Massey stu­dents help in dam­aged sea­wall re­pair

DEMM Engineering & Manufacturing - - TECHNOLOGY -

MASSEY UNIVER­SITY en­gi­neer­ing stu­dents have teamed up with one of their lec­tur­ers to de­velop a real-time sen­sor for a dam­aged sea­wall in Auck­land.

Last Au­gust, Auck­land Trans­port had dis­cov­ered dam­age to a sea­wall un­der the his­toric Auck­land Ferry Build­ing. The ma­jor scour­ing, thought to have been caused by pro­pel­ler wash from fer­ries, needed ur­gent at­ten­tion. Auck­land Trans­port moved to fix the is­sue, but they needed to con­tin­u­ously mon­i­tor the wall for po­ten­tial move­ment while divers un­der­took re­pairs. This was nec­es­sary for the safety of the divers and peo­ple within the ferry build­ing, how­ever they didn’t have the tech­nol­ogy them­selves. While such a so­lu­tion could be or­dered, it was weeks away.

Se­nior lec­turer in com­puter en­gi­neer­ing Dr Fakhrul Alam was ap­proached by Auck­land Trans­port to see if he could come up with a so­lu­tion. “We went down to take a look and saw that their method for mon­i­tor­ing in­volved man­ual mea­sure­ments be­ing taken daily,” Dr Alam says. “As these were taken at dif­fer­ent times they could be in­flu­enced by the tide, time of day, and tem­per­a­ture, ul­ti­mately mak­ing the in­for­ma­tion in­ac­cu­rate and too slow to act on if the wall started to move. They needed real-time, but they didn’t have the tech­nol­ogy to im­ple­ment real- time mon­i­tor­ing on- hand.

“We saw that the sen­sors needed to po­ten­tially ex­ist un­der­wa­ter, which meant wa­ter­proof­ing, and there was no good ac­cess for power. Our ex­per­tise is mainly in ro­bot­ics not in­fra­struc­ture, but I thought that we could give it a go, although we didn’t make any prom­ises,” he says. To help him with the project, Dr Alam en­listed fourth- year Bach­e­lor of En­gi­neer­ing ( hons) stu­dent Baden Parr and PhD stu­dent Daniel Kon­ings to do the work un­der his su­per­vi­sion. “This re­quired de­vel­op­ing a com­plete end-to- end so­lu­tion,” Dr Alam says. “Due to the ur­gent na­ture of the re­pairs, a so­lu­tion was re­quired to be ready for de­ploy­ment within t wo weeks. The time pres­sure meant the team couldn’t source ma­te­ri­als from over­seas and had to use what they had on hand or could at­tain lo­cally. “There were a lot of tech­no­log­i­cal prob­lems we needed so­lu­tions for – a com­bi­na­tion of wire­less telecom­mu­ni­ca­tion, elec­tron­ics, sig­nal pro­cess­ing, data­base and web de­vel­op­ment, and ma­chine learn­ing.”

De­spite the chal­lenges the team was able to pro­duce some sen­sors which they be­lieved would work. They found that a bucket of salt wa­ter sim­u­lated the ocean well enough to test and prove the sen­sors were wa­ter­proof be­fore putting them in a deep freeze to sim­u­late the mas­sive tem­per­a­ture ranges the sen­sors would be ex­posed to.

“In the be­gin­ning we had an is­sue with huge vari­ances in tem­per­a­ture, but with a lot of ma­chine learn­ing and ad­vanced statis­tics we were able to take out that vari­ance, and from then on, they worked per­fectly.”

The sen­sors were de­ployed be­fore the dead­line in late Novem­ber and in­stalled on the sea­wall, just above the tide­line where they were ex­posed to the el­e­ments. A router hang­ing off a barge was able to pro­vide them with Wi- Fi and 4G con­nec­tion to send data back to the Massey server. They stayed in place un­til late April this year.

Parr said the ex­pe­ri­ence was a great op­por­tu­nity to put the­ory to prac­tice. “Dur­ing the course of my un­der­grad­u­ate de­gree, I’d com­pleted a num­ber of prac­ti­cal projects but never with this kind of real- world scope. I gained valu­able ex­pe­ri­ence deal­ing with the re­quire­ments of a project this im­por­tant.” His work­mate for the project, Kon­ings, was sim­i­larly happy with the work. “I en­joyed work­ing on ad­vanced ma­chine learn­ing prin­ci­ples dur­ing this project and see­ing a pro­to­type come to­gether so quickly. This project was very dif­fer­ent to ex­ist­ing com­mer­cial projects I have been in­volved in, and it was very sat­is­fy­ing to be able to meet the strict dead­lines re­quired, whilst still de­liv­er­ing a qual­ity out­come for the client.”

“This re­ally was an amaz­ing feat of en­gi­neer­ing,” says Dr Alam. “Ev­ery­thing was de­vel­oped from scratch. Not only were the sen­sors ex­tremely ac­cu­rate, they are also rugged and ro­bust enough to with­stand the harsh con­di­tions. Their abil­ity to send data re­motely in real time was of ex­treme im­por­tance to the divers work­ing the wall – they had to be able to leave the wa­ter within t wo min­utes of a warn­ing go­ing out.

“Need­less to say, AT is suit­ably im­pressed, but so was I. This project came late in se­mes­ter t wo while the stu­dents were un­der the crunch with ex­ams. The way they man­aged their time and stepped up to the prob­lems was noth­ing short of pro­fes­sional. The project was a unique blend of the­ory and ap­pli­ca­tion for the stu­dents, and our course sets them up to re­ally thrive in both. They had to use such a broad skill set to tackle each of the is­sues and they did that fan­tas­ti­cally, from ma­chine learn­ing to man­ag­ing big data. But also deal­ing with real clients.”


Newspapers in English

Newspapers from New Zealand

© PressReader. All rights reserved.