Elsevier

Journal of Cleaner Production

Volume 131, 10 September 2016, Pages 754-764
Journal of Cleaner Production

Power consumption and tool life models for the production process

https://doi.org/10.1016/j.jclepro.2016.04.099Get rights and content

Highlights

  • Survey on modelling power consumption in manufacturing process is undertaken.

  • Need for explicit generalized power consumption and tool life models is highlighted.

  • Three advanced methods (GP, SVR and MARS) are proposed in this work.

  • GP models outperform SVR and MARS.

  • GP models can be used to improve the environmental performance of process.

Abstract

For achieving the multi-objective optimization of product quality and power consumption of any production process, the formulation of generalized models is essential. Extensive research has been done on applying the traditional statistical methods (analysis of variance, response surface methodology, grey relational analysis, Taguchi method) in formulation of these models for the processes. In the present work, a detailed survey on the applications of these methods in modelling of power consumption for the production operations specifically machining is conducted. Critical issues arising from the survey are highlighted and hence form the motivation of this study. Further, three advanced soft computing methods, namely evolutionary-based genetic programming (GP), support vector regression, and multi-adaptive regression splines are proposed in predictive modelling of tool life and power consumption of a turning phenomenon in machining. Statistical comparison based on the five error metrics and hypothesis tests for the goodness of the fit reveals that the GP model outperforms the other two models. The hidden relationships between the process parameters are unveiled from the formulated models. It is found that the cutting speed parameter is the most influential input for power consumption and tool life in the turning phenomenon. The future scope comprising of the challenges in predictive modelling of production processes is highlighted in the end.

Introduction

In view of the desire for achieving higher economic growth, emerging economies have paid significant attention in expediting the number of production industries in their respective countries. This has led to an increase in the demand for power and natural resources. Production operations include machining processes (turning, drilling, grinding and milling) and 3-D printing processes which make use of extensive amount of power drawn from the electrical grid (Kuram et al., 2013, CECIMO, 2009, Dang et al., 2007). Among these operations, machining is widely focused and performed in the production industry. Excessive use of power used in driving the machining operations has an adverse impact on the environment. Higher power consumption translates to higher emissions of the toxic gases in the air. The consequences of air pollution are negative on both the environmental and social aspects such as living conditions of the urban population. Thus, the optimization of power performance of production operations will benefit the environment and improve the economic performance of the industry (Lam and Dai, 2015). Performing the production operations for higher economic and environmental performance has led the experts to work extensively towards the environmental conscious production (ECM) or green manufacturing (GM). For example, if the power is drawn from the thermal power station, lowering the power consumption would result in the lower emission of harmful gases, and would result in saving of water resources, if the route chosen for drawn was from hydraulic plants (Bhushan, 2013, Balogun and Mativenga, 2013).

Numerous survey studies (Mukherjee and Ray, 2006, Chandrasekaran et al., 2010, Garg and Tai, 2012, Garg and Tai, 2013a, Garg and Tai, 2013b, Garg et al., 2013a, Garg et al., 2013b, Liu, 2015) conducted on the modelling of machining operations reveal that the focus was on the optimizing the cost and productivity of the production process. Majority focussed on optimizing the surface roughness of the machining operations (Aykut et al., 2007, Li et al., 2000, Kim and Ehmann, 1993, Young et al., 1994, Zheng et al., 1998). To the best of authors' knowledge, less focus has been paid on optimizing the environmental/power characteristics (power consumption and cutting forces) of the machining operations (Pawade et al., 2009, Rajemi et al., 2010, Li and Kara, 2011, Hu et al., 2012, Garg and Lam, 2015). It is known that power is consumed during the different stages (at the time of the machining, after the machining process and during the initial idle time so as to drive motors and auxiliary components such as fans and coolant pumps) of machining process. Machine tool is designed based on the peak power requirement, which is significantly higher than the non-peak power requirement of the machine tools. This results in lower power efficiency of machine tools. Optimization of the power component in machining operations can result in wide applications of lower power rated motors/auxiliary components and thus can prevent wastage of power, thereby improving the environmental impacts of the machining operations (Kant and Sangwan, 2014).

Vital standards for assessing the performance of any production industry are the product quality, tool life, and power efficiency. It is known that the efficiency of machining operations is lower than 30%, and almost 99% of the environmental impacts are from the power consumption (Kant and Sangwan, 2014). Lowering the product quality or the tool life does result in a reduction in power consumption. However, the product would not be accepted by the market. Therefore, there is a need to find a balance between power consumption, product quality, and tool life by effectively determining the optimum input process parameters of the machining operation (Kant and Sangwan, 2014). In the context of optimization of machining operations, the formulation of mathematical model representing the relationship between the outputs (power consumption, tool life, and product quality) and inputs is vital. Due to complexity of the process, the physics-based models may not be able to extrapolate the power consumption and product characteristics simultaneously beyond the input conditions range. Therefore, this problem has indeed been a strong motivation for the authors towards exploring the means of investigating the modelling of power characteristics of the machining operations by soft computing methods.

