Heat pump and AHU air conditioner cross-system AI intelligent group control

Abstract

The invention discloses a heat pump and AHU air conditioner cross-system AI intelligent group control which comprises a warehouse required temperature and AHU control model, an air control model, an AHU water temperature and water quantity and heat pump control model, a heat pump group control model and an AHU group control model, the AHU control model sends out a control signal, the air control model and the AHU water temperature and water quantity and heat pump control model are regulated and controlled to achieve temperature control, and meanwhile, the AHU air conditioner cross-system AI intelligent group control is achieved. And the number of the online air conditioners and the number of the heat pumps are controlled through the AHU group control model and the heat pump group control model. According to the invention, the external air is utilized, and the dynamically adjusted heat pump and AHU air conditioner are combined, so that accurate temperature control and energy optimization are realized, and the purposes of saving energy, protecting environment and reducing operation cost are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of air-conditioning intelligent control, and in particular relates to a cross-system AI intelligent group control of a heat pump and an AHU air-conditioning. Background Art

[0002] Temperature control is a crucial factor for warehouses that store chemicals. Chemicals are very sensitive to temperature, and once the temperature is unstable, it may cause the chemicals to deteriorate or have dangerous reactions. Since changes in the external climate will directly affect the temperature and humidity in the room, warehouses need to rely on efficient air conditioning systems to regulate and maintain them. The air conditioning system must not only ensure that the indoor temperature is within a suitable range, but also maintain appropriate humidity to prevent the chemicals from getting damp or over-drying. However, achieving this goal often requires high energy consumption, which is a major challenge facing current air conditioning systems.

[0003] The temperature and humidity requirements for warehouses vary in different seasons. For example, more cooling functions are required in summer, while strong heating capabilities are required in winter. These requirements further increase the complexity and operating costs of air conditioning systems. Therefore, when designing and managing warehouse air conditioning systems, it is necessary not only to consider how to effectively control temperature and humidity, but also to find energy-saving solutions to reduce energy consumption.

[0004] In addition, to ensure the safety of the warehouse, ventilation must be performed regularly to ensure that harmful gases and chemical odors in the air can be discharged in time. This requires that the exhaust system must have sufficient exhaust volume and be able to maintain a balance with the air supply volume of the air handling unit (AHU). Only in this way can the cleanliness of the indoor air be maintained without destroying the temperature and humidity balance in the warehouse.

[0005] In summary, temperature and humidity control in chemical warehouses is a multifaceted challenge that requires comprehensive consideration of the energy efficiency of the air conditioning system, the requirements of different seasons, and the configuration of the ventilation system. How to achieve energy-saving and efficient operation while ensuring safety and stability is a major difficulty in current research. Summary of the invention

[0006] In order to solve the difficulties of the prior art, the present invention proposes a cross-system AI intelligent group control of heat pumps and AHU air conditioners. First, the warehouse demand temperature and AHU control model are used to obtain the target values ​​of water temperature and wind temperature adjustment, as well as air conditioning energy consumption according to specific environmental parameters and operating conditions. The constantly changing target values ​​are input as parameters into the wind control model, AHU water temperature and water volume and heat pump control model, and AHU group control model on the supply side, thereby realizing cross-system group control, improving and stabilizing the supply and demand balance of warehouse ventilation and temperature, and achieving the purpose of energy saving and emission reduction.

[0007] A heat pump and AHU air conditioner cross-system AI intelligent group control, first includes a data acquisition module, which is used to obtain the following data at different times: set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 , water outlet temperature T 出水 , return water temperature T 回水 , Main pipe temperature T 管 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 , heat pump load rate L 泵 、Number of heat pumps online 泵 , Water valve adjustment proportional coefficient P 水阀 , Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀、 Target water temperature T 水 and AHU air conditioning energy consumption Q AHU 、Number of air conditioners online AHU 、AHU air conditioning load factor L AHU .

[0008] Secondly, it also includes the warehouse demand temperature and AHU control model. First, the set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , water outlet temperature T 出水 As the input of the model, the heat pump energy consumption Q 热 and AHU air conditioning energy consumption Q AHU The first data set is constructed as the output of the model, and then the deep learning network is used to train the first data set to obtain the warehouse demand temperature and AHU control model; then, the current set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , water outlet temperature T 出水 As the input of the model, the target water volume V at the current moment is obtained 泵、 Target water temperature T水 and AHU air conditioning energy consumption Q AHU .

