MATPOWER does not include a multi-period OPF. It may be possible to implement 
what you want on top of MATPOWER's extensible OPF solver, but you would need to 
know exactly what the problem formulation looks like and how you would map it 
onto the extended formulation described in Section 5.3 in the manual. You would 
need to implement the code yourself.

-- 
Ray Zimmerman
Senior Research Associate
419A Warren Hall, Cornell University, Ithaca, NY 14853
phone: (607) 255-9645




On Mar 26, 2012, at 11:16 AM, Carlos Gonzalez Almeida wrote:

> Dear Dr. Zimmerman,
> 
> Thank you so much for helpful comments. I want to maximize Social Welfare 
> considering offers and bids for 10 years by using a multi-period OPF.
> 
> Is it possible?
> 
> Best Wishes
> 
> C. Gonzalez Almeida
> 
> 
> On Mon, Mar 26, 2012 at 14:49, Ray Zimmerman <[email protected]> wrote:
> When I think of a multi-period OPF, I think of something that simultaneously 
> solves for the dispatch of several different time periods, typically with 
> some inter-temporal constraints of some sort. If that's what you mean too, 
> then I think the answer is no. The code below cannot be used to solve such a 
> multi-period OPF. If all you want to do is solve a sequence of independent 
> OPFs, or even a sequence of OPFs where the inputs of one depend on the 
> outputs of the previous, then certainly the structure of that code can be 
> used.
> 
> -- 
> Ray Zimmerman
> Senior Research Associate
> 419A Warren Hall, Cornell University, Ithaca, NY 14853
> phone: (607) 255-9645
> 
> 
> 
> 
> On Mar 23, 2012, at 5:23 PM, Carlos Gonzalez Almeida wrote:
> 
>> Dear Dr. Zimmerman,
>> 
>> Can I do a multiperiod OPF by this method, i.e. as the following code that 
>> you have written?
>> 
>> define_constants;
>> mpc = loadcase('mybasecase');
>> for k = 1:52
>>     mpc.bus(:, PD) = load_profile(:, k);
>>     mpc.gen(:, PG) = gen_profile(:, k);
>>     r = runpf(mpc);
>>     loss(k) = sum(mpc.gen(:, PG)) - sum(mpc.bus(:, PD));
>> end
>> total_loss = sum(loss);
>> 
>> Regards
>> 
>> C. Gonzalez Almeida
>> 
>> 
>> On Thu, Mar 22, 2012 at 13:39, Ray Zimmerman <[email protected]> wrote:
>> Here's what I would do. Create a matrix gen_profile with the generator 
>> output values, where each row corresponds to a generator and each column to 
>> a week of the year. You could set up another matrix load_profile with the 
>> nodal loads in it, then do something like the following ...
>> 
>> define_constants;
>> mpc = loadcase('mybasecase');
>> for k = 1:52
>>     mpc.bus(:, PD) = load_profile(:, k);
>>     mpc.gen(:, PG) = gen_profile(:, k);
>>     r = runpf(mpc);
>>     loss(k) = sum(mpc.gen(:, PG)) - sum(mpc.bus(:, PD));
>> end
>> total_loss = sum(loss);
>> 
>> Hope this helps,
>> 
>> -- 
>> Ray Zimmerman
>> Senior Research Associate
>> 419A Warren Hall, Cornell University, Ithaca, NY 14853
>> phone: (607) 255-9645
>> 
>> 
>> 
>> 
>> On Mar 22, 2012, at 4:24 AM, Ruhaizad Ishak wrote:
>> 
>>> Dear All,
>>>  
>>> First of all please forgive if the question that I wanted to ask here 
>>> sounds repetitive. I am new in this community.
>>>  
>>> I had browsed through the archive but couldn't find any similar topic or 
>>> otherwise I may missed it.
>>>  
>>> Although found one closely related on how to add generator but mine is a 
>>> liitle bit different.
>>>  
>>> Using the existing generators (same number of generators) how can I change 
>>> the output of each generator?
>>>  
>>> So far I tried the same method for changing load power demand but it 
>>> doesn't work.
>>>  
>>> Apart from changing the output, I also plan to set an output profile for 
>>> each generator for certain time frame (such as 52 weeks.)
>>>  
>>> For example:
>>> Generator 1
>>> Week 1: 20 MW
>>> Week 2: 40 MW
>>> Week 3: 25 MW
>>>  
>>> Generator 2
>>> Week 1: 10 MW
>>> Week 2: 25 MW
>>> Week 3: 13 MW
>>>  
>>> Then I would like to find the total losess for the system in MWh unit.
>>>  
>>> Can someone guide me on this?
>>>  
>>> I appreciate your helpful advice and comment in advance.
>>>  
>>> Thank you.
>>>  
>>>  
>>>  
>>>  
>> 
>> 
> 
> 

Reply via email to