A grey neural simulation method for civil buildings with non-artificial cold and heat sources at room temperature is disclosed

Abstract

A grey neural simulation method for civil buildings with non-artificial cold and heat sources at room temperature includes: step 1, collecting and grouping the measured data, step 2, constructing theindoor temperature OGM (1, N) simulation model, step 3, constructing the indoor temperature BP neural network simulation model, and step 4, simulating the indoor temperature OGMBPT and outputting theresult. The invention adopts the combination of the optimized multivariable grey OGM (1, N) simulation model and the BP neural network, groups the measured data and constructs the room temperature OGM(1, N) simulation model respectively, so as to achieve the effect of reducing the influence of the data disturbance on the simulation result. The measured data and the OGM (1, N) simulation results are input into the BP neural network at room temperature to construct the secondary simulation model, and the advantages are complementary, so as to improve the overall simulation accuracy, and the limited measured data are fully utilized to maximize the room temperature law contained in the data.

Description

technical field [0001] The invention belongs to the field of building heat and humidity environment simulation, and in particular relates to a simulation method for gray prediction of indoor temperature of civil buildings with non-artificial cold and heat sources and a neural network. Background technique [0002] Non-artificial cold and heat sources refer to rooms or areas that do not use artificial cold and heat sources, but only use natural adjustment or mechanical ventilation to regulate the heat and humidity environment. The Standard for Evaluation of Indoor Heat and Humidity Environment in Civil Buildings (GBT 50785-2012) provides two methods, calculation method and graphical method, to evaluate the indoor thermal and humid environment of non-artificial cold and heat sources. It is an important indoor heat and humidity environment parameter and an important basis for evaluation. According to the "Civil Building Indoor Thermal and Humid Environment Evaluation Standard"...

AI Technical Summary

Problems solved by technology

After decades of research, the gray forecasting method has also been continuously improved and improved. The univariate gray forecasting model GM(1,1) is widely used and simple in modeling, but this model does not consider the influence of relevant factors on the development of the system.

The multivariate gray prediction model is represented by GM(1, N), and its modeling process fully considers the influence of relevant factors on the system. However, the traditional GM(1, N) model has unstable performance and strict requirements on the range of data changes. insufficient

Generally speaking, the gray prediction method is simple in modeling and requires a small amount of data. However, if the array contains abnormalities, mutations, or large disturbances, it will have a great adverse effect on the prediction results, resulting in large prediction errors.

Images