Accepted Manuscript

Critical factors in energy demand modelling for CNC milling and impact of toolpath
strategy

Ampara Aramcharoen, Paul T. Mativenga

PII: S0959-6526(14)00422-3
DOI: 10.1016/j.jclepro.2014.04.065
Reference: JCLP 4272

To appear in: Journal of Cleaner Production

Received Date: 30 August 2013

Revised Date: 23 April 2014
Accepted Date: 25 April 2014

Please cite this article as: Aramcharoen A, Mativenga PT, Critical factors in energy demand modelling
for CNC milling and impact of toolpath strategy, Journal of Cleaner Production (2014), doi: 10.1016/
j.jclepro.2014.04.065.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
Critical factors in energy demand modelling for CNC milling and impact of toolpath strategy
1 2
Ampara Aramcharoen , Paul T Mativenga
1
Singapore Institute of Manufacturing Technology, 71 Nanyang Drive, Singapore 638075
2
School of Mechanical, Aerospace and Civil Engineering, The University of Manchester, Manchester
M13 9PL, United Kingdom

Abstract

PT
Reducing energy demand in manufacture is an urgent challenge. This challenge is driven by higher
consumer demand for manufactured products, increasing electricity and energy prices, volatility and
uncertainty in energy supply and national policy. These factors, together with a need to reduce energy
consumption derived carbon dioxide emissions, strategically call for energy efficient manufacturing. In

RI
manufacturing processes, especially mechanical machining, more than 90% of environmental impact
arises from direct electrical energy demand in machine tools. At the machine tool level, the biggest
share of the electrical energy associated with mechanical machining is required to bring the machine

SC
to a ready state and support non-cutting operations such as spindle torque requirements, auxiliary
units and movements. These activities are controlled and related to machine commands such as NC
codes. In this paper comprehensive information on energy intensity in machining process, including

U
the influence of tool wear, was studied. Key energy states were identified to build up an energy
demand for machining components. The paper defines the essential power constants for a database
AN
that can assist energy prediction for any available machines and workpiece materials. The
assessment of alternative toolpaths identified major opportunities for energy demand reduction.
M

Keyword
Energy consumption, machining, milling, tool wear, toolpath
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
1. Introduction
The increase in the population of nation states and the world is promoting higher demand for
products. The challenge that arises is how manufacturing industries can cope with increased product
demand and sales revenues as well as simultaneously supporting environmental sustainability.
Electricity use in manufacturing is one major driver for carbon emissions increase. This is the case
because carbon rich energy sources and fossil fuels dominate electrical energy generation (Jeswiet
and Kara, 2008). Thus industry needs to be smarter to increase output and productivity while reducing
the energy intensity of manufactured products and manufacturing cost.

The manufacturing sector is one of the key growth drivers in many countries and hence a major

PT
energy consumer for the world. International Energy Agency (IEA, 2013) reported the world electricity
consumption by each sector in 2011, the total figure had reached - 1582 Mtoe (18,396,735 GWh),
where industry sector is the highest consumer of electricity with a percentage of 42.6%. This energy

RI
consumption represents a high share of environmental burden in the manufacturing sector (Bonvoisin
et al., 2013).

SC
Mechanical machining is a technology widely used in manufacturing industries. Gardner Research
(2013) reported the consumption of machine tools from 28 leading manufacturing countries
(including China, Germany, Japan, US etc). Total machine tool sales increased in the past ten years
and reached $USD 93,205 million in 2012. The European commission identified machine tools in

U
manufacturing as a top three priority for product categories regulated through the EcoDesign directive
(EPTA, 2007).
AN
The European Union has set the goal to significantly increase efficiency in the use of resources in all
sectors, including industry. In many countries Small to Medium Enterprises (SMES) are significant
parts of the economy and need support to manage their energy and environmental footprints. France
M

and the UK have energy efficiency policies for SMEs, while an Industries Assessment Centre in the
US was established to improve energy management for SMEs. Implementation of energy efficiency
for SMEs in Sweden, Norway, the European Union and India showed potential energy saving of 7 to
D

22% on energy bills (Hao, 2011).

Recently, there are more consortia and collaborations around the world working on industrial
TE

sustainability. For example, The Sustainability Consortium (TSC®, 2013) is working to develop
methodologies, tools and strategies for improving sustainability of product and supply networks.
However, TSC is currently limited in scope to household and consumer products. Cooperative Effort
EP

on Process Emissions in Manufacturing so-called CO2PE!-Initiative is focusing on manufacturing

processes. They are working on a framework of environment footprint with energy consumption and
CO2 emission, developing methodology and identifying opportunity for improvement (CO2PE!, 2013 ).
The proposed CO2PE! taxonomy for energy consumption in milling processes is composed of basic
C

energy, idle energy and milling energy. These initiatives do not clarify machine tool design and its
complexity, including operations specific tasks done by machine tool axis movements and or defined
AC

by CNC codes. Thus a comprehensive methodology and robust energy model will enhance the
information on process level energy modelling.

1.1 Energy consumption in machining process

Mechanical machining processes such as turning, milling and drilling are widely used in
manufacturing production. Their popularity is due to the capability to fabricate 3D complex geometries
to high dimensional accuracy and cost efficiency. The machining process is performed on machine
tool centres. Electricity is the main power source for machine tool centres in cutting processes. The
energy consumption of machine tools includes the demands for the spindle, axes motion, cutting
resistance (workpiece materials, cutting tool and cutting conditions) and others (cutting fluid pump,
cooling device, computer controller). Kordonowy (2002) categorised energy consumption for
 ACCEPTED MANUSCRIPT
mechanical machining into constant and variable parts. The former, composed of two sub-groups
(start-up and run time) depends on machine tool modules (servo motor, computer, spindle, cutting
fluid etc), while the latter was influenced by machining and material removal rate. The energy in
manufacturing system can be analysed at the unit process level, multi-machine system, factory level
and supply-chain level (Duflou et al., 2012). This enables linking the energy demand of machine tools
to factory wide performance.

It has been suggested that more than 90% of environmental impact from machine tools is due to
electrical energy consumption (CECIMO, 2009). This electrical power consumption is the biggest
source of environmental impact because carbon dioxide is released when the electricity is generated

PT
from carbon rich fuel sources. Hence the energy used in machining process can be used to calculate
the associated carbon emission through the carbon emission signature, CES (Jeswiet and Kara,
2008).