[0009] The cross-system AI intelligent group control of heat pumps and AHU air conditioners also includes wind control models.

[0010] The wind control model first uses the AHU air conditioning energy consumption Q AHU , Fan outlet temperature T 出风 , Fan return air temperature T 回风 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 As the input of the model, the fan outlet temperature T 出风 , fan valve opening O 风 , fan frequency F 风 The second data set is constructed as the output of the model, and the wind control model is obtained by training the second data set using a deep learning network; then, the AHU air conditioning energy consumption Q at the current moment is calculated. AHU , Fan outlet temperature T 出风 , Fan return air temperature T 回风 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 As the input of the model, the fan outlet temperature T at the current moment is obtained. 风 , fan valve opening O 风 , fan frequency F 风 .

[0011] The cross-system AI intelligent group control of heat pumps and AHU air conditioners also has AHU water temperature and water volume and heat pump control models;

[0012] The AHU water temperature and water volume and heat pump control model includes a pre-trained heat pump main pipeline temperature demand model and a terminal pressure difference-heat pump frequency model.

[0013] The heat pump main pipe temperature demand model first takes the target water temperature T 水 , water outlet temperature T 出水 , return water temperature T 回水 , Main pipe temperature T 管 As the input of the model, the outlet water temperature T 出水 The third data set is constructed as the model output, and then the third data set is trained using the deep learning network to obtain the heat pump main pipeline temperature demand model; then the current target water temperature T 水 , water outlet temperature T 出水 , return water temperature T 回水 , Main pipe temperature T 管 As the input of the model, we get the current T 出水 .

[0014] The terminal pressure difference-heat pump frequency model first takes the target water volume V 泵 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 As the input of the model, the heat pump water valve opening O 水 , heat pump frequency F 泵 The fourth data set is constructed as the model output, and the terminal pressure difference-heat pump frequency model is obtained by training the third data set using the incremental PID algorithm. Then, the target water volume V at the current moment is used as the 泵 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 As the input of the model, the output is the heat pump water valve opening degree O at the current moment 水 , heat pump frequency F 泵 .

[0015] The cross-system AI intelligent group control of heat pumps and AHU air conditioners also includes: a heat pump group control model applied to warehouse heat pump systems.

[0016] The heat pump group control model first uses the main pipe temperature T 管、 Heat pump load factor L 泵 、Number of heat pumps online 泵 As the input of the model, the number of heat pumps online N 泵 The fifth data set is constructed as the model output, and then the fifth data set is trained using the deep learning network to obtain the heat pump group control model; then, the main pipeline temperature T at the current moment is 管、 Heat pump load factor L 泵 、Number of heat pumps online 泵 As the input of the model, the number of heat pumps online at the current moment is obtained 泵 .

[0017] Cross-system AI intelligent group control of heat pumps and AHU air conditioners also includes: AHU group control model.

[0018] The AHU group control model first takes the number of online air conditioners N as AHU With fan frequency F 风 、AHU air conditioning load factor L AHU As the input of the model, the number of online air conditioners N AHU , fan frequency F 风The sixth data set is constructed as the output of the model, and the AHU group control model is obtained by training the sixth data set using the deep learning network; then the number of air conditioners online at the current moment N is used AHU With F 风 、AHU air conditioning load factor L AHU As the input of the model, the number of online air conditioners N AHU , fan frequency F 风 as the output of the model.

[0019] The warehouse demand temperature and AHU control model also includes an outside air-fresh air control module and a room temperature-water valve control module.

[0020] The outside air-fresh air control module includes an AI automatic mode and a manual mode.

[0021] In manual mode, the staff manually switches the fresh air valve on and off.

[0022] In AI automatic mode, the outside air-fresh air control module includes an air valve automatic control model.

[0023] The air valve automatic control model is based on the outdoor temperature T 外 With indoor temperature T 内 And set temperature T S As the input of the model, the fan valve opening O 风 As the output of the model, the outdoor temperature T is calculated using a deep learning network. 外 , Indoor temperature T 内 , set temperature T S , Engine air valve opening O 风 The air valve automatic control model is obtained by training; then, the outdoor temperature T 外 With indoor temperature T 内 And set temperature T S As the input of the model, the output is the fan valve opening O at the current moment 风 .

