Awareness of Energy Consumption in ... - ScienceDirect.com

0 downloads 0 Views 848KB Size Report
Abstract. Several literature have reported on the low level of awareness for energy .... is a framework documented in the book by Blessing and Chakrabarti [7].

Available online at www.sciencedirect.com

ScienceDirect Procedia Manufacturing 8 (2017) 152 – 159

14th Global Conference on Sustainable Manufacturing, GCSM 3-5 October 2016, Stellenbosch, South Africa

Awareness of energy consumption in manufacturing processes Oladele Owodunnia* a

Faculty of Engineering and Science, University of Greenwich, Chatham Campus, ME4 4TB, UK

Abstract Several literature have reported on the low level of awareness for energy consumption in manufacturing processes. This paper presents results aimed at providing some evidence for this low level of awareness. The results of energy consumption as measured by various metering devices on different manufacturing equipment are presented. Predictive equations of energy consumption are also presented. The low awareness and meaningfulness of this results among a spectrum of students and manufacturing practitioners is reported to indicate the existence of a gap between the measured/predicted results and the human perception of energy consumption. Reasons for the gap between measurements/prediction and perception as well as methods for bridging the gap are proposed. The results are evidenced by experiments in turning, milling and rapid prototyping and have been conducted on manufacturing equipments which include manual desktop machine tool, desktop CNC machines and industrial scale manual and CNC machines © 2016 The Authors. B.V.This is an open access article under the CC BY-NC-ND license Crown Copyright © 2017Published Publishedby byElsevier Elsevier B.V. Peer-review under responsibility of the organizing committee of the 14th Global Conference on Sustainable Manufacturing. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of the 14th Global Conference on Sustainable Manufacturing Keywords: Energy consumption; Energy awareness; machining; sustainable machining

1. Introduction It is now widely accepted that the reduction of energy consumption in industry is a performance measure to be considered along with the traditional measures of cost, delivery time and quality. Various targets to reduce energy consumption have therefore been set and being constantly monitored at national, continental (e.g. in EU countries) and international levels. It has been reported that manufacturing accounts for over 30% of global CO2 emissions and energy consumption [1], a trend that is growing, especially in emerging industrial economies like China, India and

* Corresponding author. Tel.: +44 (0)1634 88 3576 E-mail address: [email protected]

2351-9789 Crown Copyright © 2017 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of the 14th Global Conference on Sustainable Manufacturing doi:10.1016/j.promfg.2017.02.018

Oladele Owodunni / Procedia Manufacturing 8 (2017) 152 – 159

Brazil. The adverse impact of manufacturing activities on global sustainability means the manufacturing community need to be at the forefront of responding to the global sustainability challenge. This response is reflected in the various global sustainable manufacturing research activities and international standards (e.g. ISO 14000, ISO 50000, ISO 14955 series). With the vast amount of growing literature in the field of sustainable manufacturing, most of these were only conducted in an academic environment and those tested in industrial environments do not give indication of sustained adoption in those settings. This paper addresses this problem, with a focus on how academic results in measuring/predicting energy consumption in manufacturing processes could be used to raise energy awareness and auditing in industrial environments. 1.1. Aim and structure of paper The paper reports a number of related investigation aimed at improving the energy awareness and adoption of energy efficiency measures in industry. Specifically the research question is: ”What are the technical factors affecting a common-man understanding of energy consumption in manufacturing processes and what ways can be used to address this problem?” After presenting the literature review in section 2, the paper considers the design research methodology employed in section 3. The background to the research reported in this paper is the results (presented in section 4) of the laboratory-based investigations that measured and predicted energy consumption in various manufacturing operations in addition to an effort to disseminate it in industry. Section 5 then considers the set of investigations carried out to address the problem of making energy measurement industrially-relevant. 2. Review of the literature From the pioneering works in the early 1990s, the literature on sustainable manufacturing is now vast and growing. The fragmentation have been identified by some researchers [2] and they suggested that the use of ontology may address this problem. Various literature review papers have made contributions in identifying some structure in the field, an example of which is a review paper by a group of researchers from the International academy of production research [3] which has identified the following 3 areas that the contributions in sustainable manufacturing can be classified into: x Performance measures for sustainable manufacturing and their prediction/measurements; x Improvement of performance of sustainable manufacturing through optimizing/improving the process parameters and x Improvement of performance of sustainable manufacturing through designing new the process technologies/methods. The key performance measures identified in literature that relates to energy consumption are power, energy consumed and energy consumed in processing unit volume of material (referred to as Specific Energy Consumption, SEC). Researchers such as Gutowski and collaborators at MIT [4] carried out some of the earliest comprehensive investigations, documenting these measures for a wide range of manufacturing processes. Various definitions of energy efficiency or Efficiency Ratios, ER, were addressed by researchers such as Rahimifard et al. [5]. Currently, there are over 50 performance measures and having these streamlined and in a form that can be transferable across industrial activities is still a challenge. Factors that affect the adoption of energy efficient manufacturing have been considered by a number of researchers. Cagno and Trianni [6] identified a mixture of technical and non-technical factors which include: Technology-related, economic, organisational, behavioural, competence-related and awareness. The importance of information and how it is presented has been identified as being significant to raising awareness of energy efficiency issues. The technical factors that affect how practitioners understand energy consumption information is therefore considered important to address. It is however understood that this problem needs to be considered within a framework that promotes shared understanding so as to avoid the fragmentation developing in the field mentioned

