marine vessels is challenging and costly. Early detection before hull performance is significantly affected is desirable, especially if "grooming" is an option. Here, a system is described to detect marine fouling at an early stage of development. In this study, an image of fouling can be transferred wirelessly via a mobile network for analysis. The proposed system utilizes transfer learning and deep convolutional neural network (CNN) to perform image recognition on the fouling image by classifying the detected fouling species and the density of fouling on the surface. Transfer learning using Google's Inception V3 model with Softmax at last layer was carried out on a fouling database of 10 categories and 1825 images. Experimental results gave acceptable accuracies for fouling detection and recognition.

show abstract

Section: Introductionmentioning

confidence: 99%

Intelligent Image Recognition System for Marine Fouling Using Softmax Transfer Learning and Deep Convolutional Neural Networks

Chin

Clare

2017

Complexity

View full text Add to dashboard Cite

show abstract

“…This phenomenon, called fouling, affect the equipment operation by reducing their thermal effectiveness and by involving a considerable pressure drops. This leads to significant economic losses due to the pumping and the frequent cleaning of the installations [2,3]. Fouling in heat transfer systems is often unavoidable and reduces energy efficiency and plant operability.…”

Section: Introductionmentioning

confidence: 99%

The fouling in the tubular heat exchanger of Algiers refinery

2015

View full text Add to dashboard Cite

List of symbols AThe heat transfer external surface (m 2 ) a ⊥ , a c⊥ The fluid passage area at side tube and side calender, respectively (m 2 ) C P The heat-storage capacity (kJ/kg °C) C Pf , Cp c The heat-storage capacity (kJ/kg °C) d The fluid circulating density at side tube di Tubes interns diameter (m) d 0,e Diamètre externe des tubes (m) d 4 15 Density D c , D e Diamètre de la calandre et diamètre hydraulique, respectivement (m) f t The friction factor F The correction factor G t The mass speed fluid circulating at side tube (kg/h m 2 ) h i The heat transfer coefficient at internal film (kW/m 2 °C) h 0 The heat transfer coefficient at external film (kW/m 2 °C) h i0 The heat transfer coefficient at internal film brought back to the external surface (kW/ m 2 °C) J hf , J hc Coefficient de transfert pour le fluide froid et chaud, respectivement (kW/m 2 °C) l Tubes length (m) m The cold fluid mass flow (the crude oil) (m 3 /h) m f , m c Débit massique pour le fluide froid et chaud, respectivement (kg/s) m f⊥ , m c⊥ Mass flow for the cold and hot fluid, respectively, concerning the straight section (kg/s) MLDTThe difference logarithmic temperature (°C) n c , n tThe devices and the master keys numbers P e , P s Inlet and outlet pressure of the crude oil respectively (Pascal)Abstract Crude oil fouling in refinery preheat exchangers is a chronic operational problem that compromises energy recovery in these systems. Progress is hindered by the lack of quantitative knowledge of the dynamic effects of fouling on heat exchanger transfer and pressure drops. In subject of this work is an experimental determination of the thermal fouling resistance in the tubular heat exchanger of the crude oil preheats trains installed in an Algiers refinery. By measuring the inlet and outlet temperatures and mass flows of the two fluids, the overall heat transfer coefficient has been determined. Determining the overall heat transfer coefficient for the heat exchanger with clean and fouled surfaces, the fouling resistance was calculated. The results obtained from the two cells of exchangers studies, showed that the fouling resistance increased with time presented an exponential evolution in agreement with the model suggested by Kern and Seaton, with the existence of fluctuation caused by the instability of the flow rate and the impact between the particles. The bad cleaning of the heat exchangers involved the absence of the induction period and caused consequently, high values of the fouling resistance in a relatively short period of time.

show abstract

“…One of them has to do with the selection of a proper model structure. Indeed, looking closer at literature, most of the developments dedicated to heat exchanger identification deal with linear time-invariant (LTI) models or black-box artificial neural networks [7], [15], [17], [31], [37]. Now, when thermodynamics or heat transfer laws are studied [23], it can be shown that important physical parameters, such as the heat transfer coefficients, are directly related to specific state vector components such as the inlet mass flow rates [9], [11].…”

Section: Introductionmentioning

confidence: 99%

“…This simulator is similar to those used in [14] or in [28]. It was also used in [17] with minor changes in the numerical discretization procedure. This simulator was validated for steady-state conditions in [17] and transient results also showed a good agreement with the expected results.…”

Section: Introductionmentioning

confidence: 99%

“…It was also used in [17] with minor changes in the numerical discretization procedure. This simulator was validated for steady-state conditions in [17] and transient results also showed a good agreement with the expected results. Here it is assumed that the numerical model can simulate the transient behavior of different inputs and properties with sufficient accuracy, based on the steady-state validation, which is necessary when mimicking real inflow situations with and without fouling.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Continuous-Time Linear Parameter-Varying Identification of a Cross Flow Heat Exchanger: A Local Approach

Mercère

Pálsson

Poinot

2011

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

In this paper, the problem of deriving a dynamical model of a cross flow heat exchanger is considered. In order to take into account the dependency of the system's dynamics on the hot and the cold mass flow rates in an explicit way, an input-output linear parameter-varying (LPV) model is used. A local approach composed of three steps is carried out to identify this LPV model. A parameter estimation scheme is introduced in which cost functions are minimized by using specific nonlinear programming methods. In this study, a finite volume physical model simulator is exploited to simulate and to generate the data. Simulations are performed to demonstrate the benefits of the suggested approach.Index Terms-Cross flow heat exchanger, interpolation, linear parameter-varying (LPV) model, local approach, numerical modeling, output-error method.

show abstract

Detection of fouling in a cross-flow heat exchanger using a neural network based technique

Cited by 48 publications

References 12 publications

Intelligent Image Recognition System for Marine Fouling Using Softmax Transfer Learning and Deep Convolutional Neural Networks

Intelligent Image Recognition System for Marine Fouling Using Softmax Transfer Learning and Deep Convolutional Neural Networks

The fouling in the tubular heat exchanger of Algiers refinery

Continuous-Time Linear Parameter-Varying Identification of a Cross Flow Heat Exchanger: A Local Approach

Contact Info

Product

Resources

About