Real-time Occupancy Prediction in Large Exhibition Halls
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods are ineffective in predicting the number of occupants in large exhibition halls due to the lack of physical division and significant changes in occupancy, which hampers energy efficiency optimization for heating and cooling.
Innovation Solution
A deep learning-based method that divides the space into zones, preprocesses data using image sensors and counting devices, generates time-series data, trains a model using LSTM, and predicts occupancy through real-time input data via socket communication, allowing for accurate and real-time prediction of occupant numbers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the HMM-based prediction method is used, then prediction can be performed in physically divided spaces with stable occupancy, but it becomes impossible to predict occupants in large exhibition halls where the number of occupants changes greatly and space is not physically divided
Solution Approach 1:
The patent divides the large exhibition hall into multiple virtual zones without physical partitions. Each zone is assigned an occupancy state, allowing the system to track and predict occupancy changes in each segment independently while maintaining overall system coherence. This segmentation enables the HMM to handle large-scale spaces that would otherwise be impossible to monitor effectively.
Solution Approach 2:
The patent transitions from static physical zone divisions to dynamic virtual zone assignments. Zones are defined by occupancy patterns and can change over time, allowing the system to adapt to varying occupancy distributions. This dynamic approach enables reliable prediction in spaces where physical division is not feasible, resolving the contradiction between maintaining prediction reliability and adapting to different space configurations.
2Use of energy by stationary object
If real-time occupancy prediction is implemented in large exhibition halls, then energy efficiency for heating and cooling can be optimized, but existing methods fail due to the lack of physical division and significant occupancy changes
Solution Approach 1:
By segmenting the large exhibition hall into virtual zones, the system enables zone-specific occupancy prediction and control. This allows heating and cooling systems to be optimized for each zone based on actual occupancy, improving overall energy efficiency while maintaining prediction reliability through the mathematical framework of HMM applied to each segment.
Solution Approach 2:
The patent changes the fundamental parameters of the prediction system by transitioning from physical to virtual zone definitions and from static to dynamic occupancy modeling. This parameter transformation enables the application of HMM to large exhibition halls, making reliable prediction feasible and thereby enabling energy efficiency optimization that was previously unattainable in such spaces.
Data Source
AI summary
Disclosed are a method and apparatus for predicting a change in the occupants within a large exhibition hall in real time based on deep learning. A proposed method of predicting a change in the number of occupants within a space in real time includes dividing, into zones, a space where a number of occupants is to be predicted and pre-processing data related to a number of occupants within the space collected through simulations, generating the pre-processed data in a form of time-series data for deep learning, training a deep learning model for predicting a number of occupants in each divided zone using the generated time-series data, and predicting the number of occupants within the space by inputting, to the trained model, the data related to a number of occupants within the space collected in real time through socket communication with a server.