CN-121977283-A - Combined optimization control method for air system and water system of ventilation air-conditioning system

Abstract

The invention discloses a combined optimization control method of a ventilation air-conditioning system air system and a water system, which takes key parameters of the air system, the water system and the environment as states, takes an adjustable control parameter set of the air system and the water system as actions, trains a reinforcement learning model based on a depth certainty strategy gradient DDPG and a preset rewarding function, deploys the trained reinforcement learning model to an actual ventilation air-conditioning system, acquires system running state data in real time and further outputs optimal control actions to realize combined optimization control of the air system and the water system.

Inventors

• QIAO KUANGHUA
• GU SIYUAN
• Jing Chenying
• HAN QIANG

Assignees

• 西安思安云创科技有限公司

Dates

Publication Date

20260505

Application Date

20260129

Claims (10)

1. 1. A combined optimization control method for a ventilation air-conditioning system wind system and a water system is characterized by taking key parameters of the wind system, the water system and the environment as states, taking an adjustable control parameter set of the wind system and the water system as actions, training a reinforcement learning model based on a depth deterministic strategy gradient DDPG and a preset reward function; The reinforcement learning model is composed of a current Actor network, a target Actor network, a current Critic network and a target Critic network.
2. 2. The method for combined optimization control of a ventilation air conditioning system and a water system according to claim 1, wherein key parameters of the ventilation air conditioning system are blower air quantity, blower frequency, return air machine air quantity, return air machine frequency, fresh air blower air quantity, fresh air blower frequency, air supply temperature and return air temperature.
3. 3. The method for combined optimization control of a ventilation air conditioning system and a water system according to claim 1, wherein key parameters of the water system are chilled water supply and return water temperature difference, chilled water supply pressure, chilled water pump frequency, cooling tower fan frequency, water chilling unit evaporation temperature, water chilling unit condensation temperature and water chilling unit COP.
4. 4. The method for combined optimal control of a ventilation and air conditioning system and a water system according to claim 1, wherein the key parameters of the environment are indoor average temperature, indoor average relative humidity, outdoor temperature, outdoor relative humidity and solar radiation intensity.
5. 5. The method for combined optimization control of a ventilation air conditioning system and a water system according to claim 1, wherein the adjustable control parameter set of the ventilation air conditioning system is a blower frequency adjustment amount, a return air machine frequency adjustment amount and a fresh air blower frequency adjustment amount.
6. 6. The method for combined optimization control of a ventilation air conditioning system and a water system according to claim 1, wherein the adjustable control parameter set of the water system is a chilled water pump frequency adjustment amount, a cooling tower fan frequency adjustment amount and a chiller refrigerating capacity adjustment amount.
7. 7. The method for combined optimization control of a ventilation air conditioning system and a water system according to claim 1, wherein the preset reward function R is: R = ω e R e + ω c R c + ω s R s Wherein R e 、R c 、R s is energy consumption rewarding, comfortable rewarding and safe rewarding respectively, omega e 、ω c 、ω s is weight coefficient of energy consumption rewarding, comfortable rewarding and safe rewarding respectively, and the sum is 1.
8. 8. The method for the joint optimization control of a ventilation air conditioning system and a water system according to claim 7, wherein R e = -k e ×(E wi + E wa )/E sum ; wherein k e is the energy consumption rewarding coefficient, E wi 、E wa 、E sum is the energy consumption of the wind system, the energy consumption of the water system and the total energy consumption under the design working condition of the system respectively.
9. 9. The method for joint optimization control of a ventilation air conditioning system and a water system according to claim 7, wherein the comfort prize R c is: R c = k c is R when the PMV is less than or equal to 0.5; when 0.5< |pmv| <1.0, R c =0; R c = -k c when the I PMV I is more than or equal to 1.0; Wherein k c is a comfortable rewarding coefficient, and the PMV value is calculated according to the indoor temperature, the relative humidity, the wind speed, the clothing thermal resistance and the human metabolism rate; PMV=[0.303exp(−0.036M)+0.028]× , wherein M is the energy metabolism rate of the human body, and 110 is usually taken; Is a human body heat load; ; Wherein, the Is a cluster coefficient; 、 Sensible heat and latent heat dissipation from adult men, ; Number of persons.
10. 10. The method for joint optimization control of a ventilation air conditioning system and a water system according to claim 7, wherein the safety prize R s is: R s = k s when all the equipment operation parameters are within the safety threshold range; R s = -k s x n when any equipment operating parameter exceeds a safety threshold range; wherein k s is a safety rewarding coefficient, and n is the number of parameters exceeding a safety threshold.