[0024] The room temperature-water valve control module includes an AI automatic mode and a manual mode.

[0025] In manual mode, the staff manually switches the on / off status of the water valve.

[0026] In AI automatic mode, the room temperature-water valve control module includes a water valve automatic control model.

[0027] The water valve automatic control model first sets the temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Water valve adjustment proportional coefficient P 水阀, Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀 , Water valve adjustment amount Y 水阀 、Last water valve adjustment amount Y1 水阀、 The last water valve adjustment amount Y2 水阀 As the input parameter of the model, the heat pump water valve opening O 水 The seventh data set is constructed as the output parameter of the model, and the water valve automatic control model is obtained by training the seventh data set using the incremental PID algorithm.

[0028] Then, the current set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Water valve adjustment proportional coefficient P 水阀 , Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀 , Water valve adjustment amount Y 水阀 、Last water valve adjustment amount Y1 水阀、 The last water valve adjustment amount Y2 水阀 As the input parameter of the model, the output is the heat pump water valve opening degree O at the current moment 水 .

[0029] As a further solution, the prediction formula of the temperature prediction model is:

[0030] T2 = (T1*x) + [(T4)*(1-x)];

[0031] T 水 =T2-[V1 / (Q1*C1)];

[0032] L = (T2-T3)*Q1*ρ1*C1;

[0033] Wherein, T2 is: indoor temperature, T1 is: return air temperature, T 水 is the target water temperature, x is the opening of the fresh air valve, T4 is the indoor temperature, Q1 is the energy consumption corresponding to different operating frequencies of the fan, V1 is the amount of ice water in the heat pump, C1 is the constant-pressure specific heat capacity of A2 gas, ρ1 is the density of air, and L is the required energy corresponding to the required temperature.

[0034] As a further solution, the cost control formula of the AHU cost control model is: fan frequency = fan energy consumption 3 .

[0035] As a further solution, the incremental PDI formula of the terminal pressure difference-heat pump frequency model and the water valve automatic control model is: next adjustment amount = P*(this error-last error)+I*(this error)+D*(this error-2*last error+last error).

[0036] As a further solution, the conditions for adding machines in the heat pump group control model include at least: the temperature of the heat pump main network is greater than 9 degrees Celsius, and the average load of the heat pump during operation is greater than 90%. The conditions for reducing machines in the heat pump group control model include at least: the residual load of the heat pump during operation is greater than the load setting value.

[0037] As a further solution, the indoor temperature collected in real time in the data collection module is collected through temperature sensors set at multiple points indoors.

[0038] As a further solution, the outdoor temperature collected in real time is collected by a temperature sensor arranged outdoors.

[0039] Compared with the prior art, the present invention has at least the following advantages:

[0040] (1) By cleverly utilizing the changes in external air and matching it according to the indoor temperature requirements, the present invention makes full use of external air resources, thereby effectively reducing energy waste; when the external air cannot fully meet the indoor temperature requirements, the system automatically starts the heat pump to supplement the required temperature. This design ensures maximum energy conservation, optimizes energy utilization efficiency, and achieves the dual goals of environmental protection and economic benefits.

[0041] (2) The cross-system AI intelligent group control of heat pumps and AHU air conditioners with real-time feedback ensures real-time regulation of warehouse temperature and improves the stability and energy efficiency of the warehouse environment. According to the actual temperature requirements in the warehouse, the operating parameters of the air-conditioning equipment are dynamically adjusted to achieve precise temperature control. This not only helps to reduce the impact of temperature fluctuations on items, but also saves energy and reduces operating costs. In addition, the cross-system AI intelligent group control of heat pumps and AHU air conditioners with automatic control can also reduce manual intervention and improve management efficiency.

