Air conditioner control method, control equipment and storage medium

Abstract

The embodiment of the invention provides an air conditioner control method and device and a storage medium, the method is used for conducting cooperative control on multiple air conditioner systems deployed in different buildings of a target park, and the method comprises the steps that people flow trend prediction data of the target park in the future time period is obtained, the people flow trend prediction data represents migration and / or distribution conditions of people among different buildings; based on the people flow trend prediction data, target loads of the multiple air conditioning systems are determined; based on the respective target loads of the multiple air conditioning systems, the multiple air conditioning systems serve as a whole, combined planning is conducted on the operation parameters of the multiple air conditioning systems, and a target control strategy used for conducting cooperative control on the multiple air conditioning systems is obtained; and on the basis of the target control strategy, control operation is executed on the multiple air conditioning systems.

Description

Technical Field

[0001] This specification relates to the field of air conditioning control technology, and in particular to an air conditioning control method, control device and storage medium. Background Technology

[0002] Currently, air conditioning systems are used in various buildings to regulate ambient temperature and ensure the comfort of people's working and living environment. With the increasing requirements for energy conservation and emission reduction, air conditioning systems, due to their high energy consumption, are also required to be as energy-efficient as possible while meeting environmental needs.

[0003] For example, a campus (such as a university campus or science park) might contain multiple buildings, each equipped with an air conditioning system. The personnel controlling the air conditioning systems determine the operating times and the proper functioning of each device based on the day's weather conditions and experience, and then implement corresponding equipment controls. The air conditioning systems in all buildings are typically controlled in the same way.

[0004] This method relies on human experience, and manual control of the air conditioning system may not match the actual air conditioning needs, resulting in poor system performance or high energy consumption. Therefore, how to automate and more accurately control the air conditioning system to improve the overall energy efficiency of the air conditioning systems in various buildings within the park is a technical issue that needs to be considered.

[0005] The information in the background section is merely information known only to the inventor and does not imply that such information had entered the public domain before the date of this application, nor does it imply that it can be considered prior art in this disclosure. Summary of the Invention

[0006] This specification provides an air conditioning control method, control device, and storage medium, which can determine a suitable target control strategy for coordinated control of the air conditioning systems in various buildings within a target park, making the control of each air conditioning system more precise and improving the overall energy efficiency of each air conditioning system.

[0007] Firstly, this specification provides an air conditioning control method for the coordinated control of multiple air conditioning systems deployed in different buildings within a target park. The method includes: obtaining pedestrian flow trend prediction data for the target park in a future time period, wherein the pedestrian flow trend prediction data characterizes the migration and / or distribution of people between different buildings; determining the target load of each of the multiple air conditioning systems based on the pedestrian flow trend prediction data; jointly planning the operating parameters of the multiple air conditioning systems as a whole based on the target load of each of the multiple air conditioning systems to obtain a target control strategy for the coordinated control of the multiple air conditioning systems; and executing control operations on the multiple air conditioning systems based on the target control strategy.

[0008] Secondly, this specification also provides a control device, comprising: at least one storage medium storing at least one instruction set for controlling an air conditioning system; and at least one processor communicatively connected to the at least one storage medium, wherein the at least one processor reads the at least one instruction set during operation and executes the method described in the first aspect according to the instructions of the at least one instruction set.

[0009] Thirdly, this specification also provides a computer-readable non-volatile storage medium, wherein the computer-readable non-volatile storage medium stores at least one instruction set, which, when executed by at least one processor, implements the method described in the first aspect above.

[0010] The air conditioning control method, control equipment, and storage medium provided in this manual, for a target industrial park, determine the target load of each building's air conditioning system based on pedestrian flow trend prediction data characterizing the migration and / or distribution of people between different buildings in the future. Then, based on this target load, the operating parameters of multiple air conditioning systems are jointly planned as a whole to obtain a target control strategy for coordinated control of these systems, thereby executing control operations on each system. In this way, by considering the pedestrian flow relationships between buildings in the target industrial park, the target load of each building's air conditioning system can be determined more accurately. This allows for the determination of a target control strategy that matches the current demand for each air conditioning system as a whole. Coordinated control of each air conditioning system based on this target control strategy improves the control accuracy of each system and enhances the overall energy efficiency of the air conditioning systems in the target industrial park.

