I have modeled a PV, Battery, Load system. For my locale, the PV generates 5,785 kWh annually, which equates to about 660 W continuously. I set the load for 660 W continuously. The 16 batteries each are rated at 130 Ah with a nominal voltage of 12.8 VDC, maximum charge current of 20 A, and 92 percent efficiency. I filled in the rest of the battery info per the manufacturer's info. The electrical simulation results say there is 5 percent excess electricity and 9 percent unmet load. But it also says the batteries never reach 100 percent SOC. How can there be excess electricity that is not being used to charge the batteries?

That battery's low maximum charge rate is causing the excess electricity to occur even before the battery reaches 100% state of charge.  I changed the maximum charge rate to verify that.  Here is the scatterplot of excess electricity versus state of charge before I changed the maximum charge rate:

Now here is the same scatterplot after I increased the maximum charge rate to 9999 A/Ah:

That second graph shows that when the maximum charge rate is not a factor, excess electricity happens only when the battery is completely full. 

I changed the maximum charge rate only to determine whether it was causing the excess electricity.  You should not change it unless you think the existing value is incorrect.