Machine Learning CO2 Prediction for Proactive Ventilation Control

Solution Overview

Problem

Current ventilation control systems rely on carbon dioxide sensors to measure CO2 levels and adjust ventilation, but they struggle to accurately predict CO2 concentrations in real-time, especially when the number of people in a room changes, leading to delayed detection and inefficient ventilation.

Innovation Solution

A machine learning device that acquires environmental and number-of-people information to create a carbon dioxide concentration estimation model, predicting CO2 levels in rooms based on learned associations, allowing for proactive ventilation control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If carbon dioxide sensors are used to measure CO2 levels and control ventilation, then ventilation control can be implemented, but accurate real-time prediction of CO2 concentrations cannot be achieved, especially when the number of people changes

Engineering Contradiction:

ImproveCO2 concentration prediction accuracyVSAvoiddelayed detection

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary learning by collecting and storing the relationship between number of people, environmental information, and actual CO2 concentrations in advance. This pre-acquired knowledge enables the system to predict CO2 concentrations accurately without requiring real-time sensor measurements, thus resolving the contradiction between prediction accuracy and detection delay.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system introduces an intermediary prediction model that uses number of people and environmental information as intermediate variables to estimate CO2 concentrations. Instead of directly measuring CO2 levels with sensors, the system uses these intermediary factors that are easier to obtain and correlate with CO2 levels, achieving accurate prediction without the limitations of direct sensor measurement.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If ventilation control is based on actual CO2 sensor measurements, then CO2 levels can be monitored, but the system responds too late when the number of people changes rapidly

Engineering Contradiction:

Improveventilation control reliabilityVSAvoidresponse speed

Core Design Contradiction:

ReliabilityVSSpeed

Solution Approach 1:

The system performs preliminary learning by collecting and storing the relationship between number of people, environmental information, and actual CO2 concentrations in advance. This pre-acquired knowledge enables the system to predict CO2 concentrations accurately without requiring real-time sensor measurements, thus resolving the contradiction between prediction accuracy and detection delay.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system dynamically adapts to changes in the number of people by using the learned model to predict CO2 concentrations in real-time based on current occupancy and environmental conditions. This dynamic prediction approach allows the ventilation control to respond immediately to occupancy changes, maintaining reliability while achieving fast response speed.

Inventive Principle:

Principle #15Dynamics

3Productivity

If traditional CO2 sensor-based control is used, then ventilation can be adjusted, but energy consumption is not optimized due to delayed detection

Engineering Contradiction:

Improveventilation efficiencyVSAvoidenergy consumption

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The system performs preliminary learning by collecting and storing the relationship between number of people, environmental information, and actual CO2 concentrations in advance. This pre-acquired knowledge enables the system to predict CO2 concentrations accurately without requiring real-time sensor measurements, thus resolving the contradiction between prediction accuracy and detection delay.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system replaces the mechanical CO2 sensor measurement system with an information-based prediction system that uses number of people and environmental information. This substitution eliminates the need for continuous sensor measurements and complex real-time calculations, improving ventilation efficiency while reducing energy consumption through simpler, faster predictions.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20240011658A1Machine learning device, ventilation control device, and ventilation control method

Publication Date: 2024.01.11 DAIKIN INDUSTRIES LTD

AI summary

A machine learning device includes a first acquisition unit, a second acquisition unit, and a learning unit. The first acquisition unit acquires environmental information on a target space. The second acquisition unit acquires number-of-people information indicating a number of people in the target space. The learning unit learns the environmental information acquired by the first acquisition unit and the number-of-people information acquired by the second acquisition unit in association with each other. The environmental information includes an actual carbon dioxide concentration in the target space.