[0042] (3) Through the heat pump group control model and the AHU group control model, the operating power of the heat pump and AHU air conditioner can be finely and comprehensively controlled to optimize energy efficiency while ensuring that the indoor temperature and humidity requirements are met, thereby achieving maximum energy saving effects. This integrated system management can effectively reduce energy consumption and lower operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0044] Figure 1 This is the AI ​​intelligent group control architecture diagram for heat pumps and AHU air conditioners across systems;

[0045] Figure 2 This is the internal network architecture diagram of cross-system AI intelligent group control of heat pumps and AHU air conditioners;

[0046] Figure 3 This is a schematic diagram of the AHU air conditioning structure. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned purpose, control system and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0048] The present invention provides a heat pump and AHU air conditioner cross-system AI intelligent group control, comprising:

[0049] The data acquisition module is used to obtain the following data at different times: set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 , water outlet temperature T 出水 , return water temperature T 回水 , Main pipe temperature T 管 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 , heat pump load rate L 泵 、Number of heat pumps online 泵 , Water valve adjustment proportional coefficient P 水阀 , Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀、 Target water temperature T 水 and AHU air conditioning energy consumption Q AHU 、Number of air conditioners online AHU 、AHU air conditioning load factor L AHU .

[0050] Warehouse demand temperature and AHU control model, first set the temperature T S , outdoor temperature T外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , water outlet temperature T 出水 As the input of the model, the heat pump energy consumption Q 热 and AHU air conditioning energy consumption Q AHU The first data set is constructed as the output of the model, and then the deep learning network is used to train the first data set to obtain the warehouse demand temperature and AHU control model; then, the current set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , water outlet temperature T 出水 As the input of the model, the target water volume V at the current moment is obtained 泵、 Target water temperature T 水 and AHU air conditioning energy consumption Q AHU .

[0051] The intelligent linkage adjustment function of the warehouse demand temperature and the AHU control model can be dynamically adjusted according to the temperature demand in the warehouse and the changes in the external ambient temperature. When the external temperature changes, the system can sense and automatically adjust the AHU air conditioning energy consumption Q in real time. AHU And the target water volume V 泵、 Target water temperature T 水 , in order to maintain the optimal temperature state in the warehouse. This intelligent adjustment significantly improves the accuracy and efficiency of temperature control, reduces energy waste, and achieves the goal of energy saving and consumption reduction.

[0052] The cross-system AI intelligent group control of the heat pump and AHU air conditioner also includes a wind control model.

[0053] The wind control model first uses the AHU air conditioning energy consumption Q AHU , Fan outlet temperature T 出风 , Fan return air temperature T 回风 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 As the input of the model, the fan outlet temperature T 出风 , fan valve opening O 风 , fan frequency F 风 , construct a second data set as the output of the model, and use the deep learning network to train the second data set to obtain the wind control model; then, the AHU air conditioning energy consumption Q at the current moment is AHU , Fan outlet temperature T 出风 , Fan return air temperature T回风 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 As the input of the model, the fan outlet temperature T at the current moment is obtained. 风 , fan valve opening O 风 , fan frequency F 风 .

[0054] Specifically, the wind control model calculates the AHU air conditioning energy consumption Q AHU , Fan outlet temperature T 出风 , Fan return air temperature T 回风 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 Combine them to find the lowest power consumption strategy, and then adjust the fan outlet temperature T at the current moment. 风 , fan valve opening O 风 , fan frequency F 风 Back control ensures the warehouse air supply and exhaust air volume requirements, and ensures that the fan operating frequency is always within an appropriate range, avoiding waste of electrical energy.

[0055] The cross-system AI intelligent group control of heat pumps and AHU air conditioners also has an AHU water temperature and water volume and heat pump control model, which is used to provide the required target water volume V for the warehouse demand temperature and AHU control model. 泵、 Target water temperature T 水 .

[0056] The AHU water temperature and water volume and heat pump control model includes two models: the heat pump main pipe temperature demand model and the terminal pressure difference-heat pump frequency model. The heat pump main pipe temperature demand model is used to control the main pipe temperature T 管， The terminal pressure difference-heat pump frequency model is used to control the opening of the heat pump water valve O 水 , heat pump frequency F 泵 , thereby outputting the target water volume V 泵 .

[0057] Among them, the heat pump main pipe temperature demand model first takes the target water temperature T 水 , water outlet temperature T 出水 , return water temperature T 回水 , Main pipe temperature T 管 As the input of the model, the outlet water temperature T 出水 The third data set is constructed as the model output, and then the third data set is trained using the deep learning network to obtain the heat pump main pipeline temperature demand model; then the current target water temperature T 水 , water outlet temperature T 出水, return water temperature T 回水 , Main pipe temperature T 管 As the input of the model, we get the current T 出水 .

[0058] Terminal pressure difference-heat pump frequency model, first with the target water volume V 泵 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 As the input of the model, the heat pump water valve opening O 水 , heat pump frequency F 泵 The fourth data set is constructed as the model output, and the terminal pressure difference-heat pump frequency model is obtained by training the third data set using the incremental PID algorithm. Then, the target water volume V at the current moment is used as the 泵 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 As the input of the model, the output is the heat pump water valve opening degree O at the current moment 水 , heat pump frequency F 泵 .

[0059] The cross-system AI intelligent group control of heat pumps and AHU air conditioners also includes a heat pump group control model.

[0060] The heat pump group control model first uses the main pipe temperature T 管、 Heat pump load factor L 泵 、Number of heat pumps online 泵 As the input of the model, the number of heat pumps online N 泵 The fifth data set is constructed as the model output, and then the fifth data set is trained using the deep learning network to obtain the heat pump group control model; then, the main pipeline temperature T at the current moment is 管、 Heat pump load factor L 泵 、Number of heat pumps online 泵 As the input of the model, the number of heat pumps online at the current moment is obtained 泵 .

[0061] The heat pump group control model is used to control the heat pump load capacity and the main pipe temperature T 管, regulate the number of heat pumps. When the online heat pump load is higher than the set load and the main network temperature is higher than the set temperature, increase the number of online heat pumps. When the online heat pump load is lower than the set load and the main network temperature is lower than the set temperature, reduce the number of online heat pumps. The setting of the heat pump group control model not only helps to improve energy efficiency, but also regulates the system temperature according to the main pipe temperature to the greatest extent, thereby improving the temperature stability of the system. In addition, the heat pump group control model can respond flexibly to load changes, reduce the risk of overload of a single heat pump, and improve the reliability and service life of the entire system. The use of AI technology to achieve automated control can reduce manual intervention and decision-making, improve management efficiency and reduce the occurrence of human errors. More importantly, the number of operating heat pumps is optimized, and timely adjustments are made when demand fluctuates, reducing unnecessary equipment operating time, reducing equipment energy consumption, and optimizing operating costs.

[0062] The heat pump and AHU air conditioner cross-system AI intelligent group control also includes: an AHU group control model.

[0063] The AHU group control model is used to control multiple AHU air conditioners in a group. It achieves the purpose of saving electricity through motor frequency conversion joint control. Since the fan frequency and fan power are in a cubic relationship, the same air supply volume can be achieved by different startup numbers and frequencies, and the strategy with the lowest total energy consumption is sought. For the AHU group control model, firstly, the number of online air conditioners N is used. AHU With F 风 、AHU air conditioning load factor L AHU As the input of the model, the number of online air conditioners N AHU 、F 风 The sixth data set is constructed as the output of the model, and the AHU group control model is obtained by training the sixth data set using the deep learning network; then the number of air conditioners online at the current moment N is used AHU With fan frequency F 风 、AHU air conditioning load factor L AHU As the input of the model, the number of online air conditioners N AHU , fan frequency F 风 as the output of the model.

[0064] The warehouse demand temperature and AHU control model also includes an outside air-fresh air control module and a room temperature-water valve control module.

[0065] The external air-fresh air control module includes an AI automatic mode and a manual mode;

[0066] In manual mode, the staff manually adjusts the switch status of the fresh air valve. The fresh air valve is used to control the circulation of indoor and outdoor air. When the warehouse needs ventilation, the staff sets the fresh air valve to automatic mode to introduce outdoor fresh air or exhaust indoor air to achieve air circulation and renewal.

[0067] When ventilation is not required, the fresh air valve control module is generally in automatic mode.

[0068] The fresh air valve control module collects the outdoor temperature T of the warehouse in real time. 外 With indoor temperature T 内 , based on the outdoor temperature T 外 With indoor temperature T 内 The size relationship between them automatically controls the fan valve opening O 风 .

[0069] The outside air-fresh air control module includes an air valve automatic control model in the AI ​​automatic mode.

[0070] The air valve automatic control model is based on the outdoor temperature T 外 With indoor temperature T 内 And set temperature T S As the input of the model, the fan valve opening O 风 As the output of the model, the outdoor temperature T is calculated using a deep learning network. 外 , Indoor temperature T 内 , set temperature T S , Engine air valve opening O 风 The air valve automatic control model is obtained by training; then, the outdoor temperature T 外 With indoor temperature T 内 And set temperature T S As the input of the model, the output is the fan valve opening O at the current moment 风 .

