Predictive modelling and Pareto optimization for energy efficient grinding based on aANN-embedded NSGA II algorithm

Wang, Jinling; Tian, Yebing; Hu, Xintao; Li, Yang; Zhang, Kun; Liu, Yanhou

doi:10.1016/j.jclepro.2021.129479

Cited by 28 publications

(7 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Modern modeling techniques based on neural networks and the support vector method are increasingly used to solve complex engineering problems, which can potentially provide better prediction accuracy. Wang et al [14], based on a neural model using non-dominated sorting genetic algorithm II (NSGA II), solved the problem of multi-criteria optimization in the grinding process. They presented compromise solutions between criteria related to energy and time efficiency and product quality.…”

Section: Introductionmentioning

confidence: 99%

Analytical Approach for Forecasting the Load Capacity of the EN AW-7075-T6 Aluminum Alloy Joints Created Using RFSSW Technology

Kluz,

Bucior,

Kubit

et al. 2024

Materials

View full text Add to dashboard Cite

To ensure the high reliability of aircraft structures, the Refill Friction Stir Spot Welding (RFSSW) process must be characterized by a high load capacity of the welds and a small standard deviation of the load capacity spread. This allows us to obtain uniform functional properties in the connections, ensuring the high quality of the process. This work aims to select the most favorable technological parameters for the welding process of EN AW-7075-T6 Alclad aluminum alloy sheets, which are used for the production of aircraft structures. The best networks were calculated using the Statistica 13.3 program. The obtained results were compared with the results of previous investigations. It has been shown that a model using neural networks allows for the determination of connection parameters with much greater accuracy than the classical model. The maximum error in estimating the load capacity of the connection for the mathematical model was 6.13%, and the standard deviation was 14.51%. In the case of neural networks, the maximum error value did not exceed 1.55%, and the standard deviation was 3.74%. It was shown that, based on the neural model, it is possible to determine the process parameters that ensure the required quality capacity of the process, ensuring a probability of obtaining the required load capacity of the connections amounting to P = 0.999935 with a defect rate of 0.0065%. This possibility is not provided by the classical model due to its large error in estimating the process spread and the high sensitivity of the process input parameters to the output parameters.

show abstract

Section: Introductionmentioning

confidence: 99%

Analytical Approach for Forecasting the Load Capacity of the EN AW-7075-T6 Aluminum Alloy Joints Created Using RFSSW Technology

Kluz,

Bucior,

Kubit

et al. 2024

Materials

View full text Add to dashboard Cite

show abstract

“…Dai et al [23] studied the effect of grinding speed on grinding temperature and power consumption and analyzed the grinding surface performance from the perspective of undeformed chip thickness. Wang et al [24] developed a nonparametric model based on an improved adaptive Artificial Neural Network (aANN) to predict surface quality, machining time, total power consumption, and effective power consumption.…”

Section: Introductionmentioning

confidence: 99%

Study on a Novel Strategy for High-Quality Grinding Surface Based on the Coefficient of Friction

Li,

Jiao,

Liu

et al. 2023

Lubricants

Self Cite

View full text Add to dashboard Cite

Surface quality has a significant impact on the service life of machine parts. Grinding is often the last process to ensure surface quality and accuracy of material formation. In this study, a high-quality surface was developed by determining the coefficient of friction in grinding a quartz fiber-reinforced silica ceramic composite. By processing the physical signals in the grinding process, a multi-objective function was established by considering grinding parameters, i.e., surface roughness, coefficient of friction, active energy consumption, and effective grinding time. The weight vector coefficients of the sub-objective functions were optimized through a multi-objective evolutionary algorithm based on the decomposition (MOEA/D) algorithm. The genetic algorithm was used to optimize the process parameters of the multi-objective function, and the optimal range for the coefficient of friction was determined to be 0.197~0.216. The experimental results indicated that when the coefficient of friction tends to 0.197, the distribution distance of the microscopic data points on the surface profile is small and the distribution uniformity is good. When the coefficient of friction tends to 0.216, the surface profile shows a good periodic characteristic. The quality of a grinding surface depends on the uniformity and periodicity of the surface’s topography. The coefficient of friction explained the typical physical characteristics of high-quality grinding surfaces. The multi-objective optimization function was even more important for the subsequent high-quality machining of mechanical parts to provide guidance and reference significance.

show abstract

“…The specific grinding energy and material removal energy of the spindle has been well-modelled [ 38 , 39 ]. In our previous studies, both total and active energies of the spindle were analysed and optimized using machine learning and genetic algorithms [ 40 , 41 , 42 ]. However, repeated and intermittent infeed movements and high-speed approaching stages consume a large portion of energy and their influence on grinding energy cannot be ignored.…”

Section: Introductionmentioning

confidence: 99%

Energy Prediction Models and Distributed Analysis of the Grinding Process of Sustainable Manufacturing

Tian

Wang

et al. 2023

Micromachines

Self Cite

View full text Add to dashboard Cite

Grinding is a critical surface-finishing process in the manufacturing industry. One of the challenging problems is that the specific grinding energy is greater than in ordinary procedures, while energy efficiency is lower. However, an integrated energy model and analysis of energy distribution during grinding is still lacking. To bridge this gap, the grinding time history is first built to describe the cyclic movement during air-cuttings, feedings, and cuttings. Steady and transient power features during high-speed rotations along the spindle and repeated intermittent feeding movements along the x-, y-, and z-axes are also analysed. Energy prediction models, which include specific movement stages such as cutting-in, stable cutting, and cutting-out along the spindle, as well as infeed and turning along the three infeed axes, are then established. To investigate model parameters, 10 experimental groups were analysed using the Gauss-Newton gradient method. Four testing trials demonstrate that the accuracy of the suggested model is acceptable, with errors of 5%. Energy efficiency and energy distributions for various components and motion stages are also analysed. Low-power chip design, lightweight worktable utilization, and minimal lubricant quantities are advised. Furthermore, it is an excellent choice for optimizing grinding parameters in current equipment.

show abstract

Predictive modelling and Pareto optimization for energy efficient grinding based on aANN-embedded NSGA II algorithm

Cited by 28 publications

References 44 publications

Analytical Approach for Forecasting the Load Capacity of the EN AW-7075-T6 Aluminum Alloy Joints Created Using RFSSW Technology

Analytical Approach for Forecasting the Load Capacity of the EN AW-7075-T6 Aluminum Alloy Joints Created Using RFSSW Technology

Study on a Novel Strategy for High-Quality Grinding Surface Based on the Coefficient of Friction

Energy Prediction Models and Distributed Analysis of the Grinding Process of Sustainable Manufacturing

Contact Info

Product

Resources

About