It's better to upload things like subcircuit files to Files => Temp rather than emailing them.

Go to the web page: https://groups.io/g/LTspice/topics. Click on Files in the list on the left. Then click on Temp. Then click on New Upload in the blue box at top left. Click on Upload File in the drop-down menu. Then send a message to tell us that you did that.

In the sub-circuit, RLEAK = 1k is not realistic. I would expect a good cap to have RLEAK = 1 megohm or even more.

I got this model off ChatGPT, possible starting point....

 

* Electrolytic Capacitor Model
.SUBCKT ELECTROLYTIC_CAP C1 C2
* C1 is the capacitance, C2 is a parallel resistance to simulate leakage
* The temperature coefficient and voltage effects can be incorporated with controlled sources

* Parameters
.PARAM C0=100u        * Nominal capacitance at room temperature
.PARAM ESR=0.1       * Equivalent Series Resistance
.PARAM RLEAK=1k      * Leakage resistance
.PARAM TCOEF=0.0039   * Temperature coefficient (for capacitance)
.PARAM VMAX=25       * Maximum voltage rating
.PARAM TEMP=25       * Nominal temperature (in Celsius)

* Capacitance model
C1 N001 0 {C0*(1 + TCOEF*(TEMP - 25))}
* ESR and Leakage
R1 N001 N002 {ESR}
RLEAK N002 0 {RLEAK}

* Voltage dependence (simplified)
* The capacitance reduces as voltage increases; modify as needed
* Here we assume a linear decrease in capacitance over the voltage range
V1 N003 0 DC {V(VCC)}
C2 N003 N004 {C0*(1 + TCOEF*(TEMP - 25)) * (1 - (V(VCC)/VMAX))}

* Connections
VCC N004 0 DC 0

.ENDS ELECTROLYTIC_CAP

-- 
OOO - Own Opinions Only
Best Wishes
John Woodgate
Keep trying

I got this model off ChatGPT, possible starting point....

 

* Electrolytic Capacitor Model
.SUBCKT ELECTROLYTIC_CAP C1 C2
* C1 is the capacitance, C2 is a parallel resistance to simulate leakage
* The temperature coefficient and voltage effects can be incorporated with controlled sources

* Parameters
.PARAM C0=100u        * Nominal capacitance at room temperature
.PARAM ESR=0.1       * Equivalent Series Resistance
.PARAM RLEAK=1k      * Leakage resistance
.PARAM TCOEF=0.0039   * Temperature coefficient (for capacitance)
.PARAM VMAX=25       * Maximum voltage rating
.PARAM TEMP=25       * Nominal temperature (in Celsius)

* Capacitance model
C1 N001 0 {C0*(1 + TCOEF*(TEMP - 25))}
* ESR and Leakage
R1 N001 N002 {ESR}
RLEAK N002 0 {RLEAK}

* Voltage dependence (simplified)
* The capacitance reduces as voltage increases; modify as needed
* Here we assume a linear decrease in capacitance over the voltage range
V1 N003 0 DC {V(VCC)}
C2 N003 N004 {C0*(1 + TCOEF*(TEMP - 25)) * (1 - (V(VCC)/VMAX))}

* Connections
VCC N004 0 DC 0

.ENDS ELECTROLYTIC_CAP

Sorry to disappoint you and ChatGPT, but this won't work. You can't use node voltages in normal capacitor values. Even if it did, presumably VCC is a fixed voltage, so the capacitance wouldn't actually change according to the voltage across the terminals. If the node used (VCC) were actually the voltage across the terminals, consider what would happen when it was reversed biased - the capacitance would get bigger, such that it would be twice the capacitance at VCC=-VMAX as it was at VCC=0!

For temperature variation of capacitance, which normally wouldn't be expected to change dynamically, that is already dealt with in the default capacitor. Please read the Help chapter on capacitors.

For non-linear capacitors, there is a variation of the standard one called the charge-controlled model. This uses the special variable "x" to denote the voltage across the terminals. This is also covered in the Help chapter.

For some worked examples of this, go to the group Homepage > Files and type "non-linear capacitor" in the search box and click on the magnifying glass.

--