RER Renewable Energy Research  français
   Overview       Expertise       Publications       Resources        Photos    
    All       By Topic       Articles       Reports       Non-technical       Presentations    

Analysis of First Two Hybrid Test Bench Sensitivity Runs

Michael M.D. Ross
RER Renewable Energy Research

Full Text Report
Link to CETC-Varennes


Research conducted on behalf of the Photovoltaics and Hybrid Systems Program at the CETC-Varennes (Natural Resources Canada) with partial funding from the Panel on Energy Research and Development (PERD).


Ross, Michael M. D. Analysis of First Two Hybrid Test Bench Sensitivity Runs. Report to CETC-Varennes (Natural Resources Canada). Montreal, Qc: RER Renewable Energy Research, 2005.


Two tests were run on the CETC-Varennes hybrid test bench with the intention of examining the role of various parameters in operation of small PV-battery-genset hybrid system. These parameters included the solar fraction, the genset stop criterion, the genset charge current, and load variability. The first test, which achieved a solar fraction of 0.53 (considerably lower than in previous tests), used a constant load and an elapsed time criterion for determining when the genset must be stopped; the battery was cycled between 40 and 65% state-of-charge under genset charging. The second test, which achieved a solar fraction of 0.64, used both constant and varying loads and a voltage threshold to determine when to stop the genset. The battery voltage thresholds used to start and stop the genset were intended to cycle the battery between 30% state of charge and the beginning of absorb charging.

Partial state-of-charge cycling due to low solar fractions led to non-ideal battery behaviour in the two sensitivity tests analysed in this report. In the first test, which used a constant load and an elapsed time criterion to shut down the genset, this non-ideal battery behaviour manifested itself as a progressively higher battery voltage at the termination of genset charging. In the second sensitivity test, where both constant and varying loads were employed and a voltage threshold was used to determine when the genset should be shut down, the time of each genset run progressively declined as the battery voltage reached the threshold more quickly. As a consequence, the genset cycled more and more frequently, with less and less charge being returned to the battery. The PVToolbox could simulate the first test with high accuracy; simulations of the second test had errors of up to 10% in the total energy flows. For reasons of predictability, user satisfaction, and battery longevity, dispatch strategy should be designed to avoid the behaviour of the second test.

Analysis of the two sensitivity tests described in this document suggests that further tests should be run, both to examine other sets of parameters and to confirm the results of the tests already run. These tests should examine system operation with solar fractions of around 0.7 to 0.8, investigate more robust methods for determine when the genset should be started, and determine how operating the genset for one, two, or three hours of absorb charging would affect the persistence of non-ideal battery behaviour observed in the sensitivity tests studied in this report.