Incorporation of prior knowledge in neural network model for continuous cooling of steel using genetic algorithm

Chakraborty, Subhamita; Chattopadhyay, Partha; Ghosh, S. K.; Datta, Someswar

doi:10.1016/j.asoc.2017.05.001

Cited by 40 publications

(15 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As reported earlier (Chakraborty et al , 2017), the correlation coefficient value is quite good in ANN-BP model, but it fails to make feasible predictions from the materials standpoint. In some of the predicted CCT diagram of the ANN-BP model, pearlite start temperature goes above ferrite finish temperature and pearlite finish temperature goes below bainite start temperature which is not possible during transformation of steel, as explained earlier.…”

Section: Ann Modeling Of Cct Diagram and Its Problemsmentioning

confidence: 62%

“…To improve the prediction of the ANN-BP models through the incorporation of knowledge, an attempt has been made to modify the weights and biases of ANN-BP in a multi-objective optimization-based mode. For this optimization, GA is employed (Chakraborty et al , 2017; Venkatesh et al , 2013; Xu et al , 2012; Srinivasu and Babu, 2008). GA is a search algorithm based on the mechanics of the natural selection process (biological evolution).…”

Section: Modeling Approachesmentioning

confidence: 99%

“…Artificial neural network (ANN) (Haykin, 1999; Fausett, 2008) is a proven tool for developing predictive models for complex material systems (Datta and Chattopadhyay, 2013; Datta, 2016; Sinha et al , 2013; Kondo et al , 2017), including steel (Mohanty et al , 2009; Kundu et al , 2009; Ghosh et al , 2009; Lu, van der Zwaag and Xu, 2017; Razavi et al , 2016). For designing new alloys with targeted microstructure, various attempts have been made in predicting austenite transformation using ANN models with chemical composition, austenitizing temperature and CR as input parameters for transformation (Trzaska and Dobrza’nski, 2007; Wang et al , 1999; Dobrza’nski and Trzaska, 2004; Wei et al , 2007; Vermeulen et al , 1996; Chakraborty et al , 2017). From the perspective of the domain knowledge of materials engineering, data-driven modeling has an inherent problem.…”

Section: Introductionmentioning

confidence: 99%

“…Then the model weights and biases are modified using multi-objective optimization using genetic algorithm (GA) (Deb, 2010), where the prediction error of the model is considered as one objective, whereas one of the factors in the prediction required avoiding predictions infeasible from metallurgy stand point. Some other such factors are considered as constraints in the optimization process (Chakraborty et al , 2017). GA has been successfully used in the area of materials engineering by several workers (Datta and Chattopadhyay, 2013; Datta, 2016; Chakraborti, 2004; Phan et al , 2015; Nandi et al , 2012; Das et al , 2009) for single or multi-objective optimization problems (MOOPs), mostly for designing composition and/or processing of materials.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

MCDM towards knowledge incorporation in ANN models for phase transformation in continuous cooling of steel

Chakraborty

Das²,

Kaveti

et al. 2019

MMMS

Self Cite

View full text Add to dashboard Cite

Purpose The purpose of this paper is to incorporate prior knowledge in the artificial neural network (ANN) model for the prediction of continuous cooling transformation (CCT) diagram of steel, so that the model predictions become valid from materials engineering point of view. Design/methodology/approach Genetic algorithm (GA) is used in different ways for incorporating system knowledge during training the ANN. In case of training, the ANN in multi-objective optimization mode, with prediction error minimization as one objective and the system knowledge incorporation as the other, the generated Pareto solutions are different ANN models with better performance in at least one objective. To choose a single model for the prediction of steel transformation, different multi-criteria decision-making (MCDM) concepts are employed. To avoid the problem of choosing a single model from the non-dominated Pareto solutions, the training scheme also converted into a single objective optimization problem. Findings The prediction results of the models trained in multi and single objective optimization schemes are compared. It is seen that though conversion of the problem to a single objective optimization problem reduces the complexity, the models trained using multi-objective optimization are found to be better for predicting metallurgically justifiable result. Originality/value ANN is being used extensively in the complex materials systems like steel. Several works have been done to develop ANN models for the prediction of CCT diagram. But the present work proposes some methods to overcome the inherent problem of data-driven model, and make the prediction viable from the system knowledge.

