This paper proposes a method of determining statistical enhancement factors to apply to single pedestrian responses to obtain corresponding crowd-induced vibration responses.

(42 hits)

The full reference for the paper is:

Caprani, C.C., Keogh, J., Archbold, P. and Fanning, P. (2012), ‘Enhancement factors for the vertical response of footbridges subjected to stochastic crowd loading’, Computers & Structures, in press.

And it is available from: http://dx.doi.org/10.1016/j.compstruc.2012.03.006.

Abstract

The vertical acceleration response of a hypothetical footbridge is predicted for a sample of single pedestrians and a crowd of pedestrians using a probabilistic approach. This approach uses statistical distributions to account for the fact that pedestrian parameters are not identical for all pedestrians. Enhancement factors are proposed for predicting the response due to a crowd based on the predicted accelerations of a single pedestrian. The significant contribution of this work is the generation of response curves identifying enhancement factors for a range of crowd densities and synchronization levels.

The construction of the bridge is extremely good. Every connection and member is in perfect position, and this means that load paths are equally distributed. As we in DIT have learned from our Spaghetti Bridge tests, it is often the best constructed bridge, and not necessarily the best design that wins.

(390 hits)

See the news report here (390 hits).

The model is based on that given in Advanced Analysis and Design of Steel Frames by Li and Li, Wiley 2007.

In Visual C++, the MFC Feature Pack gives the property grid control, but one of the questions regularly asked (see stackoverflow.com) is how to return properties once changes by the user. Part of the problem is the COLEVariant is the return type from the framework, but what if you want a double, or CString for example?

I bashed together a few different ideas and have a nice little functoid that essentially overloads the output type and so this can be used as an argument to your function. The references I used were

• http://www.codeproject.com/KB/cpp/returnoverload.aspx (197 hits)
• http://stackoverflow.com/questions/226144/puzzle-overload-a-c-function-according-to-the-return-value (134 hits)

I defined the functoid in the header file for the CPropertiesWnd class (which contains the CMFCPropertyGridProperty control) as:

1
2
3
4
5
6
7
8
9
10
11
12

class getPropValue
{
public:
    getPropValue(CMFCPropertyGridProperty* pProp):m_pProp(pProp) {};
    operator CString()        {return (LPCTSTR)(_bstr_t)m_pProp->GetValue();}
    operator int()            {return m_pProp->GetValue().iVal;}
    operator unsigned int()   {return (unsigned int)m_pProp->GetValue().iVal;}
    operator double()         {return m_pProp->GetValue().dblVal;}
    operator bool()           {return m_pProp->GetValue().boolVal == VARIANT_TRUE;}
private:
    CMFCPropertyGridProperty* m_pProp;
};

An example call using this is then:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

LRESULT CPropertiesWnd::OnWindowPropertyChanged(WPARAM wparam, LPARAM lparam)
{
    CMFCPropertyGridProperty *pProp = (CMFCPropertyGridProperty*)lparam;
    if (!pProp) return 0;
    CMyView* pView = getActiveView();  // Helper function uses CMainframe
    const int id = (int)pProp->GetData();  // storing property id as a DWORD using SetData
    switch(id)
    {
    case idWindow_OverallScale:
        pView->setScale( getPropValue(pProp) );    // double as an argument to func
        break;
    case idWindow_PenWidth:
        pView->setPenWidth( getPropValue(pProp) ); // int as an argument to func
        break;
    case idWindow_Title:
        pView->setTitle( getPropValue(pProp) );    // CString as an argument to func
        break;
    }
}

There you have it – it could be exactly what you need, or maybe not!

The output can be converted to time and length (in mm say) using the rulers to get a scale (mm per pixel say) and the frame rate of the video. The ‘redness’ of the liquid is the key here: the video is split into RGB and the red channel isolated and converted to black and white once a filter is applied. This image is then scrubbed to get rid of any objects smaller than a threshold size, leaving only the liquid of the column. The image is then split in two at the centre of the bounding box to analyze each column separately. Since a black and white image is represented by a matrix of 1 and 0s, summing the columns of the matrix (ie the vertical direction) gives the column height in pixels. A few other adjustments are needed, but that’s the basic idea.

(253 hits)

I’m pleased to say my implementation of the Nelder-Meade Simplex Algorithm (on my C++ page here (271 hits)) was able to optimize it for each minima (depending on the starting locations of course). I did have a weird problem though – starting a solution at (0,0) failed, even though the minima are not equidistant from the origin. Go figure!

Candidates interested in pursuing PhD research in Bridge Loading: statistics and dynamics should send a CV to info@colincaprani.com (532 hits).

Candidates interested in self-funding research to MSc level are also invited to send their CVs.

Again, some motivation is given in the chart below, and in this post (215 hits).

(200 hits)

The Worst Bridge Collapses in the Past 100 Years (480 hits)

(404 hits)

Mianus Bridge, Greenwich, Conn. (June 28, 1983)

Another collection is here:
http://www.environmentalgraffiti.com/news-seven-quaking-bridges-collapsed-costing-lives (653 hits)