RI
The estimation of cutting energy or tip energy in material removal process can be accomplished
through the specific cutting energy in chip formation, which is the minimum energy required to remove

SC
a certain volume of material (Dahmus and Gutowski, 2004). This depends on the machinability of the
material, taking into account the material properties, cutting fluid, cutting tool and cutting conditions.
The influence of chip thickness and feedrate on specific cutting energy was reported by Balogun and
Mativenga (2014). They suggested that specific energy consumption in roughing processes can be

U
reduced by machining at feedrate greater than the tool edge radius. However in practice, the energy
consumed by machining operation is much higher than the specific cutting energy. The machine tool
AN
centre is comprised of a variety of functions such as workpiece handling and movement, cutting fluid
system, chip removal, tool changing and tool measuring. All of these functions require energy to
activate and to support metal cutting by plastic deformation. Table 1 summarizes the mathematical
models of energy consumption in machining process from the early work. Gutowski et al (2006)
M

presented a fundamental model for energy demand in cutting. They reported that the energy required
for actual cutting was a very small portion in relation to the total machine tool operation energy. This is
due to the dominancy of basic energy (fixed energy) of machine tools. The percentage of actual
D

cutting energy will become even smaller when operating at low production rates (Rajemi et al., 2010).

The energy consumption for machining operation can be reduced by the selection of machine tool
TE

centres that have low basic power requirements. However in practice, some industries have limited
choice and investment and they would like to use existing resources and machines. Thus improving
energy efficiency using existing resources is a prompt solution for industry. Li et al (2011) proposed
EP

that energy demand could be reduced by improving machine tool components (hydraulic system,
cooling and lubrication system) and by optimizing machining processes (minimize standby time).

Dietmair and Verl (2009) introduced a methodology and framwork for energy consumption
C

minimization of machine or production system. Rajemi et al (2010) developed a model to select

conditions for minimum energy footprint by constraining cutting conditions by an optimum cutting
AC

velocity and tool life that satisfied the minimum energy criterion. The energy footprint model in
machining a single pass was built by considering power and energy requirements when setting up the
machine, doing actual cutting, tool change time and the embodied energy in the cutting tool.

Gutowski (2006) reported that non-cutting operations dominate power demand in machining
processes. In additional work, Rajemi et al (2010) model for energy footprint in machining was
extended by including spindle power consumption characteristics (Mativenga and Rajemi, 2011).
They showed that power consumption proportionally increased with spindle speed (without cutting
load). Their research pointed out that the selection of cutting parameters based on their minimum
energy footprint and minimum cost criterion resulted in significant reduction of energy footprint
compared to the conditions recommended by cutting tool suppliers. Furthermore, the direct energy
model was then developed for milling processes (Balogun and Mativenga, 2013). Several researchers
studied the optimization and influence of cutting parameters on power consumption in machining
 ACCEPTED MANUSCRIPT
process. For example, in CNC turning of Al alloy SiC composites, the effects of cutting speed, feed
rate, depth of cut and tool nose radius were investigated with regards power consumption and tool life
(Bhushan, 2013). Yan and Li (2013) studied the influence of spindle speed, feed rate, depth of cut
and width of cut in milling process. Campatelli et al (2014) also evaluated the influence of cutting
speed, feed rate, axial and radial depth of cut in milling of carbon steel. They showed that to achieve
a lower environmental footprint, material removal rate needs to be increased.

Mori et al from Mori Seiki Co. Ltd (2011) highlighted two major reasons for power consumption
including i) spindle and axes movement and ii) providing cutting force to overcome cutting resistance.
By synchronising spindle and axes movement, a control strategy was developed to reduce power

PT
consumption. It was suggested that the cutting energy could be reduced by minimized cutting
resistance and reduced cutting time. However, is noted here that this optimal cutting condition should
not cause any premature tool wear or unacceptable machined quality.

RI
Avram and Xirouchakis (2011) emphasized that energy demand depended on the interaction between
spindle and feed, the selection of technology, intelligent process design and effective integration of

SC
the process. Kong et al (2011) evaluated the influence of toolpath on environmental impact by
introducing web-based and application program interface (API). Their energy model was comprised of
constant part, run-time and cutting stage. Their research showed that the selection of toolpath
influences energy consumption. He et al (2012) presented an energy consumption model related to

U
numerical control (NC) program in machining. They suggested that this method will help in effective
energy efficient NC program selection.
AN
Diaz et al. (2009) compared conventional to high speed machining and showed that high speed
machining was more energy efficient due to reduced processing time. Additionally (Diaz et al., 2011),
investigated the influence of material removal rate and selection of machine tool (nano-, micro- and
M

macro machining centred) on energy consumption. Further investigation by Balogun et al (Balogun et

al., 2013), also reported the impact of machine tool selection and geographical location on energy
intensity and carbon footprint when milling with a standardized NC toolpath. An interesting point was
D

that for machine tools located in higher carbon intensity regions, greater improvement in energy
efficiency is required if the similar carbon footprint reductions for the process is to be realised.
TE

The influence of machining strategies such as roughing, finishing, dry machining or flood was
highlighted by Neugebauer et al (2011). Kara and Li (2011) presented their model for specific energy
consumption based on fixed, operational, tool tip and unproductive energy. In this model, machine
EP

start-up, stand-by, clamping and positioning had not been explicitly considered. Their model
suggested that at higher material removal rate, lower energy would be consumed due to improved
productivity. An on-line energy efficiency monitoring model was proposed by Hu et al (2012). Their
energy model for the spindle system captured load loss influenced by electrical loss, mechanical loss
C

in motor and transmission system.

AC

Recently, a framework for estimating energy consumption of product oriented manufacturing system
was presented by Bonvoisin et al (2013). It provided an overview to estimate energy consumption for
product fabrication which composes of i) direct consumption (consumed by processing machine) and
ii) indirect consumption (consumed by technical building service). The direct consumption was divided
into processing energy and add-on energy. To predict energy consumption as a function of material
removal rate, they adapted an empirical model developed by Li and Kara (2011). The model did not
consider toolpath axes movement. The add-on energy was non-productive mode (or idle stage of the
machine tools without material removal). Additionally, the model did not consider tool life.

Frigerio et al (2013) proposed of build energy state-based models by using automata formalism. Their
method shows the relationship between functional modules such as spindle. They highlighted that this
method will help select energy efficient functional modules or evaluate the impact of different design
for machine tools. However, the model did not explicitly model the tip energy or actual material
 ACCEPTED MANUSCRIPT
removal energy. Abele et al (2012) presented a simulation based method to calculate energy
consumption at functional module level such as exhaust, cooling, hydraulic and cutting fluid units.
Nevertheless, they did model some operations such as axes movement, tool change, tool life and NC
codes.

Calvanese et al (2013) developed an energy model which composes of spindle, axis, chillers, tool
change system, auxiliaries and cutting process. However, they only investigated the face milling
where there was a single axis movement, either X, Y or Z direction, without considering simultaneous
movement. Additionally they did not consider the influence of tool wear on the energy consumption.

