Indoor environment state prediction method based on BIM and cross sample learning

Abstract

The invention provides an indoor environment state prediction method based on BIM and cross sample learning, and the method comprises the steps that a space diagram model is built through extracting the space geometric data of a BIM model, cross sample learning is designed, on the space diagram model, node weights (clue factors) and edge weights of known points are adaptively fused to associate spatial features of indoor environment states, the state of the whole indoor space is predicted, a distribution rule of humidity values of the whole indoor space of a building is simulated through an ML-IDW algorithm, the state of the whole indoor space is predicted through a small number of collection nodes, a training model is constructed according to actually measured data, the accuracy of the model is verified by comparing with actual data, the humidity distribution of the whole indoor space in an experiment scene is predicted, the training time is short, and real-time performance is achieved.

Description

technical field [0001] The invention relates to the field of indoor environment state prediction based on BIM and cross-sample learning, in particular, to a method for predicting indoor environment state based on BIM and cross-sample learning. Background technique [0002] Indoor architecture has become the main space for human activities, and indoor environmental issues have gradually become a research hotspot, with a lot of research, and indoor architectural personalized service is an important research content and trend of intelligent buildings and smart cities; a key prerequisite for building indoor personalized service It is the overall perception of the interior of the building. [0003] Taking indoor humidity as an example, humidity is not only closely related to many health problems, but also has a great impact on building energy consumption and durability. A large number of studies have proved that low humidity environments can cause sensory irritations such as dry ...

AI Technical Summary

Problems solved by technology

[0003] Taking indoor humidity as an example, humidity is not only closely related to many health problems, but also has a great impact on building energy consumption and durability. A large number of studies have proved that low humidity environments can cause sensory irritations such as dry eyes, dry skin, and upper respiratory tract. In order to To ensure human health, the relative humidity inside the building must be kept above 10%-30%. High humidity environment will induce mold growth, affect the health of the upper respiratory tract, and increase the risk of mite allergies. Building damp and mold are related to various breathing and asthma A 30%-50% increase in related health outcomes correlates and points to the need to prevent moisture build-up, mold contamination from excessive indoor humidity often leading to very serious remedial measures, deterioration of building materials and loss of energy consumption

[0004] However, indoor sensors in buildings are usually installed at specific locations, such as the walls near the door, and the measured values ​​cannot reflect the real value of the indoor working area. Requirements, influence of indoor environmental parameters such as human activities, equipment, plants, building indoor environment and outdoor environment, etc.

Moreover, the smart sensor nodes are limited, the deployment location is specific, the sensing range is limited, and the indoor environment factors are complex, so it is impossible to realize the monitoring of the whole space state. Under the interference of various factors, it is quite difficult to predict the state of the indoor environment in real time.

Images