---
language:
- en
license: mit
tags:
- process-control
- industrial-automation
- qlora
- llama
- fine-tuned
- mpc
---

# Experiment C — Equal-Weight Multivariable MPC, 3,000 Training Records

**Paper**: LLMs as Autonomous Controllers for Multivariable Industrial Fluid Processes
**Author**: Vidyashree Rayar — BTU Cottbus-Senftenberg

Equal-weight MPC cost function (1.5×level + 1.5×temp). Best temperature accuracy across all experiments: 67.8% with SP+CoT+P. Isolates the effect of cost function design from data quantity.

## Base Model

[meta-llama/Llama-3.2-3B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct) fine-tuned with QLoRA (4-bit NF4, LoRA r=16, alpha=32, rsLoRA enabled).

## Code & Results

[github.com/vidyashreerayar/festo-llm-process-control](https://github.com/vidyashreerayar/festo-llm-process-control)