PT
1.2 Research motivation and aim

RI
The development of a more robust and accurate model for energy demand in machining requires
more explicit modelling of Basic Energy (energy demand by the machine at no load) and Tip Energy
(energy for actual material removal). At present there is a huge gap between the development models

SC
and the ability to link the model to toolpath, to capture effect of cutting variables and hence enable
energy benign process planning. For the models to be able to handle energy states for GOTO, G01,
G02 and G03 machine interpolation moves, it is essential to model energy demand by different
machine axis. This aspect has not been comprehensively investigated in literature. The purpose of

U
this paper is to consolidate on the capability of existing models, combining the best contributions from
different models, and filling the gaps that are required to models essential features Automatically
AN
Programmed Tool (APT) files or computer numerically controlled toolpaths. The APT and computer
numerically controlled (CNC) nomenclature is used to program numerically controlled machine tools
to create complex parts using a cutting tool moving in and out of workpiece space. This paper aims to
M

critically define the essential building blocks for machine tool actions and energy states and undertake
an assessment of energy consumption for toolpaths. At present many machining operations are done
by computer numerical control (CNC) or through Computer Aided Machining (CAM). In this approach
there is no unique way of manufacturing or removing material. Thus it is important to explore how
D

different toolpath can influence the direct energy demand in machining and hence suggest a way of
significantly reducing energy intensity at the material removal level.
TE

2. Direct Electrical Energy Model for Machining

EP

In this study, the system boundary, in Figure 1, was selected in order to observe and estimate energy
consumption and hence carbon footprint for the machine tool and cutting process, focussing on direct
energy demand. The selected boundary is shown in a dashed line which covers the cutting process
C

and milling machine tool. The motivation is to better model an aspect that has an influence on a
company’s electricity bill. On the basis of this focus, the embodied energy for inputs, which are cutter
AC

and workpiece material, was not considered in this evaluation. Figure 1 shows the overall machining
procedure including (1) “inputs” of cutting tools and workpiece material, (2) “process” of cutting and
machine tool and (3) “outputs” of machined component.

Most machine tools are controlled by numerical control (NC) codes. These codes are input to
machine tool controller for defining the operations and the movement of cutting tool, workpiece and
cutting activities. The NC codes are linked to the components in the machine tools such as axis feed,
servo drive system, spindle, cutting fluid pump and tool change system through the machine
controller. The design of the machined tool and its modules influence the basic energy consumption
(or fixed energy) in machining process. Table 2 summaries the common NC codes as related to
machine components and their operations. The knowledge that needs to be filled is to develop
machine tool energy state equations that can be related to the CNC codes.
 ACCEPTED MANUSCRIPT
Energy consumption in machining process can be estimated by power consumption and time, as
illustrated in Equation (1). The power consumption can be derived from the current monitored and
captured using a power meter/sensor. Figure 2 shows the power profile of milling process done on a
Hitachi Seiki VG45. The process begins with machine start-up, workpiece setup, machine warm-up,
tool change, and the cutting process until the machine is switched OFF. From this figure, it can be
seen that the energy requirement in milling process needs to be built up from distinct events or states
experienced by the machine. The total energy states has to take into account energy for: operating
the machined tool (Pbasic), tool change (Ptool), spindle run (Pspindle), material removal (Pcutting), table
movement and cutting feed (Pfeed) and cooling system (Pcutting-fluid). Thus the total energy consumption

PT
of each machining process for machined part (discrete component) can be calculated from the energy
model as shown in Equation (2). The total energy consumption can be decreased either by reducing
the energy consumed by each module/activity or by reducing the process time.

RI
n ts
Energy = ∑ ∫ Pdt
0
(1)
0

SC
Where n is the number of energy states in the total machining operation that can be defined by a
unitary energy density function, ts is the duration of such machine events/states, that occur at power
P.
Etotal = Ebasic + Etool + Espindle + Ecutting + Efeed + Ecutting− fluid (2)

U
Basic Energy, Ebasic, is the energy required when the machine tool is switched ON but without the
AN
spindle running, without feed movement and without cutting. This basic power is influenced by the
design of the machine tool and its modules for example servo system, fan motor, computer, controller
and lighting. This keeps the machine tool in operational stage and ready for cutting activities. The
basic energy can be calculated from basic power, Pbasic, and the duration of the machine in this state
M

as shown in Equation (3).

Ebasic = Pbasic⋅ tbasic (3)

D

Where tbasic = tend-tstart is the total time required throughout operational process or the NC program
TE

(from start to the end).

Tool change Energy, Etool, is the energy required for changing the cutting tools. Tool change includes
the movement of tool turret for changing the tools, the hydraulics to unload and load the cutting tool.
EP

The tool change motor rotates the turret to a specific position to pick an appropriate tool as requested
by the NC code. The energy consumption for tool change, Etool, can be calculated by Equation (4).

Etool = Ptool ⋅ ttool ⋅ ntool (4)

C
AC

Where Ptool is the power demand by the tool change motor, ttool is time required for tool change (turret
rotating time) and ntool is the number of cutting tools used in one cutting operation. The number of
cutting tools can be determined by the cutting tool life (T) and the total cutting time for material
removal (tcutting), ntool= tcutting/T.

Spindle energy, Espindle, is influenced by the energy required for spindle transmission module which is
activated by “M-code” (M03 code for spindle ON and M05 for spindle OFF) and “S-code” (to define
specific spindle speed). The energy consumption for running the spindle can be estimated by spindle
power (Pspindle) as shown in Equation (5).
 ACCEPTED MANUSCRIPT
Espindle = Pspindle ⋅ t spindle (5)

where tspindle = tspindleend-tspindlestart is the total time required to run spindle. This spindle power is an
unloaded power of the spindle motor where there is motion of running the spindle but no cutting action
or no material removal. Hence the unloaded power is a function of the spindle rotation (N).

Cutting energy, Ecutting, is the energy used for cutting the workpiece removing workpiece material into
the form of chip. It can be estimated by cutting power, Pcutting at the period of cutting time, in Equation
(6).

PT
E cutting = Pcutting ⋅ t cutting (6)

RI
where tcutting is the cutting time which can be estimated based on toolpath and feedrate.

SC
The cutting or tip power can be simply defined using specific cutting energy, (k) as in Equation (7).

Pcutting = kQ (7)

U
The specific cutting energy is the energy required per unit volume of cut and depends on types of
AN
workpiece material and reflect machinability i.e. how easy it is to cut materials with an acceptable tool
life and workpiece quality. When cutting the same material with a dull tool, the specific cutting energy
will be higher (Kalpakjain and Schmid, 2003). Hence cutting energy can determined as shown in
Equation (8).
M

E cutting = ( kQ ) ⋅ t cutting (8)

D

3
Where Q is material removal rate (mm /s).
TE

Feed energy, Efeed, is the energy required to move the work table or cutting tool in X, Y and Z
direction at the given feedrate (Vf). This includes X-axis, Y-axis and Z-axis feed motors. The
movements are controlled by NC codes (G01, G02, G03) at certain feed rate or rapid feed rate (G00).
EP

The feed energy can be calculated by the summation of feed energy for each axis (He et al., 2012) as
shown in Equation (9).

m ti
E feed = ∑∫ end
Pfeed i ⋅ dt
C

(9)
ti
i =1 start
AC

Where Pfeed i is power requirement for i-axis movement (X, Y or Z axis), m is the number of axis
motors for machine tool. However, the NC-code for feed movement can be classified in two different
ways, which are rapid movement (G00) and defined feedrate (G01). Thus the feed energy
r
requirement can be summarized for rapid feed (E feed) as shown in Equation (10) and for energy for
f
cutting feed (E feed) as shown in Equation (11).

