Would you please explain a bit about infeasible solutions to a unit commitment (UC) problem and what dangers and consequences it may lead to ?
by solution to UC I mean a day-ahead schedule of course.
regarding UC I am studying different paradigms used to tackle this problem. Some conservative approaches like Robust Optimization(Ben-Tal) and Info-Gap ( Ben-Haim). Their conservatism can be justified due to risk averse nature of UC.
For me, as a water resources engineering student, this risk averse and avoiding infeasibility, needs to be explained in a general fashion,
thank you in advance.
Hello,
As you may know, power systems may be subjected to contingencies that result in forced outages of generators. Therefore, the generators fleet should include a time-dependent minimum spinning reserve for handling such contingencies. Another requirement is the optimization economic operation of power systems. A third requirement is the consideration of the startup time and its related costs, and ramp rate limits of generators. All these constraints and objectives contribute in the secure and resilience operation of power systems. Security constrained UC is then among the methods of realizing secured operation of the generation subsystem with time. Improper UC may force the system to operate in an alert state due to spinning reserve shortage or even an emergency state due to steady power imbalance. You may read the relevant parts of the following publications for more knowledge about unit commitment, power system security requirements, and corrective actions.
Wood AJ, Wollenberg BF. Power generation, operation, and control. John Wiley & Sons; 2012.
Book Dynamic Security of Interconnected Electric Power Systems - Volume 1
Book Dynamic Security of Interconnected Electric Power Systems - ...
Article Stability-based minimization of load shedding in weakly inte...
Article The Penetration of Pollutant Productions on Dynamic Generate...
Thank you dear sir.
I will take a close look at those referenced mentioned above.
The one question which remains in my mind is, that if our supply do not meet demands, what happens ? do we experience any technical issue at the power plant, regardless of what will happen in demand sites.
You are welcome. Active power imbalances cause deviations in the grid frequency that probably will be global. On the other hand, reactive power imbalances cause local voltage profile and/or voltage stability problems. In the former situation, some generators may be taken out of the service due to either over or under frequency by the relevant relays' actions. That depends on the sign of the power imbalance and its magnitude. The later situation, phenomenon such as over or under voltage may be detected according also to the sign and magnitude of the reactive power imbalance. Some buses may also be disconnected due to these voltage problems. The worst situation is the voltage collapse that may be extended to large areas. In some situations, cascaded outages of components may occur leading eventually to blackout which is either frequency stability or voltage stability caused. This is a brief summary, while the elaborated descriptions are available in the previously given publications. Good luck
You specifically ask about the impact on power plants of low frequency (due to a shortfall of generation wrt demand).
If sustained (which implies frequency is only slightly below nominal), it will affect the performance of electrically-driven auxiliaries at thermal power plants (such as boiler feed pumps, coolant circulation pumps, etc.) reducing their output and hence degrading the overall power plant efficiency/output. It may also degrade the efficiency of generator excitation systems, but probably not to the extent of reducing system voltage or causing reactive power problems (in that respect I'm afraid I do not see the relevance of Mohamed's mention of MVAR imbalance/voltage instability)
Thank you James .
I appreciate all the answers.
From what I get, there would some minor technical issues at the power plant itself. which seemingly are too technical for non-power-engineers. But I am very eager to learn about them all.
now if seek to asses the effects of imbalance between demand and load generation, in a power plant, or cascades of them, on a national or region scale, how should we look at this question.
let me say once more what I said earlier, I am trying to comprehend the risk-averse nature of UC problems and justify my conservative optimization approach toward them.
Conservative optimization approaches, like Robust Optimization, are totally justifiable while it is being used in Aircraft Routing. Because of the nature of the problem, there can not be a mistake or infeasible solution. Because if we do make a mistake, there will be an air crash !
Now in UC problem, what are the risks, the safety measures, in the users places, or in the power plants, which encourages us toward conservative optimization approaches ?
Alireza,
Not sure I have all the answers.
But as I understand the situation, thermal power plant operators are very risk-averse when it comes to electrically-powered pumps that are critical to their units' operation. They would therefore be very uneasy if the TSO planned to operate the system with a risk of supply deficit, leading to a risk of low frequency. They might even elect to trip their unit(s), which might lead to cascading failures: lower frequency leading to more units tripping , and eventually a blackstart rehearsal.
Thus I find the concept of 'optimal' UC a bit academic: the solution is not equally balanced - the cost of under-commitment is much higher than the cost of committing a sufficient margin of extra plant. I think I know which way most TSOs would bias their solution! And how much margin is enough?
The problem is made more complex when you consider increasing penetration of intermittent generation: how do you price in the intermittency of must-run WTGs or the uncertainty of wind-speed forecasts?
Thanks again James,
My focus is exactly on this sentence you have mentioned:
"the cost of under-commitment is much higher than the cost of committing a sufficient margin of extra plant" .
Im looking for a comprehensive analysis of cost of under-commitment in power plants and specifically hydro-power plants.
and I understand when you say optimal solutions are rather academic and not practical, I guess due to the quoted reason to some extent.
Maybe not using reserve spins is not acceptable by operators at all. Having said that, still there is an area of research to asses a hybrid approach while we regard the UC as a stochastic process beside having minimum spin reserves.
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity