A Reinforcement Learning Modeling Method for Demand Response in Building Air Conditioning Systems

Abstract

The invention discloses a reinforcement learning modeling method for building air-conditioning system demand response, which comprises the following steps: using the known indoor temperature, cooling capacity, weather, and personnel data of the building to develop an RC gray box model to establish indoor temperature and air-conditioning system refrigeration relationship between quantities; establish an enhanced model based on value function linear approximation for air-conditioning system and battery control; use meteorological and personnel data to train the agent of the reinforcement learning model; apply the fully trained agent to the target time period, and obtain Control strategy for air conditioning system and battery. The invention can reflect the thermal characteristics of buildings and reduce the need for reference data and prior knowledge. Under the premise of ensuring the thermal comfort of indoor occupants, an air-conditioning system and battery operation strategy with low energy consumption and low electricity cost can be proposed.

Description

technical field [0001] The invention belongs to the intersection field of building energy management and artificial intelligence, and in particular relates to a reinforcement learning modeling method for demand response of a building air-conditioning system. Background technique [0002] Building operation energy consumption is an important aspect of my country's energy consumption, and in building operation, air conditioning energy consumption accounts for a large proportion. However, due to the delay and attenuation of the response of the building system to the external weather conditions, this adds complexity to the control of the air conditioning system. As a result, the air conditioner operation strategy is mostly formulated based on the experience of the operator, that is, the operator adjusts the air conditioner operation strategy according to the current meteorological conditions, weather forecast, past experience, operating economy and other factors. There are only...

AI Technical Summary

Problems solved by technology

There are only subjective judgments on personnel comfort and energy saving status, which cannot guarantee the comfort of indoor personnel and reduce energy consumption

[0003] At present, there are many methods for the automatic control of building air-conditioning systems. When adopting feedback control, due to its delay characteristics, it is impossible to control the indoor temperature in time and cannot guarantee that the system is running in an efficient state; when adopting supervised machine learning algorithms for control , often requires reference data and prior knowledge, and the existing conditions in actual operation are difficult to meet the algorithm requirements

At the same time, most of the existing control methods only control the air conditioning system and do not involve battery control, which cannot further achieve the purpose of power grid load transfer

Images