Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and system for building and predicting air conditioner load prediction model in office building

An air-conditioning load and forecasting model technology, which is applied in neural learning methods, biological neural network models, information technology support systems, etc., can solve problems such as insufficient forecasting accuracy, long cycle, and large resource occupation, so as to achieve accurate forecasting results and reduce Accuracy requirements, high universality effect

Pending Publication Date: 2021-07-13
XI'AN UNIVERSITY OF ARCHITECTURE AND TECHNOLOGY
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide an air conditioning load forecasting model establishment, forecasting method and system in an office building to solve the problem of long cycle, difficult forecasting, large resource occupation and forecasting accuracy for a certain area load forecasting in the prior art not enough problem

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and system for building and predicting air conditioner load prediction model in office building
  • Method and system for building and predicting air conditioner load prediction model in office building
  • Method and system for building and predicting air conditioner load prediction model in office building

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0076] This embodiment discloses a method for establishing an air-conditioning load forecasting model in an office building. The training data set includes more than 14,000 sets of data for training a single air-conditioning load forecasting model. More than 7,000 sets of data are used to test the model and train the model. The set and test set include environmental data and actual air-conditioning load data for each hour between 8:00 am and 20:00 pm every day for a month, and the data set is input into the prediction network to train the model. According to the established RBF model and the combined residual correction model, the air-conditioning load of 12 hours during the daytime on a certain day in summer is predicted. The actual value, absolute error and average relative error predicted by various methods are as follows: Figure 5 , Figure 6 , Figure 7 shown

[0077] According to the simulation results of air-conditioning load forecasting by different methods, the av...

Embodiment 2

[0085] This embodiment discloses an air-conditioning load forecasting system in an office building. On the basis of the above embodiments, the following technical features are also disclosed:

[0086] The indoor and outdoor temperature and humidity sensor selects the digital temperature and humidity sensor model HTU21D(F). It is a plug-and-play temperature and humidity measurement component. The sensor is matched by the OEM to make the measurement more reliable and accurate. It directly uses an MCU without Other peripheral circuits can output temperature and humidity as digital signals. Its temperature measurement range is: -40~+125℃, and the humidity measurement range is: 0~100%RH. Because this invention is mainly aimed at load forecasting for office buildings in cold regions, the measurement range of the sensor needs to be considered. In some areas in northern my country, the lowest temperature in winter can reach minus 30 to 40 degrees, so it is necessary to select a sensor...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the field of smart city and building energy saving, and discloses a method and system for building and predicting an air conditioner load prediction model in an office building. The method comprises the following steps: 1, collecting historical environment data and measuring daily load data; 2, establishing an RBFNN and a residual network, wherein the residual network comprises a plurality of sub-networks, the historical environment data serve as a training set, the measured daily load data serve as a label set, the RBFNN and the residual network are trained respectively, and the trained RBFNN serves as an air conditioner load prediction model. For different time periods, according to the relative prediction errors of various single prediction models, two models with the minimum relative prediction errors are dynamically selected to form a combined residual error to correct the prediction errors of a basic method, and the prediction method for dynamic combined residual error correction is used for central air conditioner load prediction; and the prediction precision is further improved.

Description

technical field [0001] The invention belongs to the field of smart cities and building energy conservation, and in particular relates to the establishment of an air-conditioning load forecasting model in an office building, a forecasting method and a system. Background technique [0002] China's energy consumption has shown a steady increase, among which building energy consumption accounts for about one-third of the country's total energy consumption, and public building energy consumption is the main component of building energy consumption, so the research on energy-saving design and load forecasting for public buildings has become a It is one of the main problems in the field of intelligent buildings, and building load forecasting is the primary problem to be solved to realize energy saving of building equipment. The prior art lacks a system that can flexibly predict any area of ​​a building. This is because the existing data used for research is usually collected manua...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F30/13G06F30/27G06N3/04G06N3/08G06F119/08
CPCG06F30/13G06F30/27G06N3/084G06F2119/08G06N3/045Y04S10/50
Inventor 冯增喜崔巍何鑫张茂强杨芸芸赵锦彤李诗妍陈海越
Owner XI'AN UNIVERSITY OF ARCHITECTURE AND TECHNOLOGY