tx ty tz
∫ ∫ ∫
r end r end r end r
E feed ( start → end ) = Pfeed ⋅ dt + Pfeed ⋅ dt + Pfeed ⋅ dt (10)
t xstart x t ystart y t zstart z
 ACCEPTED MANUSCRIPT
r r
Where Pfeed x
is power required for x-axis movement at rapid feedrate, Pfeed y is power required for y-
r
axis movement at rapid feedrate and Pfeed z
is power required for z-axis movement at rapid feedrate.

tx ty tz
∫ ∫ ∫
f end f end f end f
Efeed ( start → end ) = Pfeed ⋅ dt + Pfeed ⋅ dt + Pfeed ⋅ dt (11)
t xstart x t ystart y t zstart z

f f
Where Pfeed x is power required for x-axis movement at defined feedrate, Pfeed y is power required for

PT
f
y-axis movement at defined feedrate and Pfeed z is power required for x-axis movement at defined
f
feedrate. Hence Efeed depends on the toolpath, the movement time for each axis and feedrate which

RI
is controlled by the servo interpolation.

Cutting fluid delivery energy, Ecutting-fluid, is the energy required for cutting fluid pump motor for the
coolant. The cutting fluid helps reduce friction, cutting temperature and helps in flushing away chips.

SC
The energy is modelled by Equation (12). This will be activated by M08 and deactivated by M09 NC
codes.

Ecutting − fluid = Pcutting − fluid ⋅ t cutting− fluid (12)

U
AN
Where Pcutting-fluid is power of cutting fluid pump motor and tcutting-fluid is the time required for supplying
cutting fluid t cutting −fluid = t cutting −fluidend − t cutting −fluidstart
M

3. Case study
D

3.1. Experimental setup

The machining of a testpiece on a milling process was set as a case study to verify the energy model
TE

for estimating energy consumption. The cutting experiment was conducted on Hitachi Seiki VG45
milling centre. The selected workpiece in this study was T316L stainless steel. The workpiece was
prepared into a size of 150 mm in both width and length and 20 mm in height. Figure 3 shows the
toolpath for milling the stainless steel testpiece. Uncoated carbide tools (TPKN1603PPTR P30) with
EP

diameter of 50 mm were used in the cutting tests. The cutting conditions are listed in Table 3. NC-
codes were programmed for machining the testpiece. The main NC-codes for the toolpath program
are summarized in Table 4. The power consumption was measured on Hitachi Seiki VG45 using
C

ELITEpro SP power meter.

AC

3.2. Monitoring and analysis

Power consumption of machine tool and milling process was monitored and captured. As shown
earlier in Figure 2, the power profile of milling process on Hitachi Seiki VG45 can be divided into each
energy state category which are basic, tool change, spindle, cutting, feed and cutting fluid. The basic
power, Pbasic, required for Hitachi Seiki VG45 at idle stage was on average 2070 W. This basic energy
is a constant part for machine tool centre.

Figure 4 shows the power profile for tool change. The characteristic power profile for tool change is
similar for each tool-call as controlled by the T-code in NC program. The energy consumption of tool
change (area under graph) can be assumed a fixed value for each tool call (T-code). Hence the total
energy consumption of tool change will be influenced by the number of tools used in machining one
component as in Equation (4). For Hitachi Seiki machine tool, the time taken for each tool change was
 ACCEPTED MANUSCRIPT
30 seconds and hence the average energy of 4.67 Wh. In this case study, the number of cutting tools
used in NC program is one.

Power consumption of spindle depends on the spindle design. The spindle model for this machine is
Hitachi spindle 45/4,500 rpm with rated main motor of 15 hp. To characterize this spindle model, the
power consumption of the spindle was measured at different spindle speeds without any cutting or
feed. Figure 5 shows the power consumption of the spindle running at different spindle speed. The
power required is a function of spindle speed and increases with spindle speed. The characteristic of
this model, in Figure 5, agrees with spindle-torque diagram information from machine tool manual that
there is a shift in trend at 1,000 rpm. This could be due to the shift of spindle gear. From this figure,

PT
the power consumption characteristic of this spindle model can be divided into two zones with the
transition at 1000 rpm. The first zone is for spindle speed range between 45 to 1,000 rpm and the
second zone ranges between 1,000 to 4,500 rpm. Thus the spindle power consumption for Hitachi

RI
Seiki VG 45 can be estimated depending on spindle speed (N) selection (S-code in NC program) by
Equation (13a) and (13b).

Zone 1: Pspindle = 0.2603N + 26.127 ; 45 ≤ N ≤ 1,000 rpm

SC
(13a)
Zone 2: Pspindle = 0.3252N + 553.97 ; 1,000 < N ≤ 4,500 rpm (13b)

Figure 6 illustrates power profile for cutting fluid pump motor of Hitachi Seiki machine tool. The cutting

U
fluid motor will be turned ON by M08 and turned OFF by M09 codes in NC program. The power
required to supply cutting fluid was 370 W.
AN
Power requirement for X, Y and Z axis movements at a given feedrate (G01 and F-code) is
demonstrated in Figure 7. Power consumption is in a function of feedrate. The feed power increase
with the increase of feedrate for X(+), X(-), Y(+), Y(-) and Z(+) directions while power consumption for
Z(-) direction movement reduces from the basic power requirement. This is due to gravitation. Hence
M

feed power can be estimated as a function of feedrate for each axis. Power consumption at rapid
movement (G00) is shown in Figure 8. The power consumption at rapid feedrate for X and Y axes is
similar and in the range of 160 to 175 W. The power required for Z-axis at rapid feedrate is different
D

for positive (+) and negative (-) directions follows the trend for Z-axis as shown in Figure 7(c).
TE

To estimate power requirement for cutting stainless steel, the specific cutting energy of material
needs to be identified. Figure 9 shows the power consumption required for machining stainless steel
using uncoated carbide tool at different material removal rate (different cutting speed). The specific
3
cutting energy of stainless steel was evaluated as 4.72 Ws/mm . The value is in the range reported in
EP

(Kalpakjain and Schmid, 2003). It is influenced by the interaction of cutting tool and workpiece
material. Thus the cutting power (Pcutting) required to machine stainless steel can be estimated by
Equation (7) to 500 W at 250 mm/min feedrate, 0.5 mm depth of cut and 50 mm width of cut.
C

The power consumption for this case study on Hitachi Seiki machine tool can be summarized and
shown in Table 5.
AC

3.3. Validation of energy model

Figure 10 (a) shows the power profile for the testpiece comparing the model estimate and the
measurement. It can be seen that the power model used to estimate the trend of the energy
consumption during the cutting process is based on NC program. However the model has some
limitations as it does not capture (i) the dynamic change of power due to the fluctuation in current, (ii)
the peak or high surge in electricity demand due to switching ON and switch OFF of machine
components such as spindle module and cutting fluid system. Additionally as expected the model
does not estimate tool wear. Figure 10 (b) shows the estimated energy profile in detail which linked to
the milling operation, tool engagement, cutting workpiece and axes movement.
 ACCEPTED MANUSCRIPT
Energy consumption for the case study was estimated according to Equation (1) to (13) where the
influence of tool wear was neglected in the models. The energy consumption for each component can
be estimated as shown in Table 6. The basic energy is the dominant part in the total energy
consumption in agreement with previous study by Gutowski (2006). In this case its contribution is
almost 73%.

The total energy demand on the testpiece using Hitachi Seiki VG45 machine tool was estimated to be
181.51 Wh and measured to be 191.52 Wh. The difference from the estimation to measurement was
5.23%. The reason for the deviation can be due to the tool wear progression which leads to increase
in specific cutting energy.

PT
The influence of tool wear can be observed more when repeated machining tests are done using the
st rd
same cutting tool. Figure 11 shows the power profile for three repeated cutting tests (from 1 cut to 3

RI
cut) on stainless steel testpiece. From this figure, it can be seen that the energy consumption
increases with the tool wear at the same cutting condition (machining profile, NC-program and cutting
parameters). This could be due to the influence of specific cutting energy that plays a major role

SC
between the tool and workpiece interface and causes higher cutting force and hence power and
energy consumption.

To conclude; total energy consumption in machining process depends on type of machine tool used

U
and the machining conditions (cutting speed, feedrate, material removal). This also allows the user to
estimate energy usage before actual cutting, especially for a new design component. The energy
AN
consumption model in Equations (1), composes of two main parts, which are the constant and
variable. Where the constant part can be measured and stored as a database for machine tools. The
basic or constant components are for example basic power of machine tool, cooling pumping system
and tool change. The variable parts are mainly due to the selection of cutting conditions (speed, feed)
M

and choice between cutting tool and workpiece materials. Table 7 shows a summary description of
energy model. This can be a platform methodology to build up a database for predicting energy
consumption of different components machined from the same or different machine tool. From Figure
D

11, it is clear that tool wear increases the power demand in machining and hence energy demand as
well. In addition to Table 7 which shows the database requirements for energy modelling, additional
information is needed to capture the effect of tool wear. This requires the scientific community to
TE

collaborate in developing mathematical models that capture how specific energy will vary with tool
wear for different workpiece and tooling combinations.
EP

Energy demand modelling and evaluation can be used to assess different toolpaths. Figure 12 shows
the energy consumption of pocket milling of stainless steel for different toolpath, cutting mode (up and
down milling) and axis movement. The cutting conditions are the same (spindle speed, feed rate,
depth of cut and width of cut). It can be seen that the energy consumption is cycle time dependent
C

due to the selection of toolpath and cutting mode. The assessment shows that in close pocket milling
contour offset strategy is the most energy efficient toolpath when compared to zigzag milling. The
AC

contour and spiral milling enables energy demand reduction by more than 5 times compared to zigzag
milling. This study shows that there are significant opportunities for reducing energy demand in
machining by selection of appropriate toolpath or optimising toolpaths.

Figure 12 shows that there are synergies between cycle time and energy demand in that reduced
cycle times lead to lower energy demand from the machining processes. Moreover, it is also shown in
Figure 12 that tool paths that produced the best surface finish also had reduced energy demand.
Again this showcases the possibility of achieving synergies between process performance and
environmental or resource efficiency objectives.
 ACCEPTED MANUSCRIPT
For the research community this paper has shown that fundamentally the modelling of energy
demand can be significantly affected by tool wear and tool paths. These two aspects are seldom
considered in research. The energy assessment of machine tools needs to be based on standardised
tool-paths and axis feed directions.

For the end user who manages a machining workshop, this paper shows that energy demand
modelling needs to be supported by appropriate databases. These databases should be populated
with the power demand models and constants for machine tools, the specific energy requirements for
the associated workpiece materials and tooling. Managing and reducing tool wear will have a
significant impact in reducing the energy intensity of machining processes. In addition, shorter cycle

PT
times and better surface finish can be achieved in synergy with reducing the energy demand of
machining processes and hence improving the environmental credentials of machined products.

RI
4. Conclusions

SC
The energy demand modelling in machining processes need more explicit modelling of energy states
in order to link energy prediction to toolpaths and capture the effect of cutting variables. This is vital in
order to raise the integrity of energy labels for machined products. In this paper, a methodology to
predict energy demand in mechanical machining processes was presented. The following points were

U
concluded.
AN
• Studying the energy profile of different machining events can be a useful way of developing more
robust models of the machining processes.
• The energy model was developed composed of two major energy states that have defined power
demand or profiles. These building block power states can be evaluated for machine tools and
M

stored as database for estimating the energy consumption.

• Total energy consumption can be estimated using the developed model presented in this paper.
The model can predict the trend of energy requirement for the process with 95% accuracy when
D

compared to online monitoring.

• The energy demand for G00 and G01, G02 and G03 can be modelled by studying the energy
TE

states of the axis.

• Tool wear plays a significant role in increasing tip energy demand. Thus process optimization to
minimise tool wear or advanced cutting tool development can help to promote more energy
efficient machining processes.
EP

• Selection of toopath strategies has a significant impact on cycle time, and can lead to significant
energy savings in machining processes. More studies are required to provide more comprehensive
guidance for industry.
C

• Environmental and manufacturing performance synergies are possible in that shorter cycle time
tool paths were associated with reduced energy demand and better surface finish on machined
AC

products.

Nomenclature

Unit
Ebasic Energy consumption for operating each machine tool centre e.g. Wh
servo system, fan motor
Ecutting-fluid Energy consumption for cutting fluid pump Wh
Ecutting Energy consumption for material removal of workpiece Wh
Efeed Energy consumption for axis feed Wh
r
E feed Energy consumption for axis feed at rapid Wh
f
E feed Energy consumption for axis feed at defined feedrate Wh
Espindle Energy consumption for operating spindle Wh
Etool Energy consumption for tool change Wh
 ACCEPTED MANUSCRIPT
Etotal Total energy consumption for NC machining Wh
Fc Cutting force N
3
K specific cutting power Ws/mm
ntool number of cutting tools used in the process or one cutting operation
N Spindle rotation speed rpm
Pbasic Power required for operating each machine tool centre e.g. servo W
system, fan motor
Pcutting-fluid Power required for cooling pump system W
Pcutting Power required for material removal of workpiece W
Pfeed Power required for table and cutting tool movement W
Pfeed i Power required for i-axis movement (X, Y or Z axis), W

PT
Pxr Rapid power for X-axis W

Pyr Rapid power for Y-axis W

RI
Pzr Rapid power for Y-axis W
Pspindle Power required for spindle run W
Ptool Power required for tool change W
3

