Direct Illuminance-Contribution-Based Lighting Control for IoT-Based Lighting Systems in Smart Buildings
Round 1
Reviewer 1 Report (Previous Reviewer 1)
Comments and Suggestions for AuthorsThe revised version has been remarkably improved and the paper can be accepted.
Author Response
Thank you for taking your valuable time to review my paper.
Reviewer 2 Report (Previous Reviewer 2)
Comments and Suggestions for AuthorsThe authors have added details in the literature review section indicating that simplified approaches to computing the reflected contribution of electric light in a space are available, but have not included such a contribution into their approach. The proposed approach is still an approach that addresses lighting control to meet a specific target illuminance value in a completely black room. The inclusion of even an estimate of the reflected light would make this approach more applicable in a real space and would improve the proposed control approach.
This omission raises major concerns regarding the details of the test simulation in the first paragraph of section 4.2 that have not yet been addressed. Is reflected light included in that simulation? The authors state that 350 lux was achieved by their lighting control system in that example, but if that considered only the direct light contribution, the actual illuminance at the workplane locations should be over 400 lux with the addition of reflected light. If the final desktop illuminance level was 350 lux, then the lighting control system cannot be operating as described in this paper (using only the direct contribution). The inconsistency in how this study is described and reported must be addressed. The paper should provide the final illuminance at the task locations that also include the reflected light. The method in Reference 2 considers the reflected light whereas the approach in this paper controls the luminaires to deliver the desired target level by direct light alone. Therefore, to make this fact clear to the readers, the final total illuminance at the target location (direct plus reflected contributions) must be reported here.
The authors indicate they computed the f(theta) axially symmetric luminous intensity distribution from the manufacturer’s report, which is most likely quadrilaterally symmetric data. This simplification introduces errors that will be large for some luminaires and small for others, but no mention of this is made in the paper. The significant limitations that luminaires cannot vary significantly from an axially symmetric distribution, and must have all light emitted below the horizontal plane, need to be added to the conclusion.
Section 2.1. The phrase “luminance polarity” is not understood. The likely correct term to apply here is “luminous intensity distribution”.
Bottom of page 2: “exchanging illuminance information with the target workplane” is not clear. Is illuminance information related to the target workplane being exchanged between control devices (not with the workplane)?
The following statement in the conclusion is not justifiable and should be removed: “While accounting for external influences and reflected light could lead to further energy savings, it was not considered in the interest of user satisfaction.” The level desired by users can be achieved by both the direct and reflected contributions to their task surface. If the user input on their preferred lighting level is addressed with feedback where the user simply raises and lowers the lighting level at their desk through a device such as a smartphone, they are responding to both the direct and reflected contribution. If they dial this in to 350 lux, for example (direct + reflected), then the control system would be able to extract the direct contribution, which, for example, might be 300 lux, and would use that level from that time forward and a final level close to 350 lux would likely be achieved. This description of how the authors’ control system can be applied and not need to consider the reflected contribution is important, but might not be able to accommodate a localized solution when the preference level was set under uniform lighting conditions across the entire space.
As written, ignoring the reflected light results in additional energy being consumed and therefore a lighting control condition that is not optimized. Therefore, the statement in the Conclusion “the proposed approach maximised energy savings and ensured preferred task illuminance” is not entirely true, since lower output levels that meet the preferred task illuminance are possible with the inclusion of the reflected light contribution within the space. However, a system that operates as described in the previous paragraph above would come close to maximizing energy savings.
Comments on the Quality of English Language
N/A
Author Response
Thank you for taking your valuable time to review my paper.
Author Response File: 
Author Response.pdf
Reviewer 3 Report (Previous Reviewer 4)
Comments and Suggestions for AuthorsI accept almost all of the authors' explanations. Unfortunately, one comment was not taken into account. As I wrote in the first review, the illuminance depends not only on the square of the distance but also on the angle of incidence of light. Formulas 1 and 8 are still incorrect because there is no cosine of the incidence angle. The absence of a cosine must lead to incorrect results.
Let's look at Figure 2. There is a Target point where the illuminance E(t) is calculated. The superposition method is used and the shares of individual luminaires are calculated. These shares depend on the distance and the luminous intensity of the luminaire at a given angle. They also depend on the cosine of the incident ligh at target pointt, and this cosine angle will be different for each luminaire fixture shown in this figure.
