Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Simulering av stomljud med hjälp av gångmönsterstatistik

Claesson, Jimmy (2008) In TVBA VTA820
Civil Engineering (M.Sc.Eng.)
Engineering Acoustics
Abstract (Swedish)
Hur människor går är väldigt individuellt och ens personliga gångstil ändras genom hela livet. Men det skulle kunna vara så att personer av samma kön med samma längd, vikt och ålder skulle ha ett gångmönster med stora likheter vad gäller fötternas vinkel och stegens längd. I detta projekt undersöktes det geometriska gångmönstret experimentellt på 150 män mellan 170 och 190 cm. Var och en av försökspersonerna fick gå med färg på fötterna en tre meter lång sträcka på papper. Ur det gångmönster
som skapades på pappret extraherades ett antal variablers värden och undersöks statistiskt. Dessa data delades upp i 5 grupper för att undersöka kopplingen till kroppslängd. De 5 grupperna var 170-175, 176-180, 181-185, 186-190 och hela provgruppen.
... (More)
Hur människor går är väldigt individuellt och ens personliga gångstil ändras genom hela livet. Men det skulle kunna vara så att personer av samma kön med samma längd, vikt och ålder skulle ha ett gångmönster med stora likheter vad gäller fötternas vinkel och stegens längd. I detta projekt undersöktes det geometriska gångmönstret experimentellt på 150 män mellan 170 och 190 cm. Var och en av försökspersonerna fick gå med färg på fötterna en tre meter lång sträcka på papper. Ur det gångmönster
som skapades på pappret extraherades ett antal variablers värden och undersöks statistiskt. Dessa data delades upp i 5 grupper för att undersöka kopplingen till kroppslängd. De 5 grupperna var 170-175, 176-180, 181-185, 186-190 och hela provgruppen.

Resultatet på undersökningen var att variablerna som beskriver gångmönstret kan delas upp i två separata grupper, de som har en koppling till kroppslängden och de som inte har det. De variabler som har en koppling till längden är de som fungerar i gångens riktning dvs. steglängden. Dock är dessa variabler antagligen inte kopplade till hela kroppslängden utan bara till benlängden. Variablerna som inte är kopplade till kroppslängden har att göra med bredden på gångmönstret och denna är antagligen kopplad till bredden på bäckenet.

Ur all denna statistiska data byggdes ett antal Markovkedjor upp som beskrev förändringen av variablernas värde från steg till steg. Dessa skulle användas för att undersöka ifall det var möjligt att, istället för att använda den konventionella metoden för stegljudisolering, utnyttja en simulering med finita element och göra det i en dator. FEM-simuleringen som utfördes visade på att det är möjligt att simulera gång
på golv och beräkna vibrationerna så att det är möjligt att utvärdera olika konstruktioner och undvika de sämre. (Less)
Abstract
How people walk is very individual and even their personal style of walking may change throughout their lives. But it could be that the people of the same sex with the same height, weight and age would have a way of walking with great similarities in terms of feet angle and length of the stride. This project examined the geometric walking pattern experimentally of 150 men between 170 and 190 cm. Each of the subjects had to walk with paint on their feet, a three-metre long stretch of paper. From the walking pattern that was created on paper the values of a number of variables was extracted and examined statistically. These data were divided into 5 groups to investigate the link with height. The 5 groups were 170-175, 176-180, 181-185,... (More)
How people walk is very individual and even their personal style of walking may change throughout their lives. But it could be that the people of the same sex with the same height, weight and age would have a way of walking with great similarities in terms of feet angle and length of the stride. This project examined the geometric walking pattern experimentally of 150 men between 170 and 190 cm. Each of the subjects had to walk with paint on their feet, a three-metre long stretch of paper. From the walking pattern that was created on paper the values of a number of variables was extracted and examined statistically. These data were divided into 5 groups to investigate the link with height. The 5 groups were 170-175, 176-180, 181-185, 186-190 and the entire sample.

The results of the survey were that the variables that describe the walking pattern can be divided into two separate groups, those who have a connection with the height and those who do not. The characteristics that have a connection to the height are the ones that work in the direction that the person is walking. However, these variables are probably not linked to the whole body, but only length of the legs of the person. The variables that are not connected to the height has to do with the width of walking pattern and this is probably linked to the width of the pelvis. From all this statistical data a number of Marko chains was constructed that described the change in the variable value from stride to stride. These would be used to examine if it was possible that, instead of using the conventional method of calculating the amount of sound that is transmitted because of footsteps, take advantage of a simulation with the finite element method and investigate the transmission in a computer. The FEM-simulation that was carried out showed that it is possible to simulate walking on the floor in a computer and calculate vibrations so that it is possible to evaluate the different
structures and to avoid the worse ones. (Less)
Please use this url to cite or link to this publication:
author
Claesson, Jimmy
supervisor
organization
course
VTA820
year
type
H3 - Professional qualifications (4 Years - )
subject
keywords
Markov chains, Stride, Walking pattern, FEM, Vibrations
publication/series
TVBA
report number
TVBA-5038
ISSN
0281-8477
language
English
id
3806147
date added to LUP
2013-08-02 12:11:33
date last changed
2013-10-07 13:50:24
@misc{3806147,
  abstract     = {{How people walk is very individual and even their personal style of walking may change throughout their lives. But it could be that the people of the same sex with the same height, weight and age would have a way of walking with great similarities in terms of feet angle and length of the stride. This project examined the geometric walking pattern experimentally of 150 men between 170 and 190 cm. Each of the subjects had to walk with paint on their feet, a three-metre long stretch of paper. From the walking pattern that was created on paper the values of a number of variables was extracted and examined statistically. These data were divided into 5 groups to investigate the link with height. The 5 groups were 170-175, 176-180, 181-185, 186-190 and the entire sample. 

The results of the survey were that the variables that describe the walking pattern can be divided into two separate groups, those who have a connection with the height and those who do not. The characteristics that have a connection to the height are the ones that work in the direction that the person is walking. However, these variables are probably not linked to the whole body, but only length of the legs of the person. The variables that are not connected to the height has to do with the width of walking pattern and this is probably linked to the width of the pelvis. From all this statistical data a number of Marko chains was constructed that described the change in the variable value from stride to stride. These would be used to examine if it was possible that, instead of using the conventional method of calculating the amount of sound that is transmitted because of footsteps, take advantage of a simulation with the finite element method and investigate the transmission in a computer. The FEM-simulation that was carried out showed that it is possible to simulate walking on the floor in a computer and calculate vibrations so that it is possible to evaluate the different 
structures and to avoid the worse ones.}},
  author       = {{Claesson, Jimmy}},
  issn         = {{0281-8477}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{TVBA}},
  title        = {{Simulering av stomljud med hjälp av gångmönsterstatistik}},
  year         = {{2008}},
}