Annoying Instructor Behaviors
-
Do you have to use complex numbers since you are dealing with AC?
At least artificial intelligence already is superior to natural stupidity
Depends on how accurate you want/need your solution to be. For just calculating the powerflow I vould reverse the problem and use the loads as sources and generators/reserves as sinks, and implement it as a network flow min cost max flow problem where power loss would be cost - in this case I would not use complex numbers.
Espen Harlinn Principal Architect, Software - Goodtech Projects & Services AS My LinkedIn Profile
-
My current class, Linear Algebra, is graded on the basis of 7 quizzes and 3 exams. I just turned in my 4th quiz, and the 1st exam begins Wednesday. I still haven't seen the results of the 3rd quiz, and the instructor hasn't been online but once in a week! :wtf: How the hell are we supposed to prepare for an exam with no feedback from the previous quizzes? To those of you teaching, or considering teaching online classes, don't you dare pull this crap on your students! Teaching an online course is an easy way to make an extra couple thousand$ with a few hours' effort, but your students deserve a quick turnaround on their work. That's your first priority - always - and it only takes a few minutes to grade a math quiz. Doing a proper job of it takes a bit more time, and a lot more dedication, but the bare minimum required is that you do it fast, especially in an accelerated class like this one (10 weeks, instead of the usual 16). I'm all for turning an easy dollar, but the students are paying a couple thousand$ each to learn from you, and you damned well owe it to them to give them a timely response! That being said, the class is great, the textbook is excellent and worth keeping for reference, and the subject is one that every programmer should master. So should every engineer, but they neglected it at my school 30+ years ago. My university assumed that I'd studied it in high school, and my high school expected that I'd take it in college. Oops... They really should communicate more, me thinks. It makes me slightly ill to think back over my career, and the many hundreds, perhaps thousands of hours I spent solving system equations for electrical and electronic networks the hard way, working slowly backwards from output to input, node by node, to determine system constants for designing functional circuits. With these techniques, I could have solved for every unknown directly, in just a few minutes, by modelling the equations in matrix form and row-reducing the resulting matrix to echelon form. It would have tremendously reduced the labor required to perform sensitivity analyses for testing and reliability studies, and made failure mode analysis a snap. I probably would have understood more of the class I took in Kalman Filtering, as well, and might have been more effective in presenting the idea to my bosses of using it to optimize testing strategies and spares stocking for ICBM repair depots, too. Crap, I hate looking back and realizing how much of my time was spent working too hard! I'm lazy by nature (all
:thumbsup: That's the reason why I'm still not certified. I'm not going to sign a certificate that say that I had an education, when that's clearly not the case. It gets worse if you can offset some of that education with money. Teaching has degraded hugely in the past decades, something which we'll learn to regret.
Bastard Programmer from Hell :suss:
-
Depends on how accurate you want/need your solution to be. For just calculating the powerflow I vould reverse the problem and use the loads as sources and generators/reserves as sinks, and implement it as a network flow min cost max flow problem where power loss would be cost - in this case I would not use complex numbers.
Espen Harlinn Principal Architect, Software - Goodtech Projects & Services AS My LinkedIn Profile
-
It has been some time since I got my head stuffed full with this. At least I still know what's coming when I open one of the books you mentioned :)
At least artificial intelligence already is superior to natural stupidity
Sedgewicks' book isn't so bad, it's written with developers in mind, the other one requires a certain level of motivation ;)
Espen Harlinn Principal Architect, Software - Goodtech Projects & Services AS My LinkedIn Profile
-
Just wanted to say that learning or refreshing your knowledge can be done in an interesting and practical way. Then the instructor is not quite as important anymore.
At least artificial intelligence already is superior to natural stupidity
-
My current class, Linear Algebra, is graded on the basis of 7 quizzes and 3 exams. I just turned in my 4th quiz, and the 1st exam begins Wednesday. I still haven't seen the results of the 3rd quiz, and the instructor hasn't been online but once in a week! :wtf: How the hell are we supposed to prepare for an exam with no feedback from the previous quizzes? To those of you teaching, or considering teaching online classes, don't you dare pull this crap on your students! Teaching an online course is an easy way to make an extra couple thousand$ with a few hours' effort, but your students deserve a quick turnaround on their work. That's your first priority - always - and it only takes a few minutes to grade a math quiz. Doing a proper job of it takes a bit more time, and a lot more dedication, but the bare minimum required is that you do it fast, especially in an accelerated class like this one (10 weeks, instead of the usual 16). I'm all for turning an easy dollar, but the students are paying a couple thousand$ each to learn from you, and you damned well owe it to them to give them a timely response! That being said, the class is great, the textbook is excellent and worth keeping for reference, and the subject is one that every programmer should master. So should every engineer, but they neglected it at my school 30+ years ago. My university assumed that I'd studied it in high school, and my high school expected that I'd take it in college. Oops... They really should communicate more, me thinks. It makes me slightly ill to think back over my career, and the many hundreds, perhaps thousands of hours I spent solving system equations for electrical and electronic networks the hard way, working slowly backwards from output to input, node by node, to determine system constants for designing functional circuits. With these techniques, I could have solved for every unknown directly, in just a few minutes, by modelling the equations in matrix form and row-reducing the resulting matrix to echelon form. It would have tremendously reduced the labor required to perform sensitivity analyses for testing and reliability studies, and made failure mode analysis a snap. I probably would have understood more of the class I took in Kalman Filtering, as well, and might have been more effective in presenting the idea to my bosses of using it to optimize testing strategies and spares stocking for ICBM repair depots, too. Crap, I hate looking back and realizing how much of my time was spent working too hard! I'm lazy by nature (all
I loved Linear Algebra. Much more than discrete math which was just a bunch of proofs.