[0011] Other functions of the air conditioning control methods, control devices, and storage media provided in this specification will be partially listed in the following description. The inventive aspects of the air conditioning control methods, control devices, and storage media provided in this specification can be fully explained by practice or use of the methods, devices, and combinations described in the detailed examples below. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic diagram illustrating an application scenario of an air conditioning control method provided according to an embodiment of this specification is shown. Figure 2 A hardware structure diagram of a control device provided according to an embodiment of this specification is shown; Figure 3 A flowchart of an air conditioning control method according to an embodiment of this specification is shown; Figure 4 A schematic diagram of the encoding structure of an individual according to an embodiment of this specification is shown; and Figure 5 A schematic diagram of a candidate control strategy provided according to an embodiment of this specification is shown. Detailed Implementation

[0014] The following description provides specific application scenarios and requirements for this specification, intended to enable those skilled in the art to make and use the contents of this specification. Various partial modifications to the disclosed embodiments will be apparent to those skilled in the art, and the general principles defined herein can be applied to other embodiments and applications without departing from the spirit and scope of this specification. Therefore, this specification is not limited to the embodiments shown, but rather to the widest scope consistent with the claims.

[0015] The terminology used herein is for the purpose of describing particular exemplary embodiments only and is not restrictive. For example, unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” used herein may also include the plural forms. When used in this specification, the terms “comprising,” “including,” and / or “containing” mean that the associated integers, steps, operations, elements, and / or components are present, but do not preclude the presence of one or more other features, integers, steps, operations, elements, components, and / or groups, or the possibility of adding other features, integers, steps, operations, elements, components, and / or groups to the system / method. The term “multiple” refers to two or more, and “at least one” refers to one or more. “At least one of A, B, or C” includes seven options: A only, B only, C only, containing both A and B, containing both A and C, containing both B and C, and containing both A, B, and C. And so on, the statement “at least one of multiple items” refers to all combinations that can be formed based on these items.

[0016] Considering the following description, these and other features of this specification, as well as the operation and function of the related components of the structure, and the economy of assembly and manufacture of the parts, can be significantly improved. All of these form part of this specification with reference to the accompanying drawings. However, it should be clearly understood that the drawings are for illustrative and descriptive purposes only and are not intended to limit the scope of this specification. It should also be understood that the drawings are not drawn to scale.

[0017] The flowcharts used in this specification illustrate operations implemented according to some embodiments of this specification. It should be clearly understood that the operations in the flowcharts may not be implemented in a sequential order. Instead, the operations may be implemented in reverse order or simultaneously. Furthermore, one or more additional operations may be added to the flowcharts. One or more operations may be removed from the flowcharts.

[0018] Improving energy efficiency in various buildings (such as commercial buildings, data centers, hospitals, and schools) is urgently needed. Buildings typically have Heating, Ventilation, and Air Conditioning (HVAC) systems, which account for a significant proportion of a building's energy consumption. A campus (such as a school, hospital, or science park) may include multiple buildings, each with its own air conditioning system. The air conditioning system regulates the temperature of each space within the building to achieve the desired temperature. This desired temperature can be a comfortable ambient temperature for people or the ambient temperature required for the normal operation of equipment installed in the space. Different types of air conditioning systems can be deployed in different buildings. For example, some air conditioning systems (such as water-cooled systems) only have cooling functions, while others (such as air-cooled heat pump systems) have both cooling and heating functions, and some systems may only have heating functions. The air conditioning control methods provided in this manual can be applied to various types of air conditioning systems.

[0019] The air conditioning control method provided in this manual is used to coordinate the control of multiple air conditioning systems deployed in different buildings in the target park. Figure 1 A schematic diagram illustrating an application scenario of an air conditioning control method provided according to an embodiment of this specification is shown. For example... Figure 1 As shown, the application scenario 00 includes a target park 100, an air conditioning system 110, and a control device 120.