SC
Q material removal rate mm /s
tspindle Total time required to run spindle s
tbasic Total time required throughout operational process or the NC s
program
tcutting-fluid Time required for activating cutting fluid s

U
tcutting Total cutting time for material removal s
tend Total time for operation process or the NC program s
AN
tstart Starting time of the operation process or the NC program s
ttool Time required for tool change (turret rotating time) s
T Cutting tool life s
Vc cutting velocity m/min
M

Vf feedrate mm/min
D

References
Abele, E., Eisele, C., Schrems, S., 2012. Simulation of the energy consumption of machine tools for a
TE

specific production task. in the 19th CIRP International Conference on Life Cycle Engineering,
Berkeley, California, USA, 23-25 May 2012, 233-237.
Avram, O.I., Xirouchakis, P., 2011. Evaluating the use phase energy requirements of a machine tool
system. Journal of Cleaner Production. 19, 699-711.
EP

Balogun, V.A., Aramcharon, A., Mativenga, P.T., Chuan, S.K., 2013. Impact of Machine Tools on the
Direct Energy and Associated Carbon Emissions for a Standardized NC Toolpath. in the 20th CIRP
International Conference on Life Cycle Engineering, Singapore, 17-19 April 2013, 197-202.
Balogun, V.A., Mativenga, P.T., 2013. Modelling of direct energy requirements in mechanical
C

machining processes Journal of Cleaner Production. 41, 179-186.

Balogun, V.A., Mativenga, P.T., 2014. Impact of un-deformed chip thickness on specific energy in
AC

mechanical machining processes. Journal of Cleaner Production, 1-9.

Bhushan, R.K., 2013. Optimization of cutting parameters for minimizing power consumption and
maximizing tool life during machining of Al alloy SiC particle composites. Journal of Cleaner
Production. 39, 242-254.
Bonvoisin, J., Thiede, S., Brissaud, D., Herrmann, C., 2013. An implemented framework to estimate
manufacturing-related energy consumption in product design. International Journal of Computer
Integrated Manufacturing. 26, 866-880.
Calvanese, M.L., Albertelli, P., Matta, A., Taisch, M., 2013. Analysis of energy consumption in CNC
th
machining centers and determination of optimal cutting conditions. in the 20 CIRP International
Conference on Life Cycle Engineering, Singapore, 17-19 April 2013, 227-232.
Campatelli, G., Lorenzini, L., Scippa, A., 2014. Optimization of process parameters using a Response
Surface Method for minimizing power consumption in the milling of carbon steel. Journal of Cleaner
Production. 66, 309-316.
CECIMO, 2009. Concept Description for CECIMO’s Self-Regulatory Initiative (SRI) for the Sector
Specific Implementation of the Directive 2005/32/EC.
 ACCEPTED MANUSCRIPT
www.energimyndigheten.se/Global/F%C3%B6retag/SRI_MachineTools_091016%20%282%29.pdf
(accessed: 30 November 2011)
CO2PE!, 2013 Cooperative effort on process emissions in manufacturing. www.co2pe.org/ (accessed:
October 2013)
Dahmus, J.B., Gutowski, T.G., 2004. An environmental analysis of machining. in the IMECE2004
ASME International Mechanical Engineering Congress and RD&D Expo, Anaheim, California USA,
13-19 November 2004, 1-10.
Diaz, N., Helu, M., Jarvis, A., Tonissen, S., Dornfel, D., Schlosser, R., 2009. Strategies for minimum
energy operation for precision machining. in the MTTRF- Annual Meeting, 1-5.
Diaz, N., Redelsheimer, E., Dornfeld, D., 2011. Energy consumption characterization and reduction
strategies for milling machine tool use. in the 18th CIRP International Conference on Life Cycle

PT
Engineering, Braunschweig, Germany, 2- 4 May 2011, 263-267.
Dietmair, A., Verl, A., 2009. A generic energy consumption model for decision making and energy
efficiency optimisation in manufacturing. International Journal of Sustainable Engineering. 2, 123-133.
Duflou, J.R., Sutherland, J.W., Dornfeld, D., Herrmann, C., Jeswiet, J., Kara, S., Hauschild, M.,

RI
Kellens, K., 2012. Towards energy and resource efficient manufacturing: A processes and systems
approach. CIRP Annals - Manufacturing Technology. 61, 587-609.
EPTA, 2007.Study for Preparing the First Working Plan of the Eco-design Directive. (accessed: 20

SC
November 2013)
Frigerio, N., Matta, A., Ferrero, L., Rusinà, F., 2013. Modeling Energy States in Machine Tools: An
th
Automata Based Approach. in the 20 CIRP International Conference on Life Cycle Engineering,
Singapore, 17-19 April 2013, 203-208.
GardnerResearch, 2013.The World Machine Tool Output & Consumption Survey

U
https://www.gardnerweb.com/cdn/cms/uploadedFiles/2013wmtocs_SURVEY.pdf (accessed: 21
October 2013)
AN
Gutowski, T., Dahmus, J., Thiriez, A., 2006. Electrical Energy Requirements for Manufacturing
Processes. in the 13th CIRP International Conference on Life Cycle Engineering, Lecture no.2 in
Cleaner Production (1) session, Leuven, Belgium, 31 May-2 June 2006.
Hao, C.W., 2011. SMEs can do bit for energy conservation, too. siew.sg/node/666 (accessed: 25
M

November 2011)
He, Y., Liu, F., Wu, T., Zhong, F.-P., Peng, B., 2012. Analysis and estimation of energy consumption
for numerical control machining. Proc. IMechE Part B: J. Engineering Manufacture. 226, 255-266
Hu, S., Liu, F., He, Y., Hu, T., 2012. An on-line approach for energy efficiency monitoring of machine
D

tools Journal of Cleaner Production. 27, 133-140.

IEA, 2013. World electricity consumption in 2011.
TE

www.iea.org/statistics/statisticssearch/report/?country=WORLD&product=electricityandheat&year=20
11 (accessed: 21 October 2013)
Jeswiet, J., Kara, S., 2008. Carbon emissions and CES in manufacturing. CIRP Annals -
Manufacturing Technology. 57, 17-20.
th
Kalpakjain, S., Schmid, S.R., 2003. Manufacturing Processes for Engineering Materials, 4 ed.
EP

Prentice Hall, New Jersey, the United States of America.

Kara, S., Li, W., 2011. Unit process energy consumption models for material removal processes.
CIRP Annals - Manufacturing Technology. 60, 37-40.
Kong, D., Choi, S., Yasui, Y., Pavanaskar, S., Dornfeld, D., Wright, P., 2011. Software-based tool
C

path evaluation for environmental sustainability. Journal of Manufacturing Systems. 30, 241-247.
Kordonowy, D.N., 2002. A power assessment of machining tools. Massachusetts Institute of
AC

Technology, Massachusetts, USA.

