Optimization Method for Energy Saving of Rural Architectures in Hot Summer and Cold Winter Areas Based on Artificial Neural Network

Abstract

With the phased spatial planning of the rural revitalization strategy, the proportion of architecture energy consumption in the overall social energy consumption is also increasing year by year. Considering the hot summer and cold winter areas, the proportion of architecture energy consumption in the total energy consumption is very large. The ecological environment and natural resources have been greatly threatened, and the issue of energy conservation and environmental protection is imminent. Energy consumption prediction and analysis is an important branch of building energy conservation in the field of building technology and science. Aiming at the energy consumption characteristics of rural architectures in areas with hot summer and cold winter, this paper proposes a method for constructing a neural network model. When building a neural network, the dataset is called and the function is applied randomly to training samples. The data are used for simulation tests to analyze the fit between the predicted results and the calculated results. Flexible forecasting of specific target building energy consumption is achieved, which can provide optimization strategies for updating and adjusting architecture energy efficiency design. The experimental analysis benchmark parameters and the output value in the dataset are compared with the target simulation value. The relative error is less than 4%, and the average relative error value (mean) and the root mean square error (RMSE) value are both controlled within 2%. It is proved that the method in this paper can directly reflect the evaluation of energy consumption by the neural network and realize the high-speed conversion of the generalized model to the concrete goal, which has a certain value and research significance.

Building Space Thermal Control Model Responding to Sharp Changes in Outdoor Temperature

“…The input layers are set up as E and ΔE from the difference between T set and T rm , and they are trained to find better control outputs to meet the appropriate value of the PMV within the initial setting range (-0.5 < a <0.5). For its simulation configuration, a scale conjugate gradient algorithm is used, and the repetition of simulations is executed up to 1000 times to get valid regression results [34,35]. From the simulation results for the statistical significance, the R 2 values were confirmed as 0.99034 for controlling the amount of supply air mass and 0.98610 for controlling its temperature, respectively.…”