Air conditioner and method of controlling the same

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional home appliances, such as air conditioners, cannot provide operations that satisfy all occupants in a home as they only determine user entry and exit without considering the number of occupants, leading to inadequate control and comfort.

Innovation Solution

An air conditioner system that uses a neural network to learn and adjust operations based on occupant information, including the number of occupants, by integrating sensors for environmental data and user inputs, and communicating with external servers to optimize temperature and mode settings.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional home appliances determine only user entry and exit, then the operation control is simple, but the operation cannot satisfy all occupants

Engineering Contradiction:
Improveoperation satisfaction for occupantsVSAvoidoccupant detection system
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The air conditioner system is designed to serve multiple occupants with different preferences by integrating multiple detection methods (access point connection detection, sensor-based presence detection) and learning capabilities. The system universally adapts to different users' temperature preferences through the neural network that learns from historical operation data of multiple occupants, enabling one system to satisfy diverse occupant needs rather than requiring separate control systems for each person.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system performs automatic learning and adaptation without requiring manual configuration for each occupant. The neural network automatically learns temperature preferences and occupancy patterns from historical operation data and sensor inputs, enabling the air conditioner to self-adjust operations to satisfy different occupants based on their implicit preferences rather than requiring explicit programming or manual input from each user.

Inventive Principle:
Principle #25Self-service

2Adaptability or versatility

If the air conditioner uses neural network learning based on setting information, then the operation can satisfy all occupants, but the system complexity increases

Engineering Contradiction:
Improveoperation adaptation to occupantsVSAvoidneural network system
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The neural network functions as an intermediary layer between the physical sensors (access point detectors, temperature sensors, occupancy sensors) and the air conditioner control system. It processes raw sensor data and historical operation information, transforming them into meaningful occupancy patterns and temperature preferences, which then guide the control decisions. This intermediary learning system enables complex adaptive behavior without requiring direct complex control logic in the hardware.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary learning during periods when occupants are not actively controlling the air conditioner, accumulating historical operation data and sensor information to build occupancy profiles and temperature preferences. This preliminary learning action enables the system to be well-prepared and immediately adaptive when occupants enter or when control decisions are needed, reducing the complexity of real-time decision-making.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If the system collects real-time occupant data and environmental conditions, then the control precision improves, but the information processing load increases

Engineering Contradiction:
Improveoccupant and environment detectionVSAvoiddata processing capacity
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The system extracts only the most relevant features from the collected data for neural network processing, such as occupancy presence/absence, temperature preferences, and basic environmental conditions, rather than processing all raw sensor data in detail. By extracting key informative features and discarding redundant information, the system maintains high measurement precision for control decisions while reducing the information processing load and preventing information overload.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3722690B1Air conditioner and method of controlling the same
Publication Date: 2022.06.01 SAMSUNG ELECTRONICS CO LTD
  • EP3722690B1 patent drawingFigure 1
  • EP3722690B1 patent drawingFigure 2
  • EP3722690B1 patent drawingFigure 3

AI summary

Provided is a home appliance of determining an operation command corresponding to an occupant through learning based on setting information of the home appliance according to the occupant to provide an operation satisfying all occupants. An air conditioner according to an embodiment of the disclosure includes: an outdoor unit; and an indoor unit including a heat exchanger, wherein the indoor unit includes: a communicator configured to communicate with an access point (AP); and a controller configured to receive information about a terminal connected to the access point through the communicator, and change at least one of operation temperature or an operation mode when a new terminal is connected to the access point or a terminal connected to the access point is disconnected from the access point.