[0020] Target campus 100 includes multiple buildings. If target campus 100 is a university campus, it may include teaching buildings, laboratory buildings, canteens, and dormitories. If target campus 100 is a science park, it may include office buildings, research and development centers, and production workshops. Figure 1 The Sino-Israeli target campus 100 includes three buildings as a simple example. At least some of the buildings in the target campus 100 may be equipped with air conditioning systems 110. This specification only describes the buildings in the target campus 100 that are equipped with air conditioning systems 110.

[0021] Figure 1This paper takes a water-cooled air conditioning system 110 deployed in a building as an example. A water-cooled air conditioning system is a complex system characterized by multiple variables, strong coupling, nonlinearity, time-varying behavior, and dynamic influence from the external environment. It utilizes water as a cooling medium, circulating water and employing evaporation and conduction to dissipate heat, achieving efficient cooling to lower indoor temperatures. The air conditioning system 110 can be divided into two main parts: the main unit side and the terminal side. The main unit side primarily involves the circulation of refrigerant and the production and flow of chilled water. The terminal side includes devices that directly provide cooling or heating to the room, regulating indoor temperature through heat exchange with chilled water.

[0022] Please continue to refer to this. Figure 1 The main unit side of the air conditioning system 110 includes the main unit equipment, cooling equipment, and water pump equipment. For example... Figure 1 As shown, the main unit can be a water-cooled main unit 111, the cooling equipment is a cooling tower 112, and the water pump equipment includes a cooling pump 113 and a chilled water pump 114. The terminal side includes fan coil units 115, air handling units (AHUs) 116, temperature control panels (i.e., indoor temperature-regulating devices, not shown in the figure), and other radiators. In this specification, the terminal-side equipment is referred to as terminal equipment. Terminal equipment is installed in spaces close to the user. For example, fan coil units 115 are distributed throughout various rooms or areas of a building, such as being installed in the ceiling or on the walls of each room.

[0023] The water-cooled unit 111 is the core component of the water-cooled air conditioning system. The water-cooled unit 111 typically consists of a compressor, evaporator, and condenser. The water-cooled unit 111 generates chilled water by compressing the refrigerant and cooling the circulating water. The chilled water generated by the water-cooled unit 111 circulates between the water-cooled unit 111 and the terminal devices. The chilled water pump 114 is located between the water-cooled unit 111 and the terminal devices. The chilled water pump 114 delivers the chilled water generated by the water-cooled unit 111 to each terminal device, ensuring the circulation of chilled water to create an effective cooling zone. The terminal devices are responsible for converting the chilled water into cool air and regulating the indoor temperature.

[0024] For example, the fan coil unit 115 in the terminal equipment delivers treated air into the room through an internally circulating chilled water coil to achieve a cooling effect. The air handling unit 116 in the terminal equipment is used to treat the air inside and outside the building, mainly responsible for heating, cooling, dehumidifying, and purifying the air to ensure indoor air quality and comfort. The air handling unit 116 can be installed in a single machine room for centralized treatment of air from multiple areas. For example, the air required for an entire floor can be centrally processed by one air handling unit 116. The air handling unit 116 delivers the treated air to the corresponding fan coil unit 115 in each room, where the fan coil unit 115 delivers air into the room. The air handling unit 116 has the ability to adjust the supply air temperature and monitor the return air temperature. Figure 1 The illustration takes an example where the terminal equipment includes both a fan coil unit 115 and an air handling unit 116. In some embodiments, depending on the building's design requirements, the fan coil unit 115 and the air handling unit 116 can be used independently.

[0025] Cooling water circulates between the water-cooled chiller 111 and the cooling tower 112. The cooling tower 112 is a heat dissipation device in the water-cooled air conditioning system. The water-cooled chiller 111 releases heat during the generation of chilled water; this heat is carried to the cooling tower 112 by the cooling water, where it is dissipated. A cooling pump 113 is located between the water-cooled chiller 111 and the cooling tower 112. The cooling pump 113 is responsible for returning the heated cooling water from the water-cooled chiller 111 to the cooling tower 112. The cooling tower 112 uses airflow to evaporate the cooling water within the tower, thereby lowering the water temperature and circulating it back to the water-cooled chiller 111. This allows the heat released by the water-cooled chiller 111 to be discharged into the atmosphere, completing heat exchange. Thus, in the air conditioning system 110, the heat generated on the chilled water circulation side is carried away by the cooling water circulation side, thereby maintaining the cooling efficiency of the air conditioning system 110.