Need custom software developed? I do custom programming based primarily on MS tools with an emphasis on C# development and consulting. I also do Android Programming as I find it a refreshing break from the MS. "And they, since they Were not the one dead, turned to their affairs" -- Robert Frost
-
James W. Nilssons Electric Circuits, third edition, used LA for analysis - pretty hard to imagine doing this stuff without LA or graph theory. If you haven't read it, read Algorithms in C++ Part 5: Graph Algorithms[^] by Robert Sedgewick. You can then have a go at Power systems applcications of graph theory[^] by Jizhong Zhu. The last one shows you how to tackle problems like:
- Network flow calculation of power flow
- Economic power dispatch
- Security constrained economic dispatch
- Multi-areas system economic dispatch
- Reactive power optimisation and pricing in multi-area environment
- Hydro-thermal power system operation
- Power system state estimation
- Secure economic automatic generation control
- Automatic contingency selection
- Distribution network optimisation
- Optimal load shedding
Some LA solvers actually tries to convert the problem into a network simplex problem before attempting to find the solution using other LA methods.
Espen Harlinn Principal Architect, Software - Goodtech Projects & Services AS My LinkedIn Profile
Thanks, Espen! The second book, by Zhu, looks interesting. I'll have to grab a copy for a little light reading. :-D I think it would help with a little problem we have at work. The reservation is split - the north is on a 12.47kV network, the south running on 24.94kV. They're joined in the middle by a step-up transformer so that, in an emergency, we can feed the entire reservation from either of two substations. Because we don't know what will happen if we loop these two sections through the step-up, we've always had to take an outage to do any switching, and that annoys customers. If I could estimate the power flow magnitude and direction, I might find that it's perfectly safe to run them interconnected all the time, providing a permanent redundancy for customers! Of course, I might also find that it's a disaster waiting for me to throw the switch...:suss: Besides, Zhu works for Areva T&D, one of my favorite suppliers. In fact, the last 69kV substation we built used Areva 69kV CCVTs for monitoring bus voltage. Thanks for the link! :-D
Will Rogers never met me.
-
Thanks, Espen! The second book, by Zhu, looks interesting. I'll have to grab a copy for a little light reading. :-D I think it would help with a little problem we have at work. The reservation is split - the north is on a 12.47kV network, the south running on 24.94kV. They're joined in the middle by a step-up transformer so that, in an emergency, we can feed the entire reservation from either of two substations. Because we don't know what will happen if we loop these two sections through the step-up, we've always had to take an outage to do any switching, and that annoys customers. If I could estimate the power flow magnitude and direction, I might find that it's perfectly safe to run them interconnected all the time, providing a permanent redundancy for customers! Of course, I might also find that it's a disaster waiting for me to throw the switch...:suss: Besides, Zhu works for Areva T&D, one of my favorite suppliers. In fact, the last 69kV substation we built used Areva 69kV CCVTs for monitoring bus voltage. Thanks for the link! :-D
Will Rogers never met me.
Brilliant :-D
Espen Harlinn Principal Architect, Software - Goodtech Projects & Services AS My LinkedIn Profile
-
This does sound interesting Roger - I'd love to know the title of the textbook, if you can recommend it. Might make good holiday reading! Al
Hi Al! In case it's not obvious, I'm enjoying the course enormously. The text is Linear Algebra And Its Applications, Third Edition (Update) by David C. Lay, ISBN 9-780321-287137, and it's dirt cheap at Amazon. I think I paid $54 there, while the school online bookstore wanted $152! Ripoff!:mad:
Will Rogers never met me.
-
I loved Linear Algebra. Much more than discrete math which was just a bunch of proofs.
Need custom software developed? I do custom programming based primarily on MS tools with an emphasis on C# development and consulting. I also do Android Programming as I find it a refreshing break from the MS. "And they, since they Were not the one dead, turned to their affairs" -- Robert Frost
Though much easier, it reminds me somewhat of Calculus. Once I got over the conceptual hurdle of limits, and the unfamiliar terminology of the delta/epsilon explanation of why it works, something clicked. From then on, calculus made all the previously hard calculations using logs, trig, and geometry much easier by integrating them all into one whole picture. Laplace transforms were just icing on the cake, so to speak, turning calculus into mere arithmetic. Linear algebra opens up a whole new world of analysis and synthesis possibilities!
Will Rogers never met me.