Please explain and correct the paper.
Author Response
Thank you for taking your valuable time to review my paper.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report (Previous Reviewer 2)
Comments and Suggestions for AuthorsThe authors have added some minor text which indicates the major limitations of their work. These limitations rule out the application of this approach in any real-world lighting control system. The fact that only the direct contribution of light from a luminaire to the work surface is considered, and that luminaire output must be 100% downlight with axial symmetry, must also be added to the abstract, and preferably also to the title (DIRECT Illuminance Contribution-based Lighting Control . . . .).
Suggested corrections/modifications:
1) In Table 1, d cannot be % in the equation. It must be a decimal fraction.
2) In section 4.1, the simulations conducted are similar to [2] in layout, but surface materials are assumed to be black (no reflection is occurring). This major difference should be stated in this paragraph and/or in the caption to Figure 3, which shows reflective surfaces. The following can be added to the caption: Reflections from room surfaces are not considered in the proposed approach.
3) Change “While energy savings” to “While additional energy savings” in conclusion sentence #4. Also, the limitation that this approach can only be applied to 100% downlight equipment should be added, perhaps by applying the following modification to the last sentence: “the luminaires must be ideally axisymmetric with all light emitted below the horizontal plane”.
4) If the only “external influence” possible is daylight, it would be better to replace this phrase with “daylight”. Otherwise, when this phrase is initially applied, the fact that it refers to daylight and possibly other influences should be stated.
Comments on the Quality of English Language-
Author Response
Thank you for your comments.
Author Response File: Author Response.pdf
Reviewer 3 Report (Previous Reviewer 4)
Comments and Suggestions for AuthorsMy previous review included comments regarding the failure to take into account the cosine angle of incidence in the illuminance formula. In the revised version of the manuscript, the cosine appears in formulas (1) and (8). However, the cosine is only in the first part of the formula. In the second one, it is no longer there. This way of writing these formulas means that I don't know whether the authors took the cosine into account in the calculations. The authors did not respond to my comments regarding the cosine. If cosine is not taken into account in the calculations, this is a fundamental mistake.
I cannot accept the manuscript in this form.
I have included this basic formula describing the inverse square of distance law.
Comments for author File: Comments.pdf
Author Response
Thank you for your comments.
Author Response File: Author Response.pdf
Round 3
Reviewer 2 Report (Previous Reviewer 2)
Comments and Suggestions for AuthorsAll suggested changes have been made.
Comments on the Quality of English Language-
Reviewer 3 Report (Previous Reviewer 4)
Comments and Suggestions for AuthorsFinally, in formulas 1 and 8, the cosine of the angle of incidence of light is included. It took a long time for the authors to learn the basic formula. I hope that this improved formula was also included in the algorithm. I accept the manuscript as is.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe manuscript entitled “Illuminance Contribution based Lighting Control for IoT-based Lighting Systems in Smart Buildings” addresses an important and transversal topic which might be worthy of publication in Sustainability. However, there are some aspects that would need improvement before I recommend publication. They are:
1. The concept of IoT is widely used nowadays, but the irruption of other related concepts like AI, makes it necessary to explain the differences to the readers of Sustainability, that might not be experts. Thus, a brief definition of IoT and its field of application is necessary in the revised version.
2. The photometric quantity producing visual perception is Luminance. The determination of luminance from a visual plane requires the knowledge of the reflecting properties of this plane. Since it is not always possible, we work with approximations that have limitations. Two relatively recent papers, one in this journal, approached this topic. The authors must read them and discuss the limitations of their proposal including those regarding the unknown reflectances.
- Cuttle, C. A fresh approach to interior lighting design: The design objective – direct flux procedure. Lighting Res Technol., 50, 1142-1163, 2018.
- - Peña-García, A., Salata, F. Indoor Lighting Customization Based on Effective Reflectance Coefficients: A Methodology to Optimize Visual Performance and Decrease Consumption in Educative Workplaces. Sustainability, 13, 119, 2021.
3. Another important topic in indoor lighting is the reflection on walls, which remarkably impacts the illuminance on the working plane. The authors should explain the potential limitations of this omission.