The present work is described in a structured manner as follows. Section 2 gives a brief description on the soft computing methods used in modelling the environmental characteristics of the machining process. Section 3 summarizes the literature review in a table format illustrating the applications of the soft computing methods used in modelling the environmental performance of the various machining processes. Section 4 provides description on the three advanced soft computing methods suggested as an alternative. Section 5 discusses the details on the experimental data used for exploring the performance of three advanced soft computing methods and the settings of parameters for their implementation. Section 6 discusses the performance of the proposed models. In Section 7, the relationships between the process parameters are revealed from the best model. In Section 8, along with conclusions, the implications and the future work arising from the present work are discussed.

Section snippets

Brief description of soft computing methods

As discussed in Section 1, several soft computing methods including analysis of variance (ANOVA), response surface methodology (RSM), grey relational analysis, Taguchi method, artificial neural network (ANN) were extensively used to predict the environmental characteristics (power consumption, cutting force, etc.) and input variables of the machining processes such as turning and milling. Formulated models predict the values of the environmental characteristics based on the set of values of

Modelling of the production and environmental characteristics of the machining processes: applications and discussions

Authors have compiled the literature (Table 1) on modelling the machining operations using soft computing methods. From Table 1, it is shown that the most widely used method is RSM because it can be applied to the limited set of experiments. Further, the analysis of variance (ANOVA) is conducted to determine the significant and redundant inputs in the production processes. However, these statistical methods hold the assumptions such as structure of a model before the problem in hand, normality

Advanced soft computing methods

A brief description of the three advanced soft computing methods, namely GP, MARS, and SVR, is given as follows.

Experimental set-up details of turning of composites for measuring the tool life and power consumption

The experiments on turning phenomenon were performed on the CNC machine. The advantages of using the CNC machine are the low cost of machine and the higher quality of the product. The inputs to this turning process are the feed rate (mm/rev), nose radius (mm), depth of cut (mm) and the cutting speed (m/min). Output process parameters considered are power consumption (W h/Watt hour) and Tool life (min). The optimum combination of these factors can improve productivity and lower the power

Statistical analysis of models formulated from advanced soft computing methods

The models formulated from the three methods (GP, SVR and MARS) are evaluated against the experimental data (Bhushan, 2013) based on the five statistical metrics given by:Cofficientofdetermination(R2)=(i=1n(AiAi¯)(MiMi¯)i=1n(AiAi¯)2i=1n(MiMi¯)2)2Meanabsolutepercentageerror(MAPE,%)=1ni|AiMiAi|×100Rootmeansquareerror(RMSE)=i=1N|MiAi|2NRelativeerror(%)=|MiAi|Ai×100Multiobjectiveerror=MAPE+RMSER2where Mi and Ai are the predicted and actual values respectively, Mi¯ and Ai¯ are the

Relationships between tool life and power consumption and the four inputs

It is true that the performance of GP model may vary depending on the selection of the training data set. Therefore, GP runs were conducted on the 10 randomly generated training data sets by the cross-validation procedure. From Fig. 4, it is observed that variation (minimum to maximum) of the testing RMSE is quite low and thus it can be concluded that the GP model is able to capture the dynamics of the turning process based on the four inputs.

The relationships of the both outputs with respect

Conclusions and recommendations for future work

The present work conducts a critical survey on applications of soft computing methods in environmental sustainability of machining processes. In addition, the need for formulation of generalized functional expressions for accessing the production and environmental aspects (power consumption) of the production operations is addressed. Alternatively, three advanced soft computing methods, i.e. GP, SVR and MARS, are proposed and their performance is explored in modelling the production and

Acknowledgements

The study was supported by Nanyang Technological University Research Grant ref. M060030008.

References (78)

  • A.H. Gandomi et al.

    Assessment of artificial neural network and genetic programming as predictive tools

    Adv. Eng. Softw.

    (2015)
  • A. Gandomi et al.

    Behavior appraisal of steel semi-rigid joints using linear genetic programming

    J. Constr. Steel Res.

    (2009)
  • A.H. Gandomi et al.

    A hybrid computational approach to derive new ground-motion prediction equations

    Eng. Appl. Artif. Intell.

    (2011)
  • A. Garg et al.

    Improving environmental sustainability by formulation of generalized power consumption models using an ensemble based multi-gene genetic programming approach

    J. Clean. Prod.

    (2015)
  • A. Garg et al.

    A molecular simulation based computational intelligence study of a nano-machining process with implications on its environmental performance

    Swarm Evol. Comput.

    (2015)
  • I. Hanafi et al.

    Optimization of cutting conditions for sustainable machining of Peek-Cf30 using Tin tools

    J. Clean. Prod.

    (2012)
  • S. Hu et al.

    An on-line approach for power efficiency monitoring of machine tools

    J. Clean. Prod.

    (2012)
  • G. Kant et al.

    Prediction and optimization of machining parameters for minimizing power consumption and surface roughness in machining

    J. Clean. Prod.

    (15 November 2014)
  • S. Kara et al.

    Unit process power consumption models for material removal processes

    CIRP Ann. Manuf. Technol.

