Long-Short Week

Posted Fri Oct 17, 2008 in

Chess SetIt was a busy week and it isn’t over yet. I spent part of the week babysitting another CLOMR submittal (finished the report last week, few details hanging over into this week) and worked on the report for yet another CLOMR submittal. While writing the report, I found a problem with the modeling.

The problem is sufficiently interesting to own a paragraph or two of its own. The proposed project needs a little fill on the floodplain fringe. Not enough to cause much of a rise in the water-surface elevation under flood conditions, but enough to require modeling to satisfy regulatory requirements. The last cross section affected by the proposed project is immediately downstream from a culvert crossing. If I set an encroachment station using Method 1 in HEC-RAS, then critical flow occurs at the section immediately upstream from the culvert. The degree of encroachment (1-ft or 500-ft) makes no difference. An ineffective flow area (not quite the same as an encroachment, but close enough) does not cause critical flow. I think (but can’t prove) that the encroachment causes the culvert computations to GO ALL WONKY.

I packaged up the model and sent it to the Corps of Engineers. I doubt they’ll respond to a private-sector engineer, but they might look at the model and figure out what’s wrong with it (or their program).

I continued writing my report, including what I learned about the affected cross section (and using the ineffective flow area work-around). That work is nearly finished — all it needs is for me to write the conclusions and the executive summary.

I left work a little early Wednesday with a headache and needs to get packed for a trip to Texas for more work. I have two defenses, one Ph.D. and one Master’s, scheduled for Friday and Monday. That leaves one graduate student to shepherd through the process. I’ll work with colleagues on project work to finish out a week or so, then return home.

Travel yesterday wasn’t too bad. The trip to Texas is manageable, although it still requires too much time in airports and on aircraft. It was good to have Young Son along. He’s good company and we had a lot of time to talk — something that doesn’t happen enough at home with all the distractions there.

Now up and at ‘em for me. I need a shower, some coffee, some food, and then to get to campus.

Oh yeah… the chess set. This is a large plastic set they have at the Lubbock airport. The king is about 24” tall. Notice anything wrong? “White on the right…” is the rule for the board. The incorrect board orientation results in the king and queen occupying reversed positions and so a king’s-side castle would be inverted. Weird… and wrong.

Comment

First CLOMR

Posted Wed Oct 1, 2008 in

After almost two years, the first approval of one of my Conditional Letter of Map Revision applications came through. It is difficult to believe just how long the process takes.

Comment [2]

D.WRE

Posted Mon Sep 29, 2008 in

I received notice today that I qualified for the Diplomate of Water Resources Engineering. This is a certification that I reached some level of expertise in my discipline specialty. Payment of the annual fee (or course) enables me to include those initials in my signature and confers some recognition that I achieved a certain level of expertise.

I suppose the D.WRE and a couple of bucks will buy a Starbucks coffee. :) But, it’s a good thing to have the certification. It lends credence to my capabilities. I’ll take that.

Comment

Ready… Set… Whoa!

Posted Thu Sep 18, 2008 in

I really need to hit the shower, but I’m going to write a bit first. (I’m also working my way through my first cup of the day!)

As I expected, yesterday was intense. The presentation started about 0930. The lead researcher went through his part, then I took over to talk about bivariate distributions and copulas, and the third man took over from me to discuss the total probability theorem.

I put them to sleep. I know, I know, not everyone thinks bivariate distributions are all that exciting, like I do. But, give me a break… This is substantial work and non-trivial and even if it is technical, it’s still good work and deserves something better than a nap!

I learned I can still put my students to sleep. However, in this case, I could not offer a quiz at the end of the hour. Humph…

After lunch, we worked through a few more questions. Then, as I expected, the most knowledgeable person on the panel decided my joint distributions were not. So, he stood up and proceeded to demonstrate why the joint distribution is not bounded by the appropriate marginal probabilities. Although I was confident in my work, I decided I would review my approach, ensure I had it right, and then rebut. However, I didn’t expect what happened as the discussion wore on.

The bulk of the panel dismissed for the remainder of the day. A few of us remained behind to work on things together. A few of the technical-types remained behind as well to continue the discussion. As we worked through a number of sticky problems, the other strong technical type reversed his position and decided that, indeed, the joint probability is bounded by the marginal probabilities. The discussion took an entirely different direction from there on. While interesting to me and the other engineers present, it will probably not be interesting to Random Ruminations readers, so I’ll leave the technical details out (or at least save them for a rainy day).

