Use of Genetic Algorithm for Spatial Layout of Indoor Light Sources

Pedro Henrique Gouvea Coelho, J. F. M. do Amaral and K. P. Guimarães

State Univ. of Rio de Janeiro, FEN/DETEL, R. S. Francisco Xavier, 524/Sala 5001E, Maracanã, RJ, 20550-900, Brazil

Keywords: Genetic Algorithms, Artificial Intelligence Applications, Spatial Layout of Light Sources.

Abstract: People spend many hours inside buildings that are naturally and artificially illuminated. Since mankind has

been able to tame the fire and use it to illuminate, the natural condition of nighttime darkness has been

modified. With the advent of electric lighting this has been intensified. The problem of indoor lighting

presents several options according to the specific purpose of the lighting. There is room for some heuristic

choices and genetic algorithms have been chosen as a computational intelligence technique that allows

multi-objective solutions and the inclusion of heuristics and versatility in specific situations that occur in

many particular applications. In this way, the main objective of this article is to optimize the number of light

sources in indoor environments with the aid of genetic algorithms to obtain a suitable light intensity with the

smallest number of light sources. One of the paramount reasons for using such algorithm is that it returns an

acceptable solution to an optimization problem with infinite possibilities in a finite number of trials. A case

study is presented in which the applicability of genetic algorithms to the problem is discussed, and the

results indicate the viability of the method.

1 INTRODUCTION

The lighting of an environment is something that

must be designed considering several factors, such

as the age group of the people who will live in that

place, the existence of natural light, type of

construction e.g. school, office, shopping center, or

sport arenas, to adjust the lighting to the performed

activity. Optimum luminous intensity is one that

allows the vision of all the points of the environment

that does not cause discomfort in the eyes of the

human being, and does not warm the environment.

The economic factor is also important, so lamps are

being manufactured to have lower energy

consumption and longer life. This saves money on

light bills and their replacement. Lighting is

responsible for about third of the energy spent in

commercial and office buildings that accounts for

more than 40% of the electrical load (Simpson,

2003). As a result, in such buildings, significant

energy savings can be achieved by reducing lights

during daylight hours or turning lights off in

unoccupied rooms. Other aspect is related to the

lighting efficiency by deploying an adequate number

of light sources suitably located to provide an

appropriate light intensity to a chosen environment,

in other words yielding a good installation with

efficient luminaires.

The main objective of this article is to optimize the

number of light sources in indoor environments with

the aid of genetic algorithms to obtain a suitable

light intensity with the smallest number of light

sources.

The remainder of this paper is organized as follows.

The second section deals with the basics of lighting

design to establish the framework of the research

done. The following section discusses the genetic

algorithm approach. A case study is presented in

section four and the paper ends with conclusion

remarks in section five.

2 LIGHTING BASICS

Light is a sort of radiant energy, detected by the

visual sense of clarity of the human eye, that

propagates through electromagnetic waves. The

radiation range of the electromagnetic waves

detected by the human eye is between 380 nm and

780 nm. The visible electromagnetic spectrum is

limited, at one extreme by the infrared (longer

wavelength) radiations, and the other by the

ultraviolet (lower wavelength) radiations.

456

Gouvea Coelho, P., Amaral, J. and Guimarães, K.

Use of Genetic Algorithm for Spatial Layout of Indoor Light Sources.

DOI: 10.5220/0006777304560460

In Proceedings of the 20th International Conference on Enterprise Information Systems (ICEIS 2018), pages 456-460

ISBN: 978-989-758-298-1

The study of the implementation and use of artificial

lighting, whether indoors or outdoors is called

lighting technology. The lighting technique is quite

old and it comes to be prior to the use of electricity

for the generation of illumination, since the first

sources of illumination were the very fire of lamps.