    (2011)
  • H.S. Kim et al.

    A cutting force model for face milling operations

    Int. J. Mach. Tools Manuf.

    (1993)
  • E. Kuram et al.

    Optimization of cutting fluids and cutting parameters during end milling by using D-optimal design of experiments

    J. Clean. Prod.

    (2013)
  • X.P. Li et al.

    An approach to theoretical modeling and simulation of face milling forces

    J. Manuf. Process.

    (2000)
  • J. Li et al.

    RBFNDOB-based neural network inverse control for non-minimum phase MIMO system with disturbances

    ISA Trans.

    (2014)
  • P.L. Liu

    Further improvement on delay-dependent robust stability criteria for neutral-type recurrent neural networks with time-varying delays

    ISA Trans.

    (2015)
  • J. Liu et al.

    Optimizing the conditions for the microwave-assisted direct liquefaction of Ulva prolifera for bio-oil production using response surface methodology

    Power

    (2013)
  • I. Mukherjee et al.

    A review of optimization techniques in metal cutting processes

    Comput. Ind. Eng.

    (2006)
  • A.K. Pani et al.

    Online monitoring and control of particle size in the grinding process using least square support vector regression and resilient back propagation neural network

    ISA Trans.

    (2015)
  • R.S. Pawade et al.

    An analytical model to predict specific shear power in high-speed turning of Inconel 718

    Int. J. Mach. Tools Manuf.

    (2009)
  • H. Quan et al.

    Uncertainty handling using neural network-based prediction intervals for electrical load forecasting

    Power

    (2014)
  • M.F. Rajemi et al.

    Sustainable machining: selection of optimum turning conditions based on minimum power considerations

    J. Clean. Prod.

    (2010)
  • R.F. Ropero

    Regression using hybrid Bayesian networks: modelling landscape–socioeconomy relationships

    Environ. Model. Softw.

    (2014)
  • M. Sarıkaya et al.

    Taguchi design and response surface methodology based analysis of machining parameters in CNC turning under MQL

    J. Clean. Prod.

    (2014)
  • S. Singh et al.

    Thermal and thermohydraulic performance evaluation of a novel type double pass packed bed solar air heater under external recycle using an analytical and RSM (response surface methodology) combined approach

    Power

    (2014)
  • H. Taghavifar et al.

    Diesel engine spray characteristics prediction with hybridized artificial neural network optimized by genetic algorithm

    Power

    (2014)
  • J.M. Wang et al.

    Numerical simulation and process optimization of an aluminum holding furnace based on response surface methodology and uniform design

    Power

    (2014)
  • W. Wang et al.

    Exponential convergence analysis of uncertain genetic regulatory networks with time-varying delays

    ISA Trans.

    (2014)
  • X. Xie et al.

    Improved delay-dependent stability analysis for neural networks with time-varying delays

    ISA Trans.

    (2014)
  • J. Yan et al.

    Multi-objective optimization of milling parameters – the trade-offs between power, production rate and cutting quality

    J. Clean. Prod.

    (2013)
  • W.H. Yang et al.

    Design optimization of cutting parameters for turning operations based on the Taguchi method

    J. Mater. Process. Technol.

    (1998)
  • Cited by (37)

    • Energy prediction for CNC machining with machine learning

      2021, CIRP Journal of Manufacturing Science and Technology
    • Axial contact points method for improving end-milling productivity

      2021, Materials Today: Proceedings
      Citation Excerpt :

      This is due to extreme loads on all elements of the machining system, mainly on cutting tools. Large cutting forces accelerate tool wear and increase the deflection force [20]. The deflection has a negative impact on the machining accuracy.

    • Use of optimization techniques for energy use efficiency and environmental life cycle assessment modification in sugarcane production

      2019, Energy
      Citation Excerpt :

      This is attained when both groups of energy use and environmental impact categories equal to values in MOGA. Garg et al. [74] reported an optimization algorithm according to genetic programming, which was applied to modify the environmental efficiency and to maintain power use in constructing procedures. Vijayaraghavan et al. [75] demonstrated that a cutting angle optimization with a genetic programming algorithm can decrease cutting power, reduce energy use and lead to a resistant production approach.

    • Power generation and pollutant emissions in the European Union: A mean-variance model

      2018, Journal of Cleaner Production
      Citation Excerpt :

      This work follows the lines initiated by several previous studies on portfolio theory and energy planning (Humphreys and McClain, 1998; Krey and Zweifel, 2006; Arnesano et al., 2012; De-Llano et al., 2014; DeLlano-Paz et al., 2015) with regard to the inclusion of the externality costs derived from power production in the cost structure of the technologies. It is a matter of adding in those costs not taken into account by the integrated power companies (except for those related to CO2 emissions), including potential damages to ecosystems and society, which have a negative effect on well-being and public health (Liu et al., 2016; Garg et al., 2016). The intent of this is to partially correct the possible market error derived from not considering these costs, which are borne by society and the environment (Eyre, 1997; Schultman et al., 2001).

    View all citing articles on Scopus
    View full text