153

154

Oladele Owodunni / Procedia Manufacturing 8 (2017) 152 – 159

earlier by other researchers [2]. Developing research within methodological frameworks shared by the research community could go a long way in addressing fragmented outputs. Also the problem of addressing how energy measurement information can be presented in industrially-relevant form relates to having research methodology that enables such research to be conducted (i.e. having a research methodology that allows industrially-relevant research to be conducted). These areas of needs are addressed in the research presented in this paper. 3. Methodology It was mentioned in section 1.1 that the methodology employed is an adaptation of the design research methodology. It is referred to as design research in that it is usually focused on designing solutions to problems in the man-made world as against just gaining understanding of a phenomenon. Perhaps beginning with the ideas articulated by the Nobel Laureate, Hebert Simon in the book titled “the science of the artificial”, several engineering/technology researchers have contributed to the development of a design research methodology. One of the most well-known of these contributions is a framework documented in the book by Blessing and Chakrabarti [7]. The methodology presented in this paper adapts that framework as shown in fig. 1 with the cycles through it shown (note the iterative model like spiral model). The cycles followed in the research programme, part of which is reported in this paper is shown by the numbered arrow-ended lines. In cycle 1, the manufacturing process with consideration of energy was investigated through experiments, mathematical modeling to obtain output in the form of descriptive/explanatory and predictive models which were represented as text augumented mathematical equations, graphs (shown as 2). This understanding was used to prescribe improvement through optimisation of process parameters and design of new processes/technologies (shown as 4) yielding optmisation algorithms, charts and implemented computer programs (5). Descriptive-II carried out evaluation of the prescriptive output (in 6) including exploratory dissemination in practice environment led to studies (9) which yielded outputs 10 which became the result of needing to seek more understanding (11), output of which (12) is expected to align more closely with practice (13). NATURAL SCIENCES & HUMAN STUDIES

DESCRIPTIVE-I

11

DESCRIPTIVE-II

PRESCRIPTIVE

1

2

SOCIETY

PRACTICE

9 4

3

6 5

8 13

7 10

12 Fig 1. Adaptation of Design Research Methodology-including cycles of activities

4. Results and Discussions I- Energy efficient manufacturing results & its dissemination in industry 4.1. Academic results in energy efficient manufacturing This aspect of the research project identified the performance measures and other factors for various manufacturing processes such as milling, turning, drilling, rapid prototyping, injection moulding and laser profiling. How these performance measures and factors are defined, represented and measured were considered. Some of the key sustainability performance indicators considered are power, energy consumed, Specific Energy Consumption, SEC and energy efficiency. Also considered are traditional performance measures such as manufacturing cost, manufacturing time and quality (e.g. as measured by the surface finish).