It is important to use some calculation method for

the design of lighting projects to define the quantity

of luminaires and equipment necessary for a given

environment to have adequate illuminance. The most

common calculation method is the lumens method,

defined by the International Commission on

Illumination (CIE) and the point-to-point method,

which is based on Lambert's Law, which defines that

the illuminance is inversely proportional to the

square of the distance of the point illuminated to the

luminous focus. The point-to-point method is also

called the light intensities method. According to this

method, it is possible to determine the illuminance at

any point of the surfaces by means of trigonometric

calculations considering the light sources present in

the environment. The total illuminance by the point-

to-point method is the sum of the illuminances

coming from each light source, whose beam reaches

the point considered (Karlen and Benya, 2004). The

lumens method is usually the most used, considering

the total amount of light flux required for a given

environment, based on the type of activity

developed, the reflectance of the surfaces (ceiling,

walls and floor) and types of equipment (luminaires,

equipment). This method considers rectangular

environments, diffuse reflection surfaces, a single

type of luminaire and considers uniform distribution

of light intensity (Taylor, 2000). In this work, the

ideas of the two methods will be used, because

analyzes will be performed in terms of rectangular

environments with uniform distribution of light

intensity, using only one type of light source. The

maximum distance that can be obtained from a light

source is sought, in order to reach the value of

illuminance required. Such value of illuminance is

usually established by a standards agency such as

the NBS in the USA. Based on these methods we

can calculate the minimum number of sources

needed to obtain the ideal illuminance.

Now it will be presented briefly some basic concepts

needed in the discussion of the problem.

Luminous Flux is the total radiation power emitted

by a light source. It can also be interpreted as the

amount of radiant energy capable of sensitizing the

human eye for 1 second. It is the SI derived unit of

luminous flux. Luminous intensity is the power of

light radiation in a given direction. The luminous

intensity is the base quantity of the international

system for illumination, and the unit is the candela

(cd). Light level or Illuminance, is the amount of

light measured in a plane surface (or the total

luminous flux incident on a surface, per unit area).

Illuminance is measured in lux in the metric SI

system.

The release of innovative products in the market is

usual, due to the great development in the last

decades. The lamps have undergone great changes

brought by energy saving needs and to technological

advances. One can easily find several types of lamps

which will now be presented in order to be

considered in the discussion of the problem in this

paper. First, the common incandescent lamps will be

discussed. The incandescent lighting results from the

passage of electric current through a spiral wire and

high electrical resistance. The higher the wire

temperature, the greater the amount of light emitted.

As one turns on and turns off the traditional

incandescent bulb, the metal wire inside the glass

bulb will wear out, consume with the heat until it

breaks and no more electric current passes, and the

lamp stops producing light. Among the various types

of light bulbs on the market, the common

incandescent is the most commonly used, especially

in homes, whether decorative or reflective, perhaps

because it is the oldest and the cheapest.

Halogen lamps have the same working principle as

ordinary incandescent lamps. However, they present

halogen gases that, within the bulb, combine with

the tungsten particles detached from the filament.

This combination, added to the thermal currents of

the lamp, causes the particles to deposit back into

the filament, constituting the regenerative cycle of

the halogen. In this way, the halogen incandescent

lamp has a longer life, greater luminous efficiency

and, as it is able to avoid the darkening of the lamp,

it has a whiter and uniform light. They are widely

used by designers and decorators, and are applied in

facades, leisure areas, theaters and even car

headlights.

Fluorescent lamps are known as cold light because

they emit less heat to the environment than

incandescent ones. They consist of a cylinder-shaped

glass tube, filled with argon, and its inner surface is

covered with a layer of fluorescent powder i.e.

phosphorus. They were designed to replace

incandescent bulbs and, when compared to

incandescent bulbs, have longer lifetimes, up to five

times the throughput, and generate up to 80% energy

savings. The energy savings that the use of this lamp

generates represents a significant reduction of the

exploitation of the natural resources, since, with

Use of Genetic Algorithm for Spatial Layout of Indoor Light Sources

457

smaller consumption, the smaller will be the need of

new plants to produce it.

LED bulbs have become increasingly popular due to

their advantage in durability and energy

consumption compared to other bulbs on the market.

Its energy consumption is up to 87% less than the

incandescent, making this product a quite attractive

in terms of economy and environmental

preservation. The LED element is a very low-light

LED, however, it has a good brightness ratio. A light

bulb has several such diodes. Today, LEDs are

present in our day-to-day backlight for LCD TVs, in

vehicle lighting and at traffic lights, for example.

LED lamps are efficient because they produce the

same amount of lumens with less energy

expenditure. For example, to generate the equivalent

of 1300 lumens, a 20-Watt LED bulb is enough,

while the same light can only be generated by a 70-

Watt incandescent bulb. Another great advantage of

LED is that its heat emission is practically non-

existent, which helps in energy saving, and its

durability can be up to 25 times greater than that of a

regular lamp. The LED bulb is more expensive, but

the high price is rewarded by the low energy

consumption and durability of the bulb.

According to their type and power, bulbs have

different luminous fluxes:

• 60 W incandescent: 778 lm;

• 70 W halogen: 1334 lm;

• 15 W fluorescent: 1357 lm;

• 6.5 W LED: 600 lm.

3 GENETIC ALGORITHM

A process to be optimized is characterized by an

objective function that provides the behavior of the

process, the constraints that define the search space

and on which the project variables tend to assume

the best value after the optimization. Many of these

processes can be modeled as problems of

maximizing or minimizing a function whose

variables must obey certain constraints. Optimizing

a process is advantageous since it allows working

with a vast contingent of variables and constraints

that are often difficult to visualize or tabulate, thus

reducing the time spent with the process and

obtaining new solutions with lower expenses.

However, the optimization can be hampered by

some factors such as: discontinuous functions, slow

convergence, functions with many local minima, and

the global minimum difficult to find, causing

computational time to become high. Over the years,

the modeling processes became more complex and

with the sophistication of the computational

resources a great advance in the techniques of

optimization was provided. These techniques can be

used in several areas such as: electrical circuit

design, energy distribution, mechanical designs, and

can also be used in biology, economics and other

scientific areas. There are many methods of

optimization and each of them achieves better

performance in certain types of problems. The

choice of method depends on a series of

characteristics of the problem to be optimized,

mainly the behavior of the function that represents it,

and this can be difficult to determine. The genetic

algorithm (GA) is inspired by biological evolution,

because it makes use of a selection of individuals,

uses genetic operators and operates in a random and

oriented way, seeking an optimal solution within a

population. In the case of the search method, a

comparison is made between the evolution of the

species and the problem in question, nature is the

problem, individuals are the possible results, fitness

is the quality of their results, in relation to the

transfer of fitness, the crossover is modeled by an

operator called crossover, and adaptive

modifications are modeled by mutation operators.

Statistically, over several generations, the results

tend to converge to the fittest results.

4 CASE STUDY

In order to realize the spatial distribution of the light

sources of an environment it is necessary to know

the size of the place, adequate lighting for the given

environment, the type of lamp to be used and the

number of lamps required for this illuminance.

According to Standard NBR 5413/1992 of ABNT

(Brazilian Association of Technical Standards), the

illuminance of normal working environments must

be at least 500 lux and at most 1000 lux (the typical

illuminance value is 750 lux). In this standard the

ABNT imposes what should be the illuminance in

several types of environment, as it is observed in

Table I.

For this case study, an illuminance of 500 to 1000

lux will be considered in an environment of 3m

high, 30m long and 15m wide. The illuminance,

which can also be called the illumination level,

depends on the distance from the light source to the

illuminated object. For example, if in a dark

environment we illuminate a nearby object with a

flashlight, we can see a circle of illumination in the

object, smaller and stronger than if we illuminate an

object farther away. What occurs with illuminance is

ICEIS 2018 - 20th International Conference on Enterprise Information Systems

458

known as the inverse law of squares, which relates

the luminous intensity and the distance from the

source. The illuminance E can be expressed as the

ratio of luminous intensity I and the area A seen by

an observer within an angle a of observation as in

equation 1 (Ma et. al. , 2017) (Simpson,2003).

E = I / A cos a

(1)

Simulations were carried out using SCILAB

software using a fitting function in terms of the

source distance to the observation point and

illuminances.

Table 1: Illuminance according to standard NBR 5413/

1992 by Brazilian Association of Technical Standards

(ABNT).

Areas

ILLUMINANCE

(lux)

ENVIRONMENT /

ACTIVITY

CLASS A

(Continuous

use)

20-50

Public streets

50-100

Little used

100-200

Storage units

CLASS B

(General

purpose)

200-500

Auditorium room

500-1000

Offices and factories

1000-2000

Special tasks

CLASS C

(Accurate

visual)

2000-5000

Continue work

5000-10000

High precision tasks

10000-20000

Surgical theaters

In this case study, the maximum distance from the

light source is to be calculated, maintaining the level

of illuminance required.