Li, W., Kara, S., 2011. An Empirical Model for Predicting Energy Consumption of Manufacturing
Processes: A Case of Turning Process Proceedings of the Institution of Mechanical Engineers, Part
B: Journal of Engineering Manufacture. 225, 1636−1646.
Li, W., Zein, A., Kara, S., Herrmann, C., 2011. An investigation into fixed energy consumption of
machine tools. in the 18th CIRP International Conference on Life Cycle Engineering, Braunschweig,
Germany, 2-4 May 2011, 268-273.
Mativenga, P.T., Rajemi, M.F., 2011. Calculation of optimum cutting parameters based on minimum
energy footprint. CIRP Annals - Manufacturing Technology. 60, 149-152.
Mori, M., Fujishima, M., Inamasu, Y., Oda, Y., 2011. A study on energy efficiency improvement for
machine tools. CIRP Annals - Manufacturing Technology. 60, 145-148.
Neugebauer, R., Wabner, M., Rentzsch, H., Ihlenfeldt, S., 2011. Structure principles of energy
efficient machine tools. CIRP Journal of Manufacturing Science and Technology. 4, 136-147.
 ACCEPTED MANUSCRIPT
Rajemi, M.F., Mativenga, P.T., Aramcharoen, A., 2010. Sustainable machining: selection of optimum
turning conditions based on minimum energy considerations Journal of Cleaner Production. 18, 1059-
1065.
TSC®, 2013. The Sustainability Consortium® www.sustainabilityconsortium.org/ (accessed: 21
October 2013)

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 1: Energy model for machining process

Model Reference Energy Model Remarks

M1 Gutowski et E = ( P0 + kQ )t Pioneer work, needs expansion to
al. (2006) model different aspects of Basic and
Where
Tip Power.
P0 is power consumed by machine module, k is specific
cutting energy requirement for particular workpiece material
Q is material removal rate, t is the cutting time
M2 Rajemi et al. t  t  Model focus on energy footprint and
(2010) E = P0 t1 + ( P0 + kQ )t 2 + P0 t 3  2  + y E  2  not direct energy demand.
T  T 
Where

PT
P0 is power consumed by machine module without cutting
t1 is the machining setup time, t2 is the actual cutting time
t3 is the tool change time, T is tool life, yE is energy footprint
per tool edge, k is specific cutting energy, v is material
removal rate
E = P1 (T1 + T2 ) + P2 (T2 ) + P3 (T3 )

RI
M3 Mori et al. Does not explicitly model tip power to
(2011) capture work piece material
Where
differences.
P1 is the constant power during machining operation
Effect of spindle speed not captured
regardless of running state (basic power), P2 is power for
Tool life and tool change not

SC
cutting by spindle and servo motor, P3 is the power
captured.
consumption to position the work to accelerate/decelerate
Feed axis not considered.
the spindle to the specified speed, T1 is the cycle time during
non-cutting stage, T2 is the cycle time during cutting stage,
T3 is the time required for position the work and to accelerate
the spindle.

U
M4 Diaz et al. E = (Pcut + Pair ) ⋅ ∆T Captures air cutting, but does not
(2011) explicitly model other Basic and Tip
Where
AN
power component details
Pcut is the power demand for cutting
Pair is the power demand for air cut .
∆T is the processing time
M5 Kong et al. E = E const + E run − time − transient Does not provide the details for
(2011) individual feed axis movement, tool
+ E run − time − steady + E cut
M

change and coolant.

Where
Econst is the energy consumed not related to machining
Erun-time-transient is the energy consumed by spindle, machine
D

axes and tool change when accelerating or decelerating to

reach specified values, Erun-time-steady is the energy consumed
the by spindle, machine axes and tool change when the
spindle motor and the axis drives keep a specified value
TE

Ecut is the energy consumed by material removal action

M6 Kara and Li E = SEC ⋅ CuttingVolume Does not explicitly or separately
(2011) C model machine tool and workpiece
SEC = C0 + 1 machinability.
EP

MRR
Where
SEC is specific energy consumption, C0 and C1 are machine
specific coefficients and MRR is material removal rate
M7 He et al. E = Espindle + Efeed + Etool + Ecoolant + Efix It is in-line with the model proposed in
C

(2012) this paper. However the model does

Where not explicitly model tip energy i.e
Espindle is the energy consumption of spindle including the capture workpiece machinability.
spindle transmission module and energy requirement for
AC

material removal from workpiece, Does not capture energy demand

Efeed is the energy consumption of axis feed, which depends on spindle speed, the
Etool is the energy consumption of tool change, number of tool changes or
Ecoolant is the energy consumption of tool change simultaneous axes movement.
Efix is the energy consumption of fan motor and servo system
M8 Calvanese E = E fixed + E axes + E axis chillers + E spindle The model does not explicitly model
et al. (2013) the energy consumption of coolant
+ E spindle chiller + E chip conveyor pumps, tip energy and material
machinability aspects, multiple axis
+ E tool change + E pallet clamp
movement.
Where
Efixed is the energy required when machine is switch on
Eaxes is the energy required for axis which composes of rapid
movement, approach workpiece at working speed and
during machining
Eaxes chillers is the energy consumed by axes chillers
Espindle is the energy of both the cutting power and the spindle
 ACCEPTED MANUSCRIPT
losses
Espindle chiller is the energy consumed by spindle chillers
Echip conveyor is the energy consumed when chip is removed
Etool change is the energy consumed for tool change
Epallet clamp is the energy consumed for clamping the pallet

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
Table 2: Relationship between NC code and energy consumption on 3-axis machine tools
Machine
NC code Operations CNC Codes
Components
Rapid move G00
G01 (linear)
G-Codes X, Y, Z Axis feed G02 (circular interpolation, clockwise)
Motion with defined feed rate
G03 (circular interpolation, anti-
clockwise)
F-Code X, Y, Z Axis feed Defined feedrate (for feed motor) F
Spindle motor on M03, M04
Spindle
Spindle motor off M00, M01, M02, M05, M30
M-Code Axis feed Stop axis feed motor M00, M01, M02, M30

PT
Coolant on M07 (mist), M08 (flood)
Coolant
Coolant off M00, M01, M02, M09, M30
S-Code Spindle Define spindle speed S
T Tool change Tool change T

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 3: cutting parameters and condition
Parameters
Spindle speed (rpm) 830
Feedrate (mm/min) 50 and 250
Depth of cut (mm) 0.5
Width of cut (mm) 50
Environment condition Flood coolant

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
Table 4: Detail of NC-codes, sequence and information of cutting toolpath
NC-code Characteristics Position
G54 G90 G00 X120 Y30 Work coordinate system, Absolute programming -
M03 S830 Spindle turn ON -
G43 H16 Z10 Tool height offset compensation O’
M08 Coolant pump turn ON -
G01 Z0.5 F200 Move in Z(-) direction (Tool approach) O’
O
G01 X30.0 Z-0.5 F50 Move in XZ (--) direction (Modal tool engagement) O
A
G01 Y120 F250 Move in Y(+) direction A
B
G01 X120 F250 Move in X(+)direction B
C
G01 Y30 F250 Move in Y(-)direction C
D
G01 X30 F250 Move in X(-) direction D
A

