Hi everyone,
I am not familiar with posting on a forum so please tell me if more (or less) information is needed, I tried to be as extensive as possible. Also, English is not my first language so if anything is unclear, please let me know.
The model :
I am working with a model that I did not design myself and I only have access to the idf file. I am using EnergyPlus 24.2.
A base model of a single-family house was modified to have different isolation levels (4) and inertia levels (3). In total, I have 8 different “building versions” of the same house.
The general model is a single-family house with 6 heated rooms and 2 not heated. The system used here is an air-water heat pump (HeatPump:PlanLoop:EIR:Heating) connected with baseboard heaters (ZoneHVAC:Baseboard:RadiantConvective:Water). The baseboard heaters are in parallel in the PlantLoop (with Connector:Splitter and Mixer). I have 2 bypass branches (Pipe:Adiabatic), one parallel to the heatpump and one parallel to the heaters. I only have one loop (CommonPipe=None) and the pump is a VariableSpeed at the beginning of the Plant Side. The pressure simulation is disabled.
I am trying to model the use of low-temperature baseboard heaters (inlet temperature 45 to 50 °C) and I found data from manufacturers with a Delta T of 10°C (I put it in Sizing:Plant), with the different Heating Design Capacities autosized and I calculated the corresponding rated water mass flow rate. The thermostat is on the Operative temperature.
In case this is relevant, I am simulating the model from python with a fmu (package fmpy), and I give schedules for heating/cooling setpoints (communication at each timestep) and other things (that are constant for now).
The issue :
When I simulate this (over a year, normal weather conditions), an instability appears after some time (1 to 300 days in). The model is stable and converges fine during the beginning of the simulation but once the instability happened, the model stays unstable until the end. When I say instability, I mean that the mass flow rate in the baseboard heaters oscillates, sometimes between 0 and a fixed value, sometime between two non-zero fixed values, and these values can change during the simulation. This causes the air and the operative temperatures, as well as the power from the heat pump to oscillate. In some of my tries, only one baseboard had an instable behavior, in others, it’s all of them. Often, not all the model are unstable, it’s 3-4 of them, but it’s not always the same ones, it changes depending of my attempts to stabilize the lot.
Therefore, I can fine-tune each of the model by hand to reach stability but it feels like covering up something I don’t understand. Also, I am trying to obtain a general rule for all of the models, because in the future I will use different weather files and I cannot fine-tune them for each case I have, it will take too much time.
My observations so far: (I did a lot of things so here is everything I can recall)
It’s easier to have stability when my heating setpoint from the fmu does not changes (I reached stability for all of the models with common “sizing rules”) but not when I vary my setpoint (usually lowered by 2°C at some point in the day and then back up 4 hours later, every day).
This unstable behavior does not appear with other systems (I have another set of models with LowTemperatureRadiant:ConstantFlow or with air heating and they are working fine).
I had a timestep of 10 minutes (6 timestep per hours) but even when I go down to 1 minute, it is not stable.
I tried replacing the VariableSpeed pump with a ConstantSpeed one but it does not solve the issue.
I tried removing one or both bypasses but it does not solve the issue either.
Currently my radiant fraction for the ZoneHVAC:Baseboard:RadiantConvective:Water:Design is 0.3. When I set it to 0, then all the models are stable but that’s not the behavior I want to model so I didn’t take that as a solution.
One parameter that seems to have a large impact is the “Maximum water flow rate” of the Baseboard models. I read the documentation so I think I understand how it is used in the equations (for calculating the air flow rate) but when I let it to autosize, instability appears in all the models, even when I keep my heating setpoint constant. If I set it to be the largest rated flow rate of the baseboard heaters for all of the heaters then some models are stable (even with the heating setpoint variation) but some are not and if I set it to an average value, some (not necessarily the same) models are stable, some are not.
I tried limiting the mass flow rate in the pump and/or PlantLoop manually to the sum of rated flow rates of the heaters but again, some models are stable, some are not.
I also tried to modify the convergence tolerance of ZoneHVAC:Baseboard:RadiantConvective:Water:Design but again, it stabilizes some models and not others.
When I look into the .err file I see nothing alarming, eventually the following error, but the model still runs and with how the “Design air flow rate” is calculated (0,0062+0,0000275*Q_des, if understood correctly), combined to my low inlet temperature, it seems logical.
* Severe * SizeHWBaseboard: ZoneHVAC:Baseboard:RadiantConvective:Water="MAISON:CHAMBRE3 RADIATEUR EAU". * ~~~ * ...Air Outlet temperature must be below the Water Inlet temperature * ~~~ * ...Air Outlet Temperature=[42.33], Water Inlet Temperature=[41.41]. * ~~~ * ...Air Outlet Temperature set to [41.40].
Sometimes, when I activate the “Display Extra Warning”, I have the “HVAC simulation not converging” warning, even if I increase the number of allowed iterations up to 100, but the model can be unstable and not show this warning.
When I activate the “Display Zone Air Heat Balance Off Balance”, there is the next warning appearing in the .err file, for all the rooms, at it happens more than 10 000 times for each room in the simulation (one year).
* Warning * Zone Air Heat Balance is out of balance for zone named MAISON:CHAMBRE2 * ~~~ * Zone Air Heat Balance Deviation Rate is more than 44.6 {W} * ~~~ * Occurrence info: Environment=IGC BORDEAUX (01-01:31-12), at Simulation time=01/08 20:00 - 20:05
Does anyone have an idea on how to solve that? Thank you in advance for your help!
I wrote a lot, sorry about that, I tried to be as concise as possible, while being exhaustive. I am not an expert in EnergyPlus, I only put my head in it 4 months ago, and on my own, so there might be obvious things I don’t know or understand properly.