[0071] The room temperature-water valve control module also includes an AI automatic mode and a manual mode.

[0072] In manual mode, the staff manually switches the water valve on and off to control the water supply. In AI automatic mode, the room temperature-water valve control module includes a water valve automatic control model. The water valve automatic control model first sets the temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Water valve adjustment proportional coefficient P 水阀 , Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀 , Water valve adjustment amount Y 水阀、Last water valve adjustment amount Y1 水阀、 The last water valve adjustment amount Y2 水阀 As the input parameter of the model, the heat pump water valve opening O 水 The seventh data set is constructed as the output parameter of the model, and the third data set is trained using the incremental PID algorithm to obtain the water valve automatic control model; then, the current set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Water valve adjustment proportional coefficient P 水阀 , Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀 , Water valve adjustment amount Y 水阀 、Last water valve adjustment amount Y1 水阀、 The last water valve adjustment amount Y2 水阀 As the input parameter of the model, the output is the heat pump water valve opening degree O at the current moment 水 .

[0073] The design of the outside air-fresh air control module and the room temperature-water valve control module takes into account flexibility, high efficiency and cost savings. On the one hand, the real-time control of the outside air-fresh air control module and the room temperature-water valve control module ensures that the warehouse environment is always in the best condition. At the same time, the existence of the manual mode allows staff to make necessary adjustments to ensure that the system can operate stably under various circumstances. In addition, the use of historical data and real-time data for training and adjustment can not only make the best decisions in the current environment, but also improve future control accuracy and efficiency through continuous learning and optimization. Finally, AI and automatic control models are used to optimize energy use and reduce unnecessary energy consumption.

[0074] Furthermore, the temperature prediction model predicts by the following formula:

[0075] T2 = (T1*x) + [(T4)*(1-x)];

[0076] T 水 =T2-[V1 / (Q1*C1)];

[0077] L = (T2-T3)*Q1*ρ1*C1;

[0078] Among them, T2 is the indoor temperature, T1 is the return air temperature, T 水 is the target water temperature, x is the opening of the fresh air valve, T4 is the indoor temperature, Q1 is the energy consumption corresponding to different operating frequencies of the fan, V1 is the amount of ice water in the heat pump, C1 is the constant pressure specific heat capacity of A2 gas, ρ1 is the density of air, and L is the required energy corresponding to the required temperature.

[0079] The AHU group control model performs cost control through the following formula: fan frequency = fan energy consumption 3 .

[0080] The incremental PDI formula of the terminal pressure difference-heat pump frequency model and the water valve automatic control model is: next adjustment amount = P*(this error-last error)+I*(this error)+D*(this error-2*last error+last error). P, I and D are three parameters of PID control. The period of collecting frequency data can be set manually, such as 1 minute or half a minute.

[0081] The heat pump group control model adding machine conditions at least include: the heat pump main network temperature is greater than 9 degrees Celsius, and the average load of the heat pump during operation is greater than 90%. The heat pump group control model reducing machine conditions at least include: the remaining load of the heat pump during operation is greater than the load setting value.

[0082] Among them, based on the results of the heat pump group control model, the target value of the main network temperature can be set to 9 degrees Celsius, the additional machine load is set to 90%, the additional machine temperature upper limit is set to 11 degrees Celsius, the waiting time after the addition condition is met is set to 120s, the additional machine cooling pump delay time is 30s, the total time for judging the failure of adding machines is 300s, the upper limit setting value of the ice machine outlet water temperature is 9 degrees Celsius, the lower limit setting value of the ice machine outlet water temperature is 6 degrees Celsius, the percentage of the small ice machine load to the large ice machine load is set to 0.6, the setting value of the residual load when reducing the large ice machine is set to 1.02, the setting value of the residual load when reducing the small ice machine is set to 0.65, the waiting time after the reduction condition is met is set to 120s, the ice machine delay time is set to 600s, the cooling pump delay time is set to 10s, and the total time for judging the failure of reducing machines is set to 1200s.

[0083] Among them, the indoor temperature T collected in real time in the data acquisition module 内 The outdoor temperature T is collected in real time by temperature sensors installed at multiple points indoors. 外 The temperature is collected through the temperature sensors installed outdoors. By installing temperature sensors at multiple locations indoors and outdoors, temperature data at different locations can be collected, thereby improving measurement accuracy and reducing local errors.

