Expert System Project - Discovery Piscine AI/ML

Overview

This project implements a simple expert system that regulates indoor temperature based on outdoor temperature fluctuations. It consists of two main components:

1. Environment Simulator (environment.py): Simulates outdoor temperature throughout the day for different seasons.
2. Expert System (expert_system.py): Controls heating and cooling systems to maintain indoor temperature within a comfortable range.

What is an Expert System?

An expert system is a basic type of artificial intelligence that uses predefined rules and knowledge to simulate decision-making. It operates within a state machine paradigm, where the system can only be in one specific state at a time, with external events triggering state changes.

Requirements

• Python 3.x

Usage

Environment Simulator

Simulates outdoor temperature variations throughout a day:

python environment.py <season>

Where <season> can be:

• "winter", "spring", "summer", "autumn" (as text)
• 1, 2, 3, 4 (as numbers representing winter, spring, summer, autumn)

The program outputs temperature values every second, with each second representing a 30-minute interval in the simulation (48 seconds = 24 hours).

Expert System

Takes input temperatures and manages heating/cooling to maintain comfort:

python environment.py <season> | python expert_system.py <min_comfort> <max_comfort>

Where:

• <min_comfort>: Lower threshold of the comfort zone temperature
• <max_comfort>: Upper threshold of the comfort zone temperature

How It Works

Environment Simulator

• Uses a sine wave to model daily temperature fluctuations
• Adjusts temperature ranges based on the selected season
• Adds small random variations for realism
• Outputs one temperature value per second

Expert System

• Reads outdoor temperature values from standard input
• Initially sets indoor temperature equal to outdoor temperature
• Activates heating when temperature falls below minimum comfort level
• Activates cooling when temperature exceeds maximum comfort level
• When active, heating increases indoor temperature by 0.5°C per cycle
• When active, cooling decreases indoor temperature by 0.5°C per cycle
• When inactive, indoor temperature gradually moves toward outdoor temperature (±0.25°C per cycle)
• Outputs current state: <external_temp> - <action> - <indoor_temp>

Example Output

7.5 - heating - 8.0
8.0 - heating - 9.0
8.5 - heating - 10.0
9.0 - heating - 11.0
10.0 - heating - 12.5
11.0 - heating - 14.0
12.0 - nothing - 13.75
13.25 - nothing - 13.5
14.5 - nothing - 13.75
15.0 - nothing - 14.0
15.5 - nothing - 14.25
15.5 - nothing - 14.5
15.5 - nothing - 14.75
16.0 - nothing - 15.0
16.0 - cooling - 14.5
15.5 - cooling - 13.5
15.5 - nothing - 13.75

Project Structure

ai-ml_discovery/
└── module0/
    ├── environment.py
    └── expert_system.py

Learning Objectives

• Understanding basic AI concepts
• Implementing state machines
• Working with rule-based systems
• Creating simulations with Python

Implementation Details

Environment Simulator

• Temperature is modeled as a sine wave with season-specific parameters
• The sine wave is shifted so that temperatures are lowest around 5 AM
• The amplitude is calculated based on seasonal min/max temperatures
• Random variations are added for more realistic simulation

Expert System

• Uses simple threshold-based rules to determine system state
• Implements gradual heating/cooling effects
• Models temperature inertia when systems are inactive
• Outputs the current state for monitoring purposes