155

Oladele Owodunni / Procedia Manufacturing 8 (2017) 152 – 159

Methods for measuring energy consumption without disrupting the manufacturing operation has been investigated through the use of 3-phase power meters (e.g. for machining) and single phase power meters (e.g. for rapid prototyping equipments). For machining, the results from the power meter were checked by using a force measurement system from Kistler Instruments. For some of the experiments such as for turning and rapid prototyping, they were exploratory and were only checked for broad reasonableness of the results and order of magnitude accuracy. The milling experiment was checked more rigorously as predictive models were to be derived from them. Using the results from the milling experiments, a predictive model for Specific Energy Consumption, SEC, has been adapted from machining science. Some of the coefficients have been obtained by fitting mathematical functions to measured data. The mathematical expression resulting is shown in equation 1 (where ap: is depth of cut; ae: width of cut; d:diameter of end mill; z:number of teeth; f:feed rate; n:spindle speed; c0-c6 are constants depending on material being machined).

(1) The SEC for a number of the processes are shown in fig. 2. The general form corresponds to the results obtained by other researchers such as that of Gutowski et al [4]. Comparison with the results of other researchers have been shown to confirm and corroborate some general results emerging in the research community. The specific energy consumption decreases with increase in material processing rate for all types of manufacturing processes. This result which is descriptive of some interesting characteristics of manufacturing processes has some predictive and prescriptive power as well. For example, given some requirements (e.g. some given specific energy consumption), the set of possible parameters (e.g. material processing rate and hence other process parameters) that could be “prescribed” can be identified. Using the descriptive/predictive/explanatory outputs of earlier stage, the prescriptive stage was to prescribe a way of improving the sustainability (e.g. energy-efficiency) of the manufacturing processes by identifying appropriate process parameters. This was solved as an optimization problem with several approaches explored (e.g. calculus-based method, direct search method, design of experiment techniques and genetic algorithm). The results obtained from the optimisation is graphically shown in fig. 3 for a case where the objective was to minimize energy consumption subject to constraints such as meeting a specified surface quality and avoiding high cutting force that could cause tool breakage. Compared to recommendations of tool manufacturers and other handbooks, the result gives over 60% improvement in energy consumption, time and cost. The prescriptive result was implemented in a form that is similar to that which practitioners are familiar with as shown in fig. 4. This is to make dissemination to them easy. Figure 3shows cutting parameters (spindle speed, federate) and 3 objectives/constraints (energy, cutting force and surface finish). The hatched area in fig. 3 shows feasible region when constraints are applied.

Fig 2. Specific Energy Consumption at different material processing rates for various manufacturing processes

