A two-step machine learning approach for dynamic model selection: A case study on a micro milling process

Cruz, Yarens J.; Rivas, Marcelino; Quiza, Ramón; Haber, Rodolfo E.; Castaño, Fernando; Villalonga, Alberto

doi:10.1016/j.compind.2022.103764

Cited by 21 publications

(9 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The scholars also studied the prediction of surface roughness in cutting chatter machining through machine learning algorithm. Cruz et al 27 predicted the apparent roughness after cutting through two-step machine learning method, and verified the accuracy of the model through experiments. Deshpande et al 28 Li predicted the apparent roughness of cutting through artificial neural network combined with cutting force, sound and vibration, and verified the validity of the model by comparing it with the modified regression model.…”

Section: Prediction Methods Of Cutting Stability Of Roadheader Based ...mentioning

confidence: 92%

Prediction method of cutting stability of roadheader based on the Newton–Lagrange mixed discrete method

Zhu

Xie

Wang

2022

Sci Rep

View full text Add to dashboard Cite

In order to study the influence of the multiple interaction of coal and rock and the velocity effect of the cutting head on the flutter stability of the cantilever roadheader, considering the physical characteristics of coal and rock, the structural parameters of the cutting head and the motion parameters of the cutting head, the functional expression of the cutting depth of the cutting head and the participating cutting teeth is fitted, the cutting state mechanical characteristic equation of the dynamic cutting of the cutting teeth is established, and based on the mapping relationship between the cutting teeth and the cutting head, a cutting dynamic model including multiple interactions between cutting head and coal and rock and the speed effect is constructed. An improved discrete method based on Newton–Lagrange mixed interpolation is proposed, and the influence law of the coupling effect of regeneration effect and velocity effect on the stability of cutting flutter under cutting state is clarified. The improved full discrete method is compared with the full discrete method and the semi-discrete method, and the superiority of the improved full discrete method based on the mixed interpolation method is proved. Based on the improved total discrete method, the influence of different cutting system dynamic parameters on the stability is studied. A cutting head coal rock system is built to simulate the flutter of the cutting system. The results show that the improved fully discrete method can reasonably predict the actual cutting state.

show abstract

Section: Prediction Methods Of Cutting Stability Of Roadheader Based ...mentioning

confidence: 92%

Prediction method of cutting stability of roadheader based on the Newton–Lagrange mixed discrete method

Zhu

Xie

Wang

2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…The usage of the term "Digital Twin" has been increasing rapidly in the academic community with the manufacturing domain as the major contributor [4,7,13,14]. This leap in the research of DTs is mainly related to the demand by the industrial sector and due to the advances in the Internet and Communication Technologies (ICT) domain [15,16].…”

Section: Literature Review On Designing and Developing Digital Twinsmentioning

confidence: 99%

“…Equation ( 11) creates the function block, while Equation (12) adds the alias to the created instance of the function blocks. Then, Equations ( 13) and (14)…”

Section: Function Block-based Digital Twin Architecturementioning

confidence: 99%

Ontology-Driven Guidelines for Architecting Digital Twins in Factory Automation Applications

2022

Self Cite

View full text Add to dashboard Cite

The rapid emerging technologies in various fields permitted the creation of simulation tools. These tools are designed to replicate physical systems in order to provide faster, cheaper and more detailed illustrative analysis of the physical system. In this regard, the concept of digital twins has been introduced to generally define these simulation tools. In fact, and according to the creator of the digital twin term Micheal Grieves, a digital twin is defined as a physical system, a digital replica of the physical system and information flow between the former parts. This definition is simple and generic for describing digital twins and yet, holistic. This broad definition creates a challenge for developers who target the development of such applications. Therefore, this paper presents a paradigm for architecting digital twins for manufacturing processes. The approach is inspired by the definitions of the ISA95 standard and the onion concept of computer applications to create multi-layer and multi-level concepts. Furthermore, and to satisfy the different required features by industries, the approach considers a multi-perspective concept that allows the separation of the digital twin views based on functionality. This paradigm aims at providing a modular, scalable, reusable, interoperable and composable approach for developing digital twins. Then, an implementation of the approach has been introduced using an ontology-based system and the IEC61499 standard. This implementation has been demonstrated on a discrete manufacturing assembly line.

show abstract

“…Te mean absolute percentage error (MAPE) is the average percentage of the absolute diference between the actual and predicted values divided by the actual value for each time period [54]. MAPE can be expressed by the following expression:…”

Section: Evaluation Parametermentioning

confidence: 99%

Air Quality Prediction Using the Fractional Gradient-Based Recurrent Neural Network

Arora

Singh

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

In this study, the air quality index (AQI) of Indian cities of different tiers is predicted by using the vanilla recurrent neural network (RNN). AQI is used to measure the air quality of any region which is calculated on the basis of the concentration of ground-level ozone, particle pollution, carbon monoxide, and sulphur dioxide in air. Thus, the present air quality of an area is dependent on current weather conditions, vehicle traffic in that area, or anything that increases air pollution. Also, the current air quality is dependent on the climate conditions and industrialization in that area. Thus, the AQI is history-dependent. To capture this dependency, the memory property of fractional derivatives is exploited in this algorithm and the fractional gradient descent algorithm involving Caputo’s derivative has been used in the backpropagation algorithm for training of the RNN. Due to the availability of a large amount of data and high computation support, deep neural networks are capable of giving state-of-the-art results in the time series prediction. But, in this study, the basic vanilla RNN has been chosen to check the effectiveness of fractional derivatives. The AQI and gases affecting AQI prediction results for different cities show that the proposed algorithm leads to higher accuracy. It has been observed that the results of the vanilla RNN with fractional derivatives are comparable to long short-term memory (LSTM).

show abstract

A two-step machine learning approach for dynamic model selection: A case study on a micro milling process

Cited by 21 publications

References 59 publications

Prediction method of cutting stability of roadheader based on the Newton–Lagrange mixed discrete method

Prediction method of cutting stability of roadheader based on the Newton–Lagrange mixed discrete method

Ontology-Driven Guidelines for Architecting Digital Twins in Factory Automation Applications

Air Quality Prediction Using the Fractional Gradient-Based Recurrent Neural Network

Contact Info

Product

Resources

About