Regulation Services and Electrolyzer
Opened this issue · 4 comments
@rkadavil @jingjingliu2018 @Hayden-Reeve
Traditional_Regulation / Electrolyzer
Drive cycle file: 08 2017 Traditional.csv
Time: 08/01/2017 16:00 to 18:59
Sim Step: 2s
Service Wt. 0.75
Assigned Service MW: 100 MW
The request from the traditional service is to reduce power to the grid over this time period, with some variation starting out as -100 MW and varying over time between -100 and -9.6 MW. The base power over this period seemed to start at -10 MW and was reduced by the request signal to -109 MW at the start of the drive cycle. Over the remainder of the drive cycle, P_response tracks P_request closely. P_togrid is lower than P_response by the approximate magnitude of P_base. P_response = P_togrid – P_base as it should for the entire period.
_
The request from the traditional service is to reduce power to the grid over this time period, with some variation starting out as -100 MW and varying over time between -100 and -9.6 MW. The base power over this period seemed to start at -10 MW and was reduced by the request signal to -109 MW at the start of the drive cycle. Over the remainder of the drive cycle, P_response tracks P_request closely. P_togrid is lower than P_response by the approximate magnitude of P_base. P_response = P_togrid – P_base as it should for the entire period.
As there is a significant difference in the power draw under the regulation signal than compared to the base electrolyzer load, is this realistic for the electrolyzer load to be so lightly loaded in the base relative to the regulation scenario? A scaling issue?
Dynamic_Regulation / Electrolyzer
Drive cycle file: 08 2017 Dynamic.csv
Time: 08/01/2017 16:00 to 18:59
Sim Step: 2s
Service Wt. 0.75
Assigned Service MW: 100 MW
Observation
The request from the dynamic regulation service is to reduce power to the grid over this time period, with some variation starting out as -99 MW and varying over time between -100 and +100 MW. The base power over this period is -10 MW (Electrolyzer Fleet Power draw) and was able to respond to the initial negative power request by increasing the power draw almost identical to the request initially in response to the first ~11 minutes of service request. The service request then goes positive and the P_response and P_togrid react to the request twice before about hour 1620 by twice showing reduced power draw by the electrolyzer fleet, with P_togrid going to 0. The P_request then goes negative again and the electrolyzer fleet reacts by increasing its load above P_base in an amount approximately equal to the request. For the remainder of the test period, the response tends to track the request where the request is negative, but the response is a maximum of +10 for positive requests greater than 10.
The positive 10 MW response reflect a reduction of the baseline power draw to 0. The P_togrid goes to 0 during these periods. The speed of response is very fast as is expected for the electrolyzer
P_response = P_togrid – P_base as it should for the entire period.
As there is a significant difference in the power draw under the regulation signal than compared to the base electrolyzer load, is this realistic for the electrolyzer load to be so lightly loaded in the base relative to the regulation scenario?
@DavidWiniarski-pnnl , were the results above run with the master code or PR #91 ?
@DavidWiniarski-pnnl , were the results above run with the master code or PR #91 ?
@DavidWiniarski-pnnl I would request you to hold on running the tests. We are still incorporating the changes for the Pbase and Pservice calculations for both FuelCell and Electrolyzer models.
@rkadavil
Will do. I will hold on any future runs till I hear back. I will note that for fuel cells there is an exception handling issue in the regulation services that is crashing the regulation run with fuel cells only. . Hayden pointed this out earlier, and we have not run this since.