Repository logo
 

Intelligent Passive Room Acoustic Technology for Acoustic Comfort in New Zealand Classrooms

Files

Thesis(5.76 MB)

Date

2022

Authors

Burfoot, Megan Jane

Supervisor

Ghaffarianhosein, Ali
Ghaffarianhoseini, Amirhosein
Naismith, Nicola

Item type

Thesis

Degree name

Doctor of Philosophy

Journal Title

Journal ISSN

Volume Title

Publisher

Auckland University of Technology

Abstract

Background: Classrooms are dynamic, progressive spaces, uniquely shaped by the very students and educators occupying the space. Just how we can change the lighting, ventilation, temperature, and layout of a classroom, we must too be able to change the acoustic ‘state’. To change the acoustic state, we need to alter the total sound absorption in the space, thereby changing the Reverberation Time (RT). RT is the time taken in seconds for the sound level to decay by 60dB, similar to the amount of echo in a space. For example, during group discussions or loud activities, we want more absorption (lower RT). During teacher lectures, we want to project and echo the sound, so we want less absorption (higher RT). Our options for changing RT are to alter the sound-absorbing properties of existing surfaces. Technology with this quality is referred to as Passive Variable Acoustic Technology (PVAT), whose properties you can vary to change the RT in the space. Conveniently, intelligent advancements can be made to PVAT to achieve a desired RT more precisely. Originality: The solution presented in this work will finally recognise classrooms as dynamic, changing spaces which require dynamic, changing RTs. The solution contains two critical components: the PVAT, and the intelligent capability. The PVAT component is achieved by overlaying reflective, rotating louvers over a sound absorbing panel. If the louvers rotate open, they allow sound through to be absorbed by the panel behind, decreasing the RT. If they rotate closed, they block and prevent sound from being absorbed, thus echoing the sound, and increasing the rooms RT. The second component aims to give the technology an intelligent mind of its own. Microphones set-up around the room detect the sound waves in the space. These waves are then transformed into an output that can be interpreted by a machine learning classifier. The classifier is trained with algorithms to recognise and define from the transformed sound waves, which acoustic ‘scene’ is happening in the room. Once the scene has been determined, a pre-programmed algorithm calculates the perfect RT for that acoustic scene. There are diverse negative effects that commonly result from inappropriate RT, which could therefore be alleviated with IPRAT. IPRAT could improve occupant health and safety, increase communication clarity, and improve occupant wellbeing. Aim: With the novel IPRAT, classrooms can finally achieve optimised RTs, for any activity happening in the space. The concept of optimising RT in real-time is new, so it was not known which RTs would optimise each classroom activity. The research question for this work thus followed; what is the effect of IPRAT on classroom acoustic comfort? The aim of this work was to quantify the effects of IPRAT on classroom acoustic comfort. This aim is satisfied by comparing the acoustic comfort of several classrooms using IPRAT, with the same classrooms not using IPRAT. An efficient and economical way this comparison was realised was to firstly use software to simulate the behaviour and results of 5 classrooms using IPRAT, and secondly, conduct a case study using an IPRAT prototype in a real classroom. Finally, the two sets of data were statistically analysed to determine the final effect of IPRAT on classroom acoustic comfort. Additionally, this data was compared with professional industry standards (for New Zealand namely, AS/NZS Recommended Design Acoustic Standards). Methodology: This thesis is submitted as Format Two: Submission by four manuscript publications. The first manuscript aimed to determine the rationale for IPRAT. A best-evidence synthesis and prior art search were conducted to determine the highest level of intelligence for passive variable acoustic technology. It was discovered that dynamic spaces should be designed with varying RT’s however, a literature gap exists for intelligently adjusting RT to suit changing space uses. The unique IPRAT solution was conceptualised, which integrates PVAT and Acoustic Scene Classification. Thus, IPRAT was proclaimed, developed, and analysed, and a use case example for IPRAT was provided. The findings from manuscript one strongly suggested the need to test or prototype IPRAT. The second manuscript aimed to establish a simulation method for testing IPRAT. Using secondary data, 20 classroom environments ‘typical’ to New Zealand were detailed and developed. Additionally, a software method was established which could be used to simulate acoustic technology. The 20 classroom profiles were detailed and demonstrated using I-Simpa, a pre/post-processor for acoustic