Information processing device and air conditioning system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing air conditioning systems fail to account for multiple users in a space, leading to reduced comfort and inefficiency when different users are present, and do not adapt quickly to changes in occupancy or user preferences.
Innovation Solution
An information processing device and air conditioning system that utilize machine learning and reinforcement learning to adjust temperature and humidity settings based on user comfort data from personal terminals, classifying users into comfort classes and optimizing control for energy efficiency and comfort using reinforcement learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If air conditioning control is based on environment parameters only, then the system is simple to operate, but user comfort is significantly reduced when users have different preferences
Solution Approach 1:
The system segments users into different comfort classes (first comfort class, second comfort class, etc.) based on their temperature preferences. By classifying users into distinct groups, the system can apply different control strategies to different segments, improving comfort accuracy without requiring complex individualized control for each user.
Solution Approach 2:
The system automatically determines user comfort classes and adjusts air conditioning control without requiring manual user input or complex interactions. Users simply need to provide basic temperature preference information, and the system self-manages the classification and control adjustment, reducing operational complexity while improving comfort measurement.
2Adaptability or versatility
If air conditioning control does not consider multiple users, then the control logic is simple, but comfort cannot be guaranteed when a plurality of users are present in the same room
Solution Approach 1:
The system divides multiple users into distinct comfort classes based on their temperature preferences. When multiple users are present, the system identifies which comfort class each user belongs to and applies appropriate control strategies for each segment, enabling multi-user adaptability through systematic classification rather than complex individual management.
Solution Approach 2:
The air conditioning control system is designed to handle multiple scenarios universally - it can manage single users, multiple users with similar preferences, and multiple users with different preferences by using the same comfort class classification framework. This universal approach allows the system to adapt to various user compositions without requiring fundamentally different control logic for each scenario.
3Adaptability or versatility
If the system does not adapt to user movement from outside, then the system operates stably, but comfort is significantly reduced immediately after a person moves from the outside
Solution Approach 1:
The system performs preliminary classification of users into comfort classes based on their movement status (indoor vs. outdoor origin). When a user moves from outside, the system proactively applies pre-determined control adjustments appropriate for that comfort class, anticipating the comfort needs before the user explicitly requests changes. This preliminary action maintains stability by using pre-established classification rules while improving adaptability to occupancy changes.
4Extent of automation
If traditional air conditioning control is used without machine learning, then the system is easier to manufacture and implement, but the system cannot automatically adjust comfort to suit different users or learn from user behavior
Solution Approach 1:
The system employs machine learning models that automatically learn user comfort preferences and behavior patterns without requiring manual programming or complex configuration. The learning models self-adjust control parameters based on collected user data, achieving automatic comfort adjustment while keeping the system architecture relatively simple by leveraging automated learning rather than manual rule-setting.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
Each of plurality of personal terminals (200) is configured to acquire first data indicating a result of inputting whether a possessor is comfortable, second data indicating a terminal location, and third data indicating a temperature at the terminal location. An information processing device (100) includes a first learning unit (102) to classify the plurality of personal terminals (200) into a plurality of classes based on the first to third data transmitted from the plurality of personal terminals (200), a storage unit (104) to store a plurality of control details each associated with a corresponding one of the plurality of classes into which the first learning unit (102) classifies the plurality of personal terminals (200), and a control unit (110) to read, from the storage unit (104), a control detail associated with a class into which a personal terminal (200) detected in an air conditioning target space is classified among the plurality of classes and control an air conditioning device.