[0084] The above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the protection scope of the present invention.

Claims

1. A heat pump and AHU air conditioner cross-system AI intelligent group control, characterized in that: include: The data acquisition module is used to obtain the following data at different times: set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 , water outlet temperature T 出水 , return water temperature T 回水 , Main pipe temperature T 管 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 , heat pump load rate L 泵 、Number of heat pumps online 泵 , Water valve adjustment proportional coefficient P 水阀 , Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀、 Target water temperature T 水 and AHU air conditioning energy consumption Q AHU 、Number of air conditioners online AHU 、AHU air conditioning load factor L AHU ; Warehouse demand temperature and AHU control model, first set the temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , water outlet temperature T 出水 As the input of the model, the heat pump energy consumption Q 热 and AHU air conditioning energy consumption Q AHU The first data set is constructed as the output of the model, and then the deep learning network is used to train the first data set to obtain the warehouse demand temperature and AHU control model; then, the current set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Fan outlet temperature T 出风 , Fan return air temperature T 回 , water outlet temperature T 出水 As the input of the model, the target water volume V at the current moment is obtained 泵、 Target water temperature T 水 and AHU air conditioning energy consumption Q AHU .

2. The heat pump and AHU air conditioner cross-system AI intelligent group control according to claim 1 is characterized in that: Also includes wind control models; The wind control model first uses the AHU air conditioning energy consumption Q AHU , Fan outlet temperature T 出风 , Fan return air temperature T 回风 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 As the input of the model, the fan outlet temperature T 出风 , fan valve opening O 风 , fan frequency F 风 The second data set is constructed as the output of the model, and the wind control model is obtained by training the second data set using a deep learning network; then, the AHU air conditioning energy consumption Q at the current moment is calculated. AHU , Fan outlet temperature T 出风 , Fan return air temperature T 回风 , fan valve opening O 风 , fan frequency F 风 , fan power P 风 As the input of the model, the fan outlet temperature T at the current moment is obtained. 风 , fan valve opening O 风 , fan frequency F 风 .

3. The heat pump and AHU air conditioner cross-system AI intelligent group control according to claim 1 is characterized in that: Also includes: AHU water temperature and water volume and heat pump control model; The AHU water temperature and water volume and heat pump control model includes a pre-trained heat pump main pipe temperature demand model and a terminal pressure difference-heat pump frequency model; The heat pump main pipe temperature demand model first takes the target water temperature T 水 , water outlet temperature T 出水 , return water temperature T 回水 , Main pipe temperature T 管 As the input of the model, the outlet water temperature T 出水 The third data set is constructed as the model output, and then the third data set is trained using the deep learning network to obtain the heat pump main pipeline temperature demand model; then the current target water temperature T 水 , water outlet temperature T 出水 , return water temperature T 回水 , Main pipe temperature T 管 As the input of the model, we get the current T 出水 ; The terminal pressure difference-heat pump frequency model first takes the target water volume V 泵 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 As the input of the model, the heat pump water valve opening O 水 , heat pump frequency F 泵 The fourth data set is constructed as the model output, and the terminal pressure difference-heat pump frequency model is obtained by training the third data set using the incremental PID algorithm. Then, the target water volume V at the current moment is used as the 泵 , heat pump frequency F 泵 , heat pump regulation proportional coefficient P 泵 , Water pump adjustment amount X 泵 、Last water pump adjustment amount X1 泵、 The last water pump adjustment amount X2 泵 As the input of the model, the output is the heat pump water valve opening degree O at the current moment 水 , heat pump frequency F 泵 ; The heat pump and AHU air conditioner cross-system AI intelligent group control also includes: a heat pump group control model applied to the warehouse heat pump system; The heat pump group control model first uses the main pipe temperature T 管、 Heat pump load factor L 泵 、Number of heat pumps online 泵 As the input of the model, the number of heat pumps online N 泵 The fifth data set is constructed as the model output, and then the fifth data set is trained using the deep learning network to obtain the heat pump group control model; then, the main pipeline temperature T at the current moment is 管、 Heat pump load factor L 泵 、Number of heat pumps online 泵 As the input of the model, the number of heat pumps online at the current moment is obtained 泵 .