codes, and Autodesk software. With these virtual environments, it was suggested that IPRAT should now be simulated, to demonstrate its potential to improve acoustic comfort. The third manuscript aimed to determine the effect of IPRAT on acoustic comfort using simulation. IPRAT was thus simulated in the 20 environments established in manuscript two, statistically analysing the effect of IPRAT on RT, sound strength and clarity. The output of this manuscript firstly included an acoustic simulation method in I-Simpa software presented for initial technology validation. Secondly, the quantified improvements of IPRAT on acoustic parameters RT, sound strength and clarity were determined. Last, a database of RTs which improve acoustic quality for four aural situations typical to classrooms was derived. In this simulation, the benefits of IPRAT were found to be statistically significant, and it was recommended that future research physically prototype the technology. The fourth and final manuscript aimed to determine the effect of IPRAT on acoustic comfort using a case study. An IPRAT prototype was deployed in a tertiary classroom by constructing and testing only the PVAT component. The IPRAT was tested by adjusting the prototype's sound absorption. Results: Despite the simulation study achieving a more significant RT reduction and RT range, when we compare it with the ASNZS recommendations, the case study data was much more significant. At a room volume of 170m3, NZS Acoustic Standards recommends a mid-frequency RT of 0.55 for 'Rooms for Speech', and 0.7 for ‘Rooms for 'Speech/Lecture'. Using the equation relating IPRAT coverage and RT from manuscript 4, the researchers could propose that at 20.5% coverage, the RT can be varied between 0.58 and 0.70s. This comes a mere 0.03 and 0.00s away from matching the industry standards for both room types. Thus, it is concluded that by using IPRAT in the case study classroom, the conditions of both room types can be satisfied – increasing the acoustic comfort in both classroom learning and classroom lecture. Existing studies in literature test single RT values, and usually aim to improve the singular RT for the classroom with some form of acoustic treatment. The thesis results can be generalised to New Zealand but may also offer benefits on a global scale. Additionally, by optimising classroom acoustics the most benefits are realized by vulnerable students. This includes children with sensory disabilities, hearing difficulties and those speaking a second language. It is probable that the adoption of this technology in industry will involve a slower progression of intelligence toward IPRAT, beginning with manual and then automated control. This is also a wise way to save development costs whilst slowly introducing the technology into opportune spaces. Trade-offs will need to be made in any spaces using IPRAT as a significant proportion of free wall space will be taken up by the technology. Future studies should compare different methods of achieving PVAT and find the most effective design. After which, the quantitative and qualitative improvements to acoustics should be researched from a human comfort perspective. Findings: The key takeaways from this thesis for industry professionals, academics and policy makers are as follows: First, acoustic comfort is largely neglected within IEQ. Acoustic optimisation is perceived as a complicated design aspect and thus is often avoided as a core topic in architecture curriculum. Thus, acoustic comfort should be taught in all architectural courses as having equal importance to other IEQ's. The discomfort associated with poor acoustics for varying space uses should be understood by students. Second, the acoustic design of spaces is often neglected by designers, as it is set as a low priority. Clients wont intuitively budget for acoustic design. Acoustic optimisation is often an afterthought. When designed for, the acoustics of a space is considered, a trade-off is made for varying space uses. Thus, the acoustics for varying space uses should be optimised. Acoustic engineers should be employed on project teams to advise the most appropriate technology to achieve the varying acoustics. Third, the development of variable acoustic technology is slow, and the current technology in development is unaffordable. No technology is in development which could provide intelligent optimisation. Thus, engineers should continue to test and develop variable acoustic technology, with the goal being to create affordable variable acoustic options. Last, acoustic standards recommended singular acoustic states for flexible and dynamic spaces, including classrooms. Therefore, acoustic standards should reflect the changing acoustic needs for flexible and dynamic spaces, beginning with recommended classrooms acoustics. This responsibility lies with policy makers.

Description

Keywords

Source

DOI

Publisher's version

Rights statement

Permanent link

https://hdl.handle.net/10292/15396

Collections

Doctoral Theses