Building energy consumption prediction system and method based on edge cloud collaborative hybrid modeling strategy

Abstract

The invention discloses a building energy consumption prediction system and method based on an edge cloud collaborative hybrid modeling strategy. The system comprises edge Internet of Things nodes arranged on various energy consumption units of a building. the edge Internet of Things nodes are in network connection with a nearby edge controller; wherein the edge controller is in network connectionwith an edge gateway deployed with a special APP, the edge gateway is in network connection with a cloud platform deployed with a hybrid modeling predictor, and the cloud platform is in network connection with a WEB-based webpage platform logged in through a personal computer and an application APP installed in an intelligent mobile terminal. According to the invention, real-time energy consumption prediction based on edge calculation is adopted; a basis is provided for controlling various energy consumption devices in the building; the cloud platform-based hybrid model energy consumption prediction is adopted, services are provided for comprehensive energy scheduling, energy consumption analysis and energy saving strategies of buildings, the method has the advantages of intelligence, high efficiency and low cost, the effectiveness, accuracy, comprehensiveness and reliability of building energy consumption prediction are improved, and a better building energy saving effect is achieved.

Description

technical field [0001] The invention belongs to the technical field of building energy consumption prediction and energy saving, and relates to a building energy consumption prediction technology, in particular to a building energy consumption prediction system and method based on an edge-cloud collaborative hybrid modeling strategy, which is applied to the building intelligence industry. Background technique [0002] Nowadays, the proportion of building energy consumption in the total energy consumption of society is increasing. By the end of 2017, the energy consumption of buildings in the world accounted for 40% of the total energy consumption of the whole society, 1.5 times that of industrial energy consumption, and more than two-thirds of the total energy consumption of buildings was used in space heating (37%), Hot water (12%), space cooling (10%) and lighting (9%). Therefore, in the context of the current building energy conservation work requiring refined management...

AI Technical Summary

Problems solved by technology

Multiple linear regression does not require parameter adjustment, and is very effective for long-term energy consumption prediction, but it cannot solve nonlinear problems; artificial neural network (ANN) has high accuracy for short-term energy consumption prediction, but this method has nothing to do with building physical parameters, and has no impact on building units and The adaptability of the system is poor; Support Vector Machine (SVM) can achieve a good balance between prediction accuracy and calculation speed, but it is not easy to determine the kernel function used by the algorithm

Images