show abstract

Section: Ann Modeling Of Cct Diagram and Its Problemsmentioning

confidence: 62%

Section: Modeling Approachesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MCDM towards knowledge incorporation in ANN models for phase transformation in continuous cooling of steel

Chakraborty

Das²,

Kaveti

et al. 2019

MMMS

Self Cite

View full text Add to dashboard Cite

show abstract

“…The artificial neural network obtained an error of 7,7 % making it a reliable mechanism. In Chakraborty (2017), an artificial neural network model was developed for predicting the phase transformation of steel by austenite, and thus the construction of the continuous cooling transformation diagram. Within the research is used a multi-layer artificial neural network and its weights are defined by a genetic algorithm.…”

Section: Introductionmentioning

confidence: 99%

System for the recognition of wear patterns on microstructures of carbon steels using a multilayer perceptron

Santoyo¹,

López²,

Mendiola³

et al. 2018

Ing. Inv.

View full text Add to dashboard Cite

This paper describes the application of a recognition system wear patterns present in carbon steel, the system classifies the microstructure of the materials which have three conditions throughout life-time in thermoelectric plants. This approach employs the artificial neural network multilayer perceptron in conjunction with the digital image processing to recognize the different physical states of the materials used as conductors in conditions of high temperatures. The studied patterns in the microstructure are spheronization, decarburization and graphitization. The microstructure is revealed from microscope images obtained in the Testing Laboratory Equipment and Materials of the Federal Electricity Commission in Mexico (LAPEM-CFE). The proposed system compared to the human expert, obtained an accuracy of 96.83 % with a shorter analysis time and inspection cost.

show abstract

Prediction of Time–Temperature–Transformation Diagrams of NiAl Alloy: An Evaluation of Intelligent Algorithms

Hernández‐Flores,

Talavera‐Rivera,

García‐Moreno

2024

Adv Eng Mater

View full text Add to dashboard Cite

Time‐temperature‐transformation diagrams are essential in the field of metallurgy. However, constructing these diagrams through empirical and simulating methods can be time‐consuming and expensive. Therefore, there is a need to develop accurate and affordable alternatives that can rapidly predict TTT diagrams. The present study offers a novel analysis of multiple algorithms for predicting TTT diagrams, as well as an examination of various data preprocessing techniques (data augmentation ‐ DA and exploratory data analysis ‐ EDA). A database is created by integrating multiple data sources and increased using DA by ensuring the synthetic data remains physically realistic and meaningful. Subsequently, an EDA is performed to prepare the data before its use in training. The influence of each hyperparameter on the prediction was studied and optimal hyperparameter configuration was defined for each algorithm. The performance of the algorithms is evaluated using RMSLE, RMSE, R2, Mean IoU, and SMAPE. The multilayer perceptron demonstrated the greatest robustness in predicting TTT curves. Additionally, the Ft‐Transformer is a viable alternative if an appropriately sized dataset is available. These results provide valuable insights into the use of Machine Learning techniques as a new alternative in predicting isothermal transformation curves.

show abstract

Incorporation of prior knowledge in neural network model for continuous cooling of steel using genetic algorithm

Cited by 40 publications

References 28 publications

MCDM towards knowledge incorporation in ANN models for phase transformation in continuous cooling of steel

MCDM towards knowledge incorporation in ANN models for phase transformation in continuous cooling of steel

System for the recognition of wear patterns on microstructures of carbon steels using a multilayer perceptron

Prediction of Time–Temperature–Transformation Diagrams of NiAl Alloy: An Evaluation of Intelligent Algorithms

Contact Info

Product

Resources

About