4. The heat pump and AHU air conditioner cross-system AI intelligent group control according to claim 1 is characterized in that: Also includes: AHU group control model; The AHU group control model first takes the number of online air conditioners N as AHU With fan frequency F 风 、AHU air conditioning load factor L AHU As the input of the model, the number of online air conditioners N AHU , fan frequency F 风 The sixth data set is constructed as the output of the model, and the AHU group control model is obtained by training the sixth data set using the deep learning network; then the number of air conditioners online at the current moment N is used AHU With F 风 、AHU air conditioning load factor L AHU As the input of the model, the number of online air conditioners N AHU , fan frequency F 风 as the output of the model.

5. According to claim 1, a heat pump and AHU air conditioner cross-system AI intelligent group control is characterized in that: The warehouse demand temperature and AHU control model also includes an outside air-fresh air control module and a room temperature-water valve control module; The external air-fresh air control module includes an AI automatic mode and a manual mode; In manual mode, the staff manually switches the fresh air valve on and off; In AI automatic mode, the outside air-fresh air control module includes an automatic air valve control model; The air valve automatic control model is based on the outdoor temperature T 外 With indoor temperature T 内 And set temperature T S As the input of the model, the fan valve opening O 风 As the output of the model, the outdoor temperature T is calculated using a deep learning network. 外 , Indoor temperature T 内 , set temperature T S , Engine air valve opening O 风 The air valve automatic control model is obtained by training; then, the outdoor temperature T 外 With indoor temperature T 内 And set temperature T S As the input of the model, the output is the fan valve opening O at the current moment 风 ; The room temperature-water valve control module includes AI automatic mode and manual mode; In manual mode, the staff manually switches the on / off status of the water valve; In AI automatic mode, the room temperature-water valve control module includes a water valve automatic control model; The water valve automatic control model first sets the temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Water valve adjustment proportional coefficient P 水阀 , Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀 , Water valve adjustment amount Y 水阀 、Last water valve adjustment amount Y1 水阀、 The last water valve adjustment amount Y2 水阀 As the input parameter of the model, the heat pump water valve opening O 水 The seventh data set is constructed as the output parameter of the model, and the water valve automatic control model is obtained by training the seventh data set using the incremental PID algorithm; Then, the current set temperature T S , outdoor temperature T 外 With indoor temperature T 内 , Water valve adjustment proportional coefficient P 水阀 , Water valve adjustment integral coefficient I 水阀 , Water valve adjustment differential coefficient D 水阀 , Water valve adjustment amount Y 水阀 、Last water valve adjustment amount Y1 水阀、 The last water valve adjustment amount Y2 水阀 As the input parameter of the model, the output is the heat pump water valve opening degree O at the current moment 水 .

6. The heat pump and AHU air conditioner cross-system AI intelligent group control according to claim 1 is characterized in that: The prediction formula of the temperature prediction model is: T2 = (T1*x) + [(T4)*(1-x)]; T 水 =T2-[V1 / (Q1*C1)]; L = (T2-T3)*Q1*ρ1*C1; Wherein, T2 is: indoor temperature, T1 is: return air temperature, T 水 is the target water temperature, x is the opening of the fresh air valve, T4 is the indoor temperature, Q1 is the energy consumption corresponding to different operating frequencies of the fan, V1 is the amount of ice water in the heat pump, C1 is the constant pressure specific heat capacity of A2 gas, ρ1 is the density of air, and L is the required energy corresponding to the required temperature; The cost control formula of the AHU cost control model is: fan frequency = fan energy consumption 3 ; The incremental PDI formula of the terminal pressure difference-heat pump frequency model and the water valve automatic control model is: next adjustment amount = P*(this error-last error)+I*(this error)+D*(this error-2*last error+last error).

7. The heat pump and AHU air conditioner cross-system AI intelligent group control according to claim 1 is characterized in that: The heat pump group control model adding machine conditions at least include: the heat pump main network temperature is greater than 9 degrees Celsius, and the average load of the heat pump during operation is greater than 90%. The heat pump group control model reducing machine conditions at least include: the remaining load of the heat pump during operation is greater than the load setting value.

8. The heat pump and AHU air conditioner cross-system AI intelligent group control according to claim 1 is characterized in that: The indoor temperature collected in real time in the data collection module is collected through temperature sensors set at multiple points indoors; The outdoor temperature collected in real time is collected by a temperature sensor arranged outdoors.

Images