4. The current trend is a total conception of installations. Thus, the savings in lighting may lead to savings in heating and cooling among other factors of the installation. A key paper on this topic demonstrated that accurate lighting may lead to important savings in conditioning. The authors should read it and comment potential applications of their proposal to these HVAC installations. The reference is:
- Golasi, I., Salata, F., de LietoVollaro, E., Peña-García, A. Influence of lighting colour temperature on indoor thermal perception: A strategy to save energy from the HVAC installations. Energy & Buildings, 185, 112–122, 2019.
If the remarks above are properly solved, I would recommend to accept the paper in a second round.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper presents some interesting work, however I have concerns regarding some of the details in the control approach and whether or not it presents a useful approach that could be applied in real spaces. A number of important details are also missing related to the approach itself, as well as to the tests that were conducted to validate it.
Specific comments on a variety of important issues are provided below.
1. It is unclear if the authors are considering reflected light in their model. I suspect they are not, but the reflected light in a space can be significant, perhaps 15-20% in a room with downlight and more in a room with uplight. Is it being considered in the DIALux simulation – I suspect so? This would likely lead to different performance s that simulation and operation in the real space, but details on the performance in the real space are missing. The authors should provide the task illuminances that result from their test of the control algorithm. All that is currently provided are the dimming levels. The authors’ control approach must address reflected light otherwise their approach, which considers only direct light from luminaire to workplane is what would be obtained in a completely black room. Such operation in a real space would then overlight the tasks. The stated approach also cannot be applied to luminaires with uplight.
2. I am a bit confused with exactly what type of reading is taken at each task location in setting up the proposed control system. There appears to be a device that is placed at these desk locations in commissioning the system, but details are lacking – is it solely for the determination of the LOS distance to each luminaire? Couldn’t it be used to determine the illuminance contribution from each luminaire and then no LOS distance is needed, simply the dimming level. This would permit interreflected light to be addressed in the control system.
3. Figures 4 and 5 do not appear to include contributions from reflected light, but reflected light will add to the contribution from each luminaire. In addition, the distribution for the luminaire shown in the image for the troffer luminaire appears narrower than what is likely emitted by that luminaire. Since the paper does not document the exact product, I found a likely match through an internet image search and the matched luminaire delivers a wider distribution and had a lower wattage (32W) at a similar lumen output. Was the authors’ 54W system for an older product? Details on the equipment being applied should be provided in the paper.
4. The authors mention visual comfort a number of times in the paper, but visual comfort generally applies to aspects of the lighting system’s effect on occupants other than simply that which results from task plane illuminance. Examples are direct glare, reflected glare off tasks, luminance ratios, etc. Turning off lights in areas away from an occupant who is alone in a space could create a very dark surround and contribute to discomfort, which is not addressed in the control approach described, which focuses on saving energy. All references to visual comfort should be eliminated and/or converted to illuminance (or preferred task illuminance).
5. Controlling luminaires solely to maximize energy savings is likely to produce poor quality task lighting. Lighting to a task from a single source is likely to present strong shadows on the work surface, whereas light from multiple directions would be of higher quality. A more appropriate control algorithm should consider this, as well as the direction from which the light is arriving at the task. Luminance ratios, such as from the task to the space surround are an additional quality metric that should be considered by such a control system to deliver a quality lighting condition.
6. The authors’ simplification of the luminaire photometry to being the product of f(theta) and the emitted lumens (which assumes axial symmetry) does not appropriately address most lighting products. Often, the luminous intensity at different horizontal angles is different. Even the Figure 4 troffer luminaire exhibits quadrilateral symmetry, which makes the author’s axially symmetric approach inaccurate. This common feature of architectural lighting equipment needs to be addressed in a model such as the one described in this paper. If applied, then the task location must also be addressed through a horizontal angle to apply to any calculations, such as those in Equation X. Finally, if the luminaire being used is larger, for example, 1.2 m in length, is it considered as a single point source, and what is the error associated with this assumption.
7. In Table X, the symbol in the rightmost column for degrees should be changed to the word “degrees”.
8. Comparison of illuminance contribution at Scenario I
9. Shadowing is mentioned, but not considered in the authors’ approach. An occupant’s body shadow could easily cause the described approach to fail.