156

Oladele Owodunni / Procedia Manufacturing 8 (2017) 152 – 159

Specific Energy Consumption kJ/cc

Surface roughness constraint

Cutting force constraint Optimum point

Fig. 3: Graphical representation of optimisation results

Fig. 4: Interface for sustainable machining

Oladele Owodunni / Procedia Manufacturing 8 (2017) 152 – 159

4.2. Dissemination of research in industry and methodological implications Efforts have been made to disseminate the results in industry. Over 40 practitioners and 5 industrial study visits/informal interviews were conducted to disseminate the results and obtain the views from potential industry users. There was also a 6 month case study carried out by a practitioner in his industry. Some awareness of the relative magnitude of energy consumed in machining processes was an eye opener to the practitioners interviewed. All practitioners interviewed were surprised to find out that the energy consumed by the lighting bulb on a machine tool or the energy required to pump the cutting fluid is much more than that required to carry out the machining operation. However, the exact results seem not to have an impact. The understanding of the author is that propositional facts representation and graphs are not the format for establishing credibility of results. Practically testing out results seems a more natural route to knowledge for the practitioners interviewed. Also the sophisticated metering methods adopted in the laboratory-based results were found out not to be transferable to industrial setting. One reason is because it required each equipment to have their own power socket which was not the case in all the industries observed. This understanding from industry made us to question whether the results obtained from our study could actually be said to be good enough. From a methodological point of view, would it have been better in fig 1. that the starting point (i.e. “1”) is placed in the practice environment? Alternatively is the results from the industry just a natural part of the iteration spiral? While such iteration spiral is acceptable within a long term research programme such as that reported by the author here, would it have been problematic for a short term project such as a PhD program? 5. Results and Discussions II-Industrial-relevance of energy metering and calibration The methodological issues with industrial relevance of the academic results obtained in this research have been briefly considered in section 4.2. This section investigates factors that could have affected the energy metering methods being industrially applicable to typical practitioners. In this investigation, factors that do not allow typical practitioners to understand information on energy metering are identified (section 5.1 report this) and ways of addressing them suggested (as presented in section 5.2) - i.e. the academic results, especially for energy metering, were investigated to identify if there are ways to make them more easily acceptable to most manufacturing practitioners. Methods employed is a mixture of experiments, reflections and informal interviews of participants in the studies. The studies were carried out in different order and only presented in the sequence below for ease of understanding. 5.1. Problem with sophisticated metering and calibration With respect to the power meters that were used for results presented in section 4.1, there were times when they gave very spurious readings and these were only detected because the readings were so far off from the power rating of the main motors of the machine tools. Except there is a mechanism for ensuring the meter reading can be transparently traced to the measured quantity, reliability of the results cannot be ensured. This complicated measuring method has become necessary because humans are no more in the loop with the measured quantity. This may mean a simple measurement method may not be possible anymore in mechanized and computerized manufacturing equipment. Practitioners and students may have to be educated about this reality of working with advanced manufacturing equipment. As mentioned in section 4.1, energy/power measurements (e.g. in milling operation) were obtained using more than 1 method (e.g. power meter and force measurement method). Also, independently conducted experiments were used to check the results. To avoid having to take a lot of readings only to find out that systematic errors make them unusable, research participants were instructed to calculate expected theoretical values (even if it is only an approximate one) to be used as an order-of-magnitude baseline. The need for these precautions in using instruments

157

158

Oladele Owodunni / Procedia Manufacturing 8 (2017) 152 – 159

often seem unnecessary to the students who participated in the experiments. This is often due to a naïve trust in readings provided by scientific instruments (without realizing the responsibility needed by the observer to calibrate each of the elements in the measurement chains). While the precautions to follow in energy measurements mentioned above are needed, intuitive ways of addressing them may be identified. So that the reading presented by power meters or other means are not just abstract numbers, some intuitive grounding could be explored (i.e. what in lay-man language is 1 W or 1 Joule?). An intuitive grounding that was found common was connecting power measurement to the power the average human being could muster or the daily recommended calorie intake for an adult. This could be taken further by using familiar objects such as the ergo-cycle found in homes or fitness centers whose readings could indicate that most adults could muster 50-75W of power in sustained activities (e.g. a 8-hour working day). Another intuitive grounding was the identification by the participants in the possibility of using an electric bulb of known wattage for rough calibration of single phase equipments (e.g. desktop machine tools or rapid prototyping machines) investigated. 5.2. Shop floor metering opportunities that can be used 5.2.1. Using central metering in manufacturing establishments This would be a good starting point as all manufacturing establishments have this metering device and due to similarity with metering device in homes, all people have some understanding and trust in the results provided by these devices. But the problem with this is that, from the experience of SMEs visited for observational studies, there is usually only one or few meters for the whole organization and all the equipments are connected to it directly, including the office equipments such as computers. Even with this, it is possible, if the number of equipments is not too large that this central reading can be used to identify the contribution from each equipment. The implementation would have to be simplified to a level similar to the method employed in multi-apartment accommodations with single electricity metering and using points allocated to each device (e.g. a 60 watt bulb has p points, an electric iron has q points) for apportioning the electric consumption of each apartment. It could be simplified for companies by having a database of different brands of manufacturing equipments with their point range (e.g. a Colchester ABC lathe has points ranging from X to Y). While this will enable energy accounting or auditing that is easy to implement, it would not necessarily on its own improve the gut level understanding of practitioners. 5.2.2. Calibrating energy/power sensors on machine tools Increasingly, modern machine tools have more sensors for monitoring various aspects of their performance. These vary from those giving load percentage on each of the axes of the machine or some that give power readings of main axes. From our observational studies (reported in section 4.2), there would be a need to educate practitioners on what these features on machine tools stand for and how best to use them. Efforts to obtain information from representatives of machine tools manufacturer have not yielded much useful results, indicating a need for these to also be included in the educational programme. To make sense of these sensors on machine tools, exploratory calibration experiments were carried out for a Haas TM-1 CE CNC milling machine. The machine offers facility for displaying the load percentage for each axis (X, Y, Z and Spindle) during machining. The experiments used a 3-component force sensing system from Kistler. Fig. 5 shows an example of results obtained for the Y axis. The object of the exploratory experiments was not to achieve accurate calibration, but to establish if there is repeatability in the correlation between the load % and the readings obtained from force measurement system. From fig. 5, there is some good degree of agreement in the early part of the operation but the later part of the operation do not correlate as much. There are factors such as the entry and exist of tool into work material that affects the quality of readings obtained. This needs to be further studied. While it would be possible to use the on-machine sensors to estimate the energy employed in machining, the energy for auxiliary functions (e.g. coolant pumps) which have been shown in several studies to constitute as much as 70% of energy consumed would not be accounted for.