Description

Combined optimization control method for air system and water system of ventilation air-conditioning system Technical Field The invention belongs to the field of joint optimization control, and particularly relates to a joint optimization control method for a ventilation air-conditioning system air system and a water system. Background The ventilation air conditioning system is a core component of building energy consumption, the energy consumption accounts for 40% -60% of the total energy consumption of the building, and the energy consumption of a wind system (comprising a wind feeding/returning fan, a fresh air fan and the like) and a water system (comprising a water chilling unit, a cooling water pump, a chilled water pump, a cooling tower and the like) accounts for more than 80% of the total energy consumption of the ventilation air conditioning system. Therefore, the realization of the efficient cooperative control of the wind system and the water system is a key for reducing the energy consumption of the building and improving the running economy of the system. The control mode of the existing ventilation air-conditioning system mostly adopts a traditional independent control strategy, namely the wind system and the water system respectively operate according to a preset threshold value or simple PID control logic, and cooperative linkage between the wind system and the water system is lacked. For example, the chilled water pump performs variable frequency adjustment according to the chilled water supply and return water temperature difference, and the blower adjusts the air supply amount according to the indoor temperature deviation, and the control targets of the chilled water pump and the blower are independent of each other, and the coupling influence of the running states of the chilled water pump and the blower is not considered. This independent control approach suffers from the following significant drawbacks: 1) The running efficiency is low, because the wind system and the water system have strong coupling relation (if the change of the air supply quantity can influence the indoor load demand, and further influence the refrigerating capacity demand of the water chilling unit and the refrigerating water flow regulation), the independent control can easily cause the system to run in a local optimal state instead of global optimal; 2) The adaptability is poor, the traditional control strategy depends on preset control parameters and fixed control logic, and the traditional control strategy is difficult to adapt to dynamic changes of building loads (such as personnel flow, outdoor temperature and humidity fluctuation, solar radiation change and the like) and ageing attenuation of system equipment. When the operation working condition deviates from the preset range, the control effect is rapidly reduced, and the balance of indoor thermal comfort requirement and energy consumption optimization cannot be ensured; 3) The control precision is insufficient, the problems of overshoot, oscillation and the like easily occur in the traditional PID control, particularly in a multivariable and strong-coupling ventilation air conditioning system, the accurate regulation and control of key parameters of a wind system and a water system are difficult to realize, and the indoor temperature and humidity deviate from set values possibly to influence the comfort level of personnel. In order to solve the above problems, some attempts of wind and water system combined control have been made in the prior art, and mainly model-based control methods, such as Model Predictive Control (MPC), are adopted. However, such methods have the following limitations: on one hand, the ventilation air-conditioning system is a complex nonlinear system, a great deal of priori knowledge and complicated parameter identification work are required for establishing an accurate mathematical model, and the precision of the model is difficult to ensure; on the other hand, most of optimization targets of model predictive control are single energy consumption and lowest, balance of multiple targets such as indoor thermal comfort, equipment operation life and the like is not fully considered, the optimization solving process is complex, the real-time performance is poor, and the on-line control requirement of an actual system is difficult to meet. Disclosure of Invention The invention aims to provide a combined optimization control method for a ventilation air-conditioning system air system and a water system, which aims to solve the problems of poor synergy, weak adaptability and high energy consumption in the control of the existing ventilation air-conditioning system air and water systems, realize the collaborative optimization adjustment of key control parameters of the air system and the water system, furthest reduce the total energy consumption of the system on the premise of ensuring the indoor thermal comfort requirement, and simultaneo