“Once in awhile, even a blind pig finds an acorn.” I did my homework and was pretty confident my approach and computations were correct. While I believed the other party was incorrect, I wasn’t sufficiently confident to take him on in a public forum, especially because so much of my work was recently completed and had not had time to settle. My modus operandi is such that I work in fits and spurts, with the in-between time dedicated to letting my subconscious mind mull over data and information and find the patterns that lend me insight. I hadn’t had enough time for this process to complete, so I wasn’t at my best. That meant I needed to let it sit for a bit and think about what he said before rebutting.

However, his own work-through-it process kicked in and he figured out the problem during the discussions that followed the formal meeting. That was pretty-damned-cool. I’ll leave here confirmed and will move my understanding further as I solidify my work and move the computations forward to the full dataset. Some damnfine work will come from this. The schedule is tight, but I’m pretty confident we’ll make it.

After the meeting we decompressed a bit at the hotel bar. I found a dark beer I liked and was about to settle in for the night when I was conscripted to go eat. I think we walked halfway across the city. It isn’t that I didn’t need the walk, my head was pretty worn out from all the thinking. Then, in proper engineer fashion, the lot of us talked shop off and on over supper. The beer was good, and the burger was great. Then, I was done. I watched a little television to settle down. But then, that’s another story.

Now I’m off to the shower and then to the meeting to wrap this thing up. I know I’ll be glad to get home again.

Comment [2]

Me-Me Meme Done

Posted Wed Jul 16, 2008 in

Finally, after dancing all around the damned thing, the me-me meme is finished, kaput. I hate leaving things undone, so I had to finish it. I think there must be more of a story there, somewhere. But, I can’t seem to ferret it out. I may have to think on this some more, and there is road time coming when I will be thinking.

I wrote this early Sunday morning and by the time it posts I’ll be at breakfast with Wife and a Friend. I’ll be thinking about other things by then and be focused on accomplishing the task-at-hand, which will be working on some professional writing. The hope will be to finish a couple of projects and generating proposals for a couple more projects.

One of the things apparent to me (now) is that I am a writer. I don’t write fiction. I mostly write about technical things — to convey results from a piece of work to a combination of technical and non-technical audiences. Some who read my work are specialists and appreciate the details of what I do. Others, though, are not specialists although they might be technical. They understand what I do in a broader context, generally in the context of what they need to do.

My research is applied research. If it is applied, it is the technical end-user who uses the technology I help develop. The idea is that what I do helps someone else do their job more effectively, ether in terms of their time spent or in terms of the cost of the project.

I also do a lot of applications these days. The end result is a report that goes to the reviewing agency (if it’s a flood hydraulics study) or to another consumer of technical products if it’s another type of study.

But, each bit of work results in a written report. For years I told my students “The principal product of an engineer is the written report. Learn to write.” Most chose not to believe me. I was right.

P.S. I see this entry morphed from one thing to another. This kind of writing is interesting in that sense. There is an element of stream-of-consciousness that I really enjoy as I sit at the keyboard for a few minutes to explore what might happen. Sometimes it’s good; other times not. I’ll leave it to others to make the value judgment.

Comment

A Week

Posted Fri Mar 28, 2008 in

It’s been a week. I have several high-priority projects all nipping at my heels. All of them want attention and all of them want attention now. There is simply not enough of me to go around, so I told my boss “I need a clone. In fact, I need two clones!” We laughed.

My biggest concern is my joint-probability project. This one requires a lot of effort to get up to speed on bivariate statistical modeling via copulas. I spent nearly two weeks reading and generating tools that will help me make the required computations. This has not been trivial.

I’ll hand off some of the bivariate work that I think is mostly complete to the project lead so he can complete the quarterly progress report. Then I’ll redirect my attention to the copula problem and think I can have some technology operational in a few more days.

I’m learning a lot.

Comment

Programming

Posted Wed Mar 19, 2008 in

I wrote my first “R”: script yesterday. It’s a simple script that computes the joint probability from the observations — the empirical distribution. I need this to use in testing the fits of the copulas I’m fitting.

The code looks like:

dufus<-function(peak,vol) {n<-length(peak) jp<-matrix(rep(0,n)) for (i in 1:n) { j2<-vol-vol[i] numv<-sum(j2<=0) jp[i]<-(numv-0.44)/(n+0.12) } jp}

Right now I’m not using the variable peak for anything, but I put it in the argument list so I could use it later as I modify the script. I see I could drop j2 and put the computation inside the sum function. That would be slightly more efficient, but I wasn’t thinking efficiency when I wrote the script. I just need to solve the problem.

That leads to another problem I’m working on. The parameter for the Frank copula requires solution of a Debye function. I’ve been doing some reading on how to either approximate the function or integrate it. I found an ACM algorithm that will do the trick. But, I’ll have to recast the code in R, I think. I’ll figure that one out later.

Now off to work…

Comment

| Next Page