10. What the authors refer to as “lightings” should be changed to “luminaires” throughout the document.
11. To avoid confusion, in describing the adjustments that are made to arrive at a set of dimming levels across all of the luminaires, the paper should explicitly state when C(i,j) values are considered versus C(i) values in selecting luminaires with high contributions.
12. For better clarity, Figure 8’s caption should be changed to something like: Comparison of the highest two illuminance contribution rate luminaires at each task loction for Scenario I.
13. The term Total Illuminance Contribution should be renamed. Given its units, it is an efficacy rather than simply a contribution. Similarly, Contribution Rate would be more appropriately named a Contribution Fraction or Contribution Percentage.
The most serious areas of concern in this study are an apparent omission of the reflected light produced by the lighting system; the assumption that all equipment is axially symmetric, when it generally is not; and the promotion of a lighting control approach that is governed solely by task illuminance and makes no attempt to address lighting quality. This control approach would save energy, but, as described by the authors in Scenario 2, would not be one that delivers a high quality lighting solution for a partially occupied space such as the one described. Also, for the actual tests of the lighting control system, reporting of the illuminances delivered to the task locations should be included in the manuscript.
Comments on the Quality of English Language
Suggested terminology changes have been provided.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe present article is a technical proposal to have an independent control of the intensity of luminaires in an indoor room based on the specific presence of persons geolocated using an emitting Wi-Fi signal.
Thus, it is intended to reduce power consumption used for lighting while it is guaranteed the “visual comfort of users”.
The mayor drawback of the paper is that this “visual comfort of users” is only evaluated considering the illuminance value of the simple point where a user is located at a specific moment. This gets an extreme state where if only a person is in one large room it is only illuminated his nearby area being the rest of the room almost dark due to energy efficiency reasons.
In the simplest of words, light uniformity refers to the uniformity of lighting in an environment. It is vital to maintain the uniformity of light in order to make sure that everything is perfectly visible in the room. When referring to light uniformity of an area, the task area in which the objects and immediate surroundings are considered. The importance of maintaining an ample amount of brightness cannot be stressed enough. However, it is common for people to omit the calculation and measurement of light uniformity in a space. When the light uniformity is very low for indoor or outdoor lighting, it would lead to athletes, workers, or citizens feeling uncomfortable as their vision would be affected.
However. Our perception of the world around us is affected by lighting beyond or present location. Hence, lighting uniformity is crucial for navigation and lighting drops hinder us from fully perceiving the space. Most focus-intensive tasks require a uniformity index of around 0.6, whereas, technical drawing and other demanding tasks require a ratio of at least 0.7.
Moreover, “the classic quality features of lighting can be divided into three basic quality features, which are weighted differently depending on the use of the room and the desired appearance: Visualization, visual comfort, and visual ambience. The following applies:
• Visual performance is influenced by the illuminance and the limitation of direct and reflected glare.
• Good color rendering and a harmonious brightness distribution ensure visual comfort.
• Visual ambience is determined by CCT, light direction, and modeling (i.e., the distribution of light and shadows).”
“A pleasant lighting climate is created when people, architecture, and room furniture are illuminated in such a way that shapes and surface structures are clearly visible.”
Stefani, O., & Cajochen, C. (2021). Should we re-think regulations and standards for lighting at workplaces? A practice review on existing lighting recommendations. Frontiers in psychiatry, 12, 671.
It is not clear, but it seems the work presented is based only in Dialux simulations and the human perception of the visual effect that generates this focus type lighting has not been evaluated.
Laws and regulations go in this same direction and minimum uniformity and glare values are required for indoor lighting installations (EN 12464‑1:2021 specifies lighting requirements for humans in indoor workplaces; Occupational Safety and Health Standards for Shipyard Employment 1915; Japanese lighting standards (JIS Z 9110: 2010). …)
From the perspective of this reviewer, this is a mayor drawback of the methodology of the work that is not analyzed and developed correctly, that different simplifications are made and relevant aspects of lighting installations are ignored. Its analysis focuses only on the validation of expected results without performing a critical analysis of them, comparison of result with other methodologies (pros and cons,…). The quality/relevance of the analysis of result and conclusions sections is low.
There is neither a critical analysis on the legal requirements of having persons or workers geolocated considering privacy, the difficulties of having all persons using all the time a signal emitter and the security aspect on evacuation or emergency situations.
