Cold-Region Pedestrian Layout Using Dynamic Thermal Comfort Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current thermal comfort prediction methods for cold-region pedestrian spaces inadequately account for dynamic environmental influences and subjective perceptions, leading to low accuracy and inefficiency in design decisions.
Innovation Solution
A method for generative design of cold-region urban pedestrian space layout based on dynamic thermal comfort prediction, involving the construction of a wayfinding agent model, mapping thermal environment data to pedestrian comfort data, and optimizing layout design using reinforcement learning and human-computer interaction for improved accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If steady-state thermal comfort models (single-node, two-node, multi-node) are used for prediction, then the prediction process is simple and widely applicable, but the prediction accuracy is insufficient because these models do not account for thermal history, thermal expectations, and dynamic environmental influences
Solution Approach 1:
The patent applies dynamics by transitioning from steady-state thermal comfort models to dynamic thermal comfort models that account for thermal history, thermal expectations, and continuous environmental influences. The model incorporates time-varying parameters and dynamic responses to accurately predict thermal comfort in non-steady-state environments, resolving the contradiction between prediction accuracy and model complexity.
Solution Approach 2:
The patent changes the parameters used in thermal comfort prediction from static inputs (air temperature, humidity, air velocity) to dynamic inputs that include thermal history, thermal expectations, and temporal variations. This parameter transformation enables the model to capture dynamic thermal comfort characteristics while maintaining computational feasibility.
2Measurement precision
If dynamic thermal comfort models accounting for thermal history and thermal expectations are developed, then prediction accuracy improves, but the computational time and complexity increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing thermal comfort data for various environmental conditions and temporal patterns before actual prediction is needed. This pre-processing enables faster prediction during actual use by retrieving pre-computed results rather than performing complex real-time calculations, thus reducing computational time while maintaining high prediction accuracy.
Solution Approach 2:
The patent uses copying by creating simplified representations or proxies of complex thermal comfort simulations. Instead of performing full dynamic thermal comfort simulations for every prediction case, the system uses pre-generated lookup tables, machine learning models trained on simulation data, or simplified analytical models that replicate key thermal comfort characteristics with reduced computational effort.
3Adaptability or versatility
If existing thermal comfort models are used without considering thermal pleasure and subjective perceptions, then the models remain simple and objective, but they fail to capture the dynamic and individualized nature of thermal sensation in real-world environments
Solution Approach 1:
The patent applies segmentation by dividing the thermal comfort prediction system into multiple independent modules: environmental parameter measurement, thermal history tracking, thermal expectation modeling, and subjective perception integration. Each module handles a specific aspect of thermal comfort prediction, making the overall system more adaptable to real-world conditions while keeping individual components manageable and easier to implement.
Data Source
AI summary
A method for generative design of a cold-region urban pedestrian space layout based on dynamic thermal comfort prediction is provided. The method includes: constructing a wayfinding agent model in a cold-region pedestrian space based on big data and IoT data to obtain pedestrian trajectories during a plurality of travel periods in the cold-region pedestrian space; constructing a mapping between thermal environment data of the cold-region pedestrian space and pedestrian thermal sensation based on physiological indicator data of pedestrians to obtain pedestrian thermal sensation under different thermal environment changes; optimizing the cold-region pedestrian space layout design based on prediction results of pedestrian thermal sensation under typical pedestrian trajectories during the plurality of travel periods; and determining a cold-region pedestrian space layout decision model guided by preferences of the designers based on machine learning models and decision feedback of designers on visual solutions.