In a paper (page 31) discussing the installation of a battery(*) in an electricity substation, UK Power discuss the ability to control the real power and reactive power components. They go so far as to show a graph showing actual results compared to the ideal:
They also say:
The ESS (ESS = Enery Storage System, i.e. the battery and inverter) had ‘out-of-the-box’ set-points available to control real and reactive power exchange between the 11kV network and ESS.
The deviation was dependent on amount of real and reactive power the ESS was set to import or export, for instance when expecting the export of 150 kW and 150 kVAr, approximately 149 kW (99%) and only 116 kVAr (77%) were delivered. At another extreme, when set to export 150 kW and import 150 kVAr, approximately 138 kW (92%) and -118 kVAr (79%) were delivered.
As far as I’m aware in AC real power is an outcome of resistive loads and reactive power an outcome of inductive loads (e.g. synchronous motors used in industry)
Let alone the fact that 2 of the quadrants above imply opposite charging/discharging for each power component, the following plot (page 23) shows the real power (ESS P) and reactive power (ESS Q) again being independent of each other:
I feel like I’m missing some understanding, such that I’m not even sure of what question to ask other than:
How is this arbitrary control managed?
How does a load cause such a lag/difference so that the plot above (fig5) is possible?
Simultaneously charging and discharging??
(*) LiPo, obviously inverted to AC, and used primarily to limit the peak of power demand from the usual supply – i.e. charging the battery when usage is low and then tapping into it when power demand is high...