There are also other elements not properly analyzed: “In this paper, the effect of daylight is excluded because the illuminance by daylight is the constant measured by the illuminance sensor at t and it can be considered by the default value.” This does not seems correct as daylight can never be considered as a constant as in depends on: the hour, the month/season, the weather,… and all these modifies the amount and quality (glare) of light that may reach to a same single point.
Other minor considerations:
Referenced are listed without a logic order. Ref number 14 appears in 7th place.
Some references are used to justify very basic sentences: “With the invention of low-voltage light emitting diode (LED) lighting sources [1,2]”
You refer several times the concept IOT but Internet does not seem to have any action in the system performance, there is only a wireless communications in the band of 2,4 GHz between electronic units but that is all.
It is not clear how many point need to be configured in a room to have a significant precision. It does not seem logical to have a matrix of 2 cm pith that need to be trained with “the illuminance measured at the sensor at time [t]” Again, ‘t’ should be considered in 1 year round?
Table 1: ‘d’ is not included
Table 2: How many of the 14 luminaires is 54W model and 19W model?
The set up of the luminaries simulated is not justified. It is abnormal to have such aleatory placement of luminaires in any real installation.
In conclusion, due to all the previous concerns detailed above I regret that I must recommend the rejection of the text.
Comments on the Quality of English LanguageThere are several English writting errors. Text would need to be reviewed. i.e.
Page 2 “so light sensors are usedor skylights are redesigned to provide a lighting environment …” ;
Page 5 "it can be considered by the default value"
Page 5 “lighting” is not equal as “luminaire” à “in which N lightings …”;
Page 10 “54 W LUMINAIRE 3D illuminance distribution diagram” …)
...
Reviewer 4 Report
Comments and Suggestions for AuthorsIn my opinion, the authors do not use the correct vocabulary in the field of lighting technology. At this stage of the review I did not have time to correct all the errors, but I will only mention a few examples. The words "lightings" and "LED lightings" are used instead of the word "luminaires". The word "luminaires" should be used. Why write LED lightings, does it matter that it is LED luminaires? This is simply luminaires.
The authors also use the term "brightness of those luminaires" - they should not write "brightness" but should write "luminous flux" of these luminaires. By the way, I haven't found any information anywhere about how the authors treat the dimming of luminaires. They write about dimming rather than writing about power regulation. For example, controlling at 50% power does not mean obtaining 50% of the luminous flux output. Do the authors know about this?
Formulas 1 and 8 should represent the inverse square of the distance law, which, under certain assumptions, can be used to calculate illuminance.
First of all, the authors make a mistake because these formulas lack the cosine of the angle of incidence of light. Illuminance depends not only on the square of the distance but also on the angle of incidence of light.
Secondly, this formula can be used for the so-called point light sources. There are instructions and recommendations on how to use this formula for sources whose sizes are larger than a point, but the authors do not write anything about it.
The assumption of the presented method is to calculate the illuminance using the so-called inverse square law formula. However, this formula can only be used to calculate the so-called direct component of illuminance. The total value of illuminance at a given point in the room has two components: a direct component and an indirect component. The indirect component depends on the multiple reflections of the luminous flux that occur between surfaces in the room (ceiling, walls and floor). The authors do not write anything about this component. If they do not take it into account, it means that they are calculating the illuminance incorrectly.
Lighting design is not only about ensuring the appropriate illuminance value at the task area. Care should also be taken to ensure other important parameters such as uniformity of illuminance, appropriate value of illuminance on the walls and ceiling, cylindrical illuminance in rooms where visual communication is important, etc. How will adjusting the power of individual luminaires affect the values of these parameters? Will the lighting design still meet the requirements after reducing the power of individual luminaires? The authors do not write anything about this.
On page 5, the authors write that "In this paper, the effect of daylight is excluded because the illuminance by daylight is the constant measured by the illuminance sensor at t and it can be considered by the default value." It's an mistake. After all, daytime lighting is characterized by high variability during the working day and cannot be excluded from this analysis.
What does equation 4 mean?
How were charts 4 and 5 made? Is it in accordance with equations 1 and 8? Is the cosine of the angle of incidence of light taken into account?