Oladele Owodunni / Procedia Manufacturing 8 (2017) 152 – 159

Fig. 5. Correlation between load percentage readings on machine tool and force measurement

6. Conclusions This paper has investigated the current state of academic understanding of energy consumption in manufacturing processes and technical factors that could be important in disseminating these results to industry. The findings include the following: x x x x x

There is a convergence of understanding that the Specific Energy Consumed (SEC) follows similar trends for various manufacturing processes, reducing as material processing rate gets higher. Industry metering is shown to still be problematic and academic energy metering results cannot be directly transferred to industrial practice. Common man understanding of energy consumed and how to meter it for manufacturing processes is affected by the out-of loop effects due to mechanization and automation. Possibility of common man understanding could be achieved using intuitive understanding of the magnitude of power measured. It is possible to calibrate the sensors currently available on machine tools and hence use them for estimating energy consumed in machining operations in industrial environments.

References [1] J.M, Allwood, M. F. Ashby, T. G. Gutowski, E. Worrell, Material efficiency: a white paper. Resources, Conservation and Recycling. 2011 Jan 31;55(3):362-81. [2] M. Dassisti, M. Chimienti, M. Shuaib, F. Badurdeen, I. S. Jawahir. Sustainable manufacturing: A framework for ontology development. In Sustainable Manufacturing 2012 (pp. 33-39). Springer Berlin Heidelberg. [3] J.R Duflou, J.W. Sutherland., D. Dornfeld, , C. Herrmann, J. Jeswiet, S. Kara, , M. Hauschild, K.Kellens, 2012. Towards energy and resource efficient manufacturing: A processes and systems approach. CIRP Annals-Manufacturing Technology, 61:2(2012):587-609. [4] T. Gutowski, J. Dahmus, A. Thiriez. Electrical energy requirements for manufacturing processes. In13th CIRP international conference on life cycle engineering 2006 May 31 (Vol. 31, pp. 623-638). [5] S. Rahimifard, Y. Seow, T. Childs. Minimising Embodied Product Energy to support energy efficient manufacturing. CIRP AnnalsManufacturing Technology, 59:1(2010):25-28. [6] E. Cagno, A. Trianni, Evaluating the barriers to specific industrial energy efficiency measures: an exploratory study in small and mediumsized enterprises. Journal of Cleaner Production, 82(2014):70-83. [7] L. T. Blessing, A. Chakrabarti, DRM: A Design Research Methodology. Springer London; 2009.

159