PT
G00 Z50 Move in Z(+) direction Rapid (Tool retract) A
F
M09 M05 Coolant turn OFF, Spindle turn OFF -
M30 End program -

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 5: Power consumption for case study using Hitachi Seiki VG45 machine tool

Power Characteristics W
Pbasic Basic power 2,070
Ptool Tool change 560
Pspindle Spindle speed at 833 rpm 243
Pcutting-fluid Cutting fluid 370
Pfeed, Pxf X(+) at 250 mm/min 7
X(-) at 250 mm/min 9
Pfeed, Pyf Y(+) at 250 mm/min 6
Y(-) at 250 mm/min 6
Pfeed, Pzf Z(+) at rapid 808
Pcutting Cutting at 50 mm/min (tool engagement) 50

PT
Cutting at 250 mm/min 500

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 6: Comparison of energy consumption between estimation and measurement for the case study

Wh
Estimation %
Ebasic 132.25 72.86
Espindle 15.52 8.55
Ecutting-fluid 21.38 11.78
Efeed 0.41 0.22
Ecut 11.95 6.59
Etotal 181.51
Measurement
Etotal 191.52
% Different between

PT
estimation and measurement 5.23

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 7: Description to develop energy model for machining process

Machine modules/Activities Characteristic Recommendation

Pbasic Power for operating machine Constant for machine tool Database for each machine tool centre
tool e.g. servo system, fan
motor, computer
Ptool Tool change Assumed as fixed value for each tool Database for tool turret for changing
change cutting tools
Pspindle Spindle speed Depend on spindle model Identify spindle speed function for
spindle speed range spindle model
Pcutting Removing workpiece material Interaction between tool and Identify the specific cutting energy for
workpiece each condition of cutting tool and
workpiece

PT
Pfeed Working table and cutting tool Depend on machine tool movement Identify feedrate function for X, Y and Z
movement in X, Y and Z Function of feedrate axis
directions
Pcutting- Cutting fluid pumping system Constant for cooling system Database of coolant system for each
machine tool centre

RI
fluid

U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Input Servo drive Output

Fan
Chiller system Loading
Hydraulic system Auxiliary
Control system

Product
Workpiece

PT
Cutting tools

Spindle drives, Axis

movement, Cooling Cutting energy

RI
system, Tool change

Figure 1: Selected boundary (dashed line) for the study of energy consumption inclusive of machine

SC
tool and cutting process

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

4
Power (kW)

PT
3 PCutting
Pready (Pcutting-fluid & Pspindle)

RI
Pspindle
2 Pbasic

SC
1

U
0
0 500 1000 1500 2000
AN
Time (s)

Figure 2: Power profile of milling process from Hitachi Seiki VG45 machine tool
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
Figure 3: Toolpath for testpiece

RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

4000 T-code

3000

Power (W)
2000

1000
1st Tool call

2nd Tool call

0

PT
0 20 40 60 80 100
Time (s)

Figure 4: Power consumption for tool change

RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

2000

1500

Power (W)
y = 0.3252x + 533.97
1000

500

y = 0.2603x + 26.127
0

PT
0 1000 2000 3000 4000 5000
Spindle (rpm)

RI
Figure 5: Power consumption characteristic for Hitachi Seiki VG45 spindle module varies by
spindle

U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
6000 Without cutting fluid
With cutting fluid (M8)

5000

4000

Power (W)
3000
Pcutting-fluid
2000

PT
1000
-25 50 125 200 275
Time (sec)
M08, tcutting-fluidstart M09, tcutting-fluidstop

RI
Figure 6: Power profile for cutting fluid pump

U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
200 200 1250
X+ Y+ Z+
X- Y- Z-
1000
150 150
y = 0.032x + 1.375
Power (W)

y = 0.029x - 1.601 750

Power (W)

Power (W)
100 100 500
y = 0.198x + 59.063
y = 0.028x - 1.405 250
50 y= 0.032x - 1.191
50
0 y = -0.0135x - 73.517
0 0 -250
0 1000 2000 3000 4000 5000 0 1000 2000 3000 4000 5000 0 1000 2000 3000 4000 5000
Feed (mm/min)

PT
Feed (mm/min) Feed (mm/min)

(a) (b) (c)

200 200
X+ X-
Y+ Y-
XY ++ XY --

RI
150 150
y = 0.037x + 5.134
Power (W)

Power (W)
y = 0.037x + 2.688
100 100

SC
50 50

0
0
0 1000 2000 3000 4000 5000
0 1000 2000 3000 4000 5000

U
Feed (mm/min) Feed (mm/min)

(d) (e)
AN
Figure 7: Power consumption of (a) X-axis, (b) Y-axis, (c) Z-axis, (d) XY++ and (e) XY—movements at
defined feedrate
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

1000
+
800 -

600

Power (W)
400

200

-200

PT
X Y Z
Axis movement

Figure 8: Power consumption of (a) X-axis, (b) Y-axis, (c) Z-axis movements at rapid feedrate (G00)

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
600

500

400

Power (W)
300 y = 4.7167x + 239.79

200

100

PT
0 20 40 60 80
Material removal rate (mm3/s)

Figure 9: Power consumption for machining stainless steel

RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
3200
7000 3400
1st Cutting D
B C A B C D
Estimated 3190 A
3200
6000
3180
3000 125 150 175 200 225
A

5000
Power (W)

Power (W)
2800 F
O
4000
2600

PT
3000
2400

2000

RI
2200

1000 2000
-25 25 75 125 175 225 275 -25 25 75 125 175 225 275

SC
Time (s) Time (s)
M03 Move M09
M03 start M09 S833 XZ Move & Cut in Y M05
M08
cutting M08 start cutting

(a) (b)

U
Figure 10: (a) comparison between estimated and measured power consumption and (b) estimated
power consumption profile in detail
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

8000
1st Cutting
2nd Cutting
7000
3rd Cutting

6000 3rd Cut

Power (W)

5000 2nd Cut

4000 1st Cut

PT
3000

2000

RI
1000
-25 25 75 125 175 225 275

SC
Time (s)

M03 M08 start cutting M09

S833 M05

Figure 11: Power profile when milling stainless steel testpiece

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Pocket milling Zag x-axis Zigzag x-axis Zag y-axis Contour Spiral
Stainless steel
N: 3,000 rpm
Vf: 300 mm/min
ap: 1 mm
ae: 2.5 mm
i: down milling iii iv V Vi
ii: up milling

250 0.26
Energy Energy
250 Time 250 Ra
0.24

PT
200
0.22

Energy (Wh)
Energy (Wh)

200

Time (sec)
200

Ra
0.20
150

RI
150 150 0.18

100 0.16

SC
100
100 i ii iii iv v vi
i ii iii iv v vi
Toolpaths
Toolpaths

Figure 12: Energy consumption for different toolpath and cutting strategy selection

U
AN
M
D
TE
C EP
AC