The main parameters used in the case study are

shown in Table 2.

Table 2: Case study parameters.

Case Study

Parameters

Choice

Number of generations

200

Precision

-6

Chromosome

60 bits - Binary

Population

150

Fitness

Illuminance

Crossover

Binary

Crossover rate

80%

Mutation rate

Figure 1 shows the curves obtained by the genetic

algorithm approach indicating the best result and the

medium one in terms of fitness degree evaluated by

the fitness function in the algorithm. The final result

regarding the number of required lamps is

summarized in Table 3 for several types of lamps.

Figure 1: Best result in red and average in blue, in terms of

fitness degree x number of generations.

Table 3: Case study results.

Case study result

Number of Lamps

Incandescent lamp(410 cd)

Halogen lamp (300 cd)

LED lamp (700 cd)

Some comments concerning the case study can be

expressed. Super-elitism did not occur due to the

high rate of crossover and mutation adopted. The

improvement could not be higher probably because

the genetic variability was reduced very quickly.

The experiment was done using the standardized

roulette to prevent the function from finding a local

minimum and not the overall optimum of the

function.

5 CONCLUSIONS

Genetic algorithm techniques were used in this paper

in order to optimize the number of light sources in

indoor environments. One of the paramount reasons

for using such algorithm is that it returns an

acceptable solution to an optimization problem with

infinite possibilities in a finite number of trials. A

case study was presented in which the applicability

of genetic algorithms to the problem was discussed,

and the viability of the method was indicated. The

work of (Zou and Li, 2010) dealt with the genetic

algorithm application for road lighting optimization.

Corcioni and Fontana, (Corcioni and Fontana, 2003)

also used genetic algorithms for the optimal design

Use of Genetic Algorithm for Spatial Layout of Indoor Light Sources

459

of outdoor lighting systems. The lighting of an

outdoor tennis court and of a football field were

considered as case-studies. The fitness function used

in all these works are different from the one used in

this paper.

Intuitively, the obtained results are consistent with

the ones yielded by traditional lighting designs.

However there is no algorithm or procedure that

should be followed directly in order to check the

number of lamps needed to get some given

illuminance for a given scenario. At most, there are

rule of thumbs or trial and error procedures followed

by experienced practitioners. For instance,

Pachamanov and Prachamanova considered

(Pachamanov and Pachamanova, 2008) optimization

of the light distribution for luminaries. The problem

was formulated as a linear optimization problem that

incorporated the geometrical parameters of the

lighting installation and the reflective properties of

the road surface. Their idea wss to incorporate

changes in the lamps themselves. For future work

one could seek benchmark problems to compare

traditional lighting techniques solutions with the

solutions yielded by the genetic algorithm approach.

REFERENCES

Taylor, A. E. F., 2000. Illumination Fundamentals.

Rensselaer Polytechnic Institute.

Simpson, R. S. A., 2003. Lighting Control Technology

and Applications. Focal Press, Oxford.

Ma X., Bader, S., and Oelmann B., 2017. Characterization

of Indoor Light Conditions by Light Source

Classification. IEEE Sensors Journal, Vol. 17, No. 12.

Zou, J., and, Li L., 2010. Optimization of Luminous

Intensity Distribution of Roadway Lighting Luminaire

Based on Genetic Algorithm. In Second WRI Global

Congress on Intelligent Systems, Wuham, China.

Corcioni, M. , and Fontana L., 2003. Optimal Design of

Outdoor Lighting Systems by Genetic Algorithms.

Lighting Research & Technology, Vol 35, Issue 3, pp.

261 - 277.

Karlen, M., and Benya, J., 2004. Lighting Design Basics.

Wiley & Sons, New Jersey, USA.

Pachamanov, A, and Pachamanova, D., 2008.

Optimization of the Light Distribution of Luminaries

for Tunnel and Street Lighting. Engineering

Optimization, 40:1, pp. 47-65.

ICEIS 2018 - 20th International Conference on Enterprise Information Systems

460