4-Room Home Heating System: Model-Based Design

This repository contains the complete model-based design, simulation, and implementation of a smart heating system for a 4-room house. The project was developed as part of the "Embedded and Real-Time Systems" course and utilizes MATLAB, Simulink, Stateflow, and Embedded Coder.

📌 Project Overview

The goal of this project is to model and control a house with 4 rooms, utilizing a maximum of 2 heaters at any given time. The system's core dynamics account for:

• Heat transfer between adjacent rooms.
• Heat loss/gain from the external environment.
• The heating power of active heaters.

The controller dynamically allocates the 2 available heaters across the 4 rooms based on predefined temperature thresholds (ON/OFF limits) and priority logic, ensuring all rooms remain within the desired temperature range (15°C to 20°C).

🏗️ System Architecture

The system architecture was implemented in Simulink and consists of three main subsystems:

1. Rooms (Plant Model): Calculates the temperature dynamics of the 4 rooms using continuous-time equations.
2. Thermostat (Controller): Implemented using Stateflow, this module dictates heater allocation and states based on current temperatures.
3. Safety Checker (Doer-Checker Pattern): Ensures that safety requirements are met (e.g., preventing more than two heaters from being active simultaneously) before passing commands to the plant.

Figure 1: High-level Simulink architecture showing the Thermostat, Rooms, and Checker modules.

🧠 Stateflow Controller (Thermostat)

The core logic for heater allocation is built using Stateflow. Because there are 4 rooms and only 2 heaters, the system has $\binom{4}{2} = 6$ primary active states.

The state machine evaluates temperature differences and thresholds ($on_i$, $off_i$, $get_i$, $dif_i$) to safely and efficiently transition heaters between rooms without violating constraints.

Figure 2: Stateflow chart demonstrating the 6 primary states and transition logic for heater allocation.

📊 Simulation & Results

The system was simulated with an initial temperature of 16.5°C across all rooms and an external temperature of 6°C. The target was to maintain room temperatures between 15°C and 20°C.

As seen in the simulation output below, the controller successfully regulates the temperatures. When a room drops below its threshold, a heater is reallocated to it, causing the characteristic oscillatory temperature curves within the safe operating envelope.

Figure 3: Scope output showing the temperature variations of the 4 rooms over time, remaining within the 15-20°C target range.

💻 Code Generation & Wokwi Simulation

As an optional extension, the system's control logic was generated into C code using Embedded Coder and tested on an Arduino framework using the Wokwi online simulator. In this pure software simulation, LEDs were utilized to visually represent the active states of the heaters in real-time.

Figure 4: Software simulation of the generated C code on an Arduino framework using Wokwi.

📂 Repository Structure

• models/: Contains the core Simulink models (.slx) and requirement files (.slreqx).
• scripts/: MATLAB scripts for initializing parameters (e.g., $A, b, c$ matrices and initial conditions).
• tests/: Simulink Test cases (blackboxTest.mldatx) for automated requirement verification.
• docs/: Project instructions, final PDF reports, and images used in this README.

---

🙌 Creators

This repository is the result of our teamwork:

• 🧑‍💻 Milad Ansari
• 👨‍💻 Sina Samadi