[0026] The temperature at which the chilled water generated by the water-cooling unit 111 is discharged from the water-cooling unit 111 is called the chilled water outlet temperature, which can be set. For example, this outlet temperature is typically set to 7 degrees Celsius. The temperature at which the chilled water is circulated back to the water-cooling unit 111 after passing through the terminal equipment is called the chilled water return temperature. In one case, this return temperature is 12 degrees Celsius.

[0027] It should be understood that Figure 1The number of water-cooled main unit 111, cooling tower 112, cooling pump 113, chilled water pump 114, fan coil unit 115, and air handling unit 116 in the description is merely illustrative. Depending on the implementation requirements, the air conditioning system 110 may include any number of these devices. Multiple units of each type can be used, and the number of units in operation and their operating parameters can be adjusted as needed. For example, the operating frequencies of the cooling pump 113 and chilled water pump 114 can be adjusted. The air conditioning system 110 in this specification can also be other types of air conditioning systems. For other types of air conditioning systems, their structure and included equipment can be similar. Figure 1 The structures shown differ to some extent; for example, the air-cooled heat pump system does not include a cooling tower.

[0028] Air conditioning system 110 can be connected to control device 120, such that each device in air conditioning system 110 can be connected to control device 120 respectively. Control device 120 is used to execute the air conditioning control method provided in this specification to achieve coordinated control of multiple air conditioning systems 110 deployed in different buildings in target park 100, that is, to achieve control of each device in each air conditioning system 110. Control device 120 can store data or instructions for executing the air conditioning control method described in this specification, and can execute or be used to execute said data or instructions. In some embodiments, control device 120 may include hardware devices with data information processing functions and the necessary programs required to drive the hardware devices to work.

[0029] The control device 120 can acquire data related to the target park 100 and the air conditioning system 110, and control each air conditioning system 110 based on this data to ensure that the air conditioning system 110 operates according to the indicated operating parameters, thereby providing cooling or heating to the corresponding buildings in the target park 100. This data may include data from each building in the target park 100 (such as actual environmental data like temperature), operating data from the air conditioning system 110, and user-related data. It should be noted that all user-related data acquired in this specification has been authorized by the user and does not infringe on user privacy. The control device 120 can be a server or a terminal device, such as a desktop computer or laptop computer.

[0030] The control device 120 and various devices in the air conditioning system 110 can be connected via a network. In some embodiments, the network can be any type of wired or wireless network, or a combination thereof. For example, the network may include a cable network, wired network, fiber optic network, telecommunications network, intranet, Internet, local area network (LAN), wide area network (WAN), wireless local area network (WLAN), metropolitan area network (MAN), public switched telephone network (PSTN), Bluetooth network, ZigBee network, near field communication (NFC) network, or similar network. In some embodiments, the network may include one or more network access points. For example, the network may include wired or wireless network access points, such as base stations or Internet exchange points, through which the control device 120 and various devices in the air conditioning system 110 can connect to the network to exchange data or information.

[0031] Those skilled in the art should understand that when the air conditioning system 110 is another type of air conditioning system, the implementation method and technical effect of the air conditioning control method provided in this specification are similar. Similar to the control method for water-cooled air conditioning systems, the control methods for other types of air conditioning systems can generally refer to the description of the control methods for water-cooled air conditioning systems in this specification, and will not be repeated here. Alternatively, the control methods for other air conditioning systems can be adaptively adjusted based on actual needs. For example, if other air conditioning systems are in heating mode, the cooling load (i.e., the required cooling capacity) required by the water-cooled air conditioning system needs to be adjusted to the heat load (i.e., the required heating capacity) required by the other air conditioning system, and then processed accordingly. This specification will not elaborate on this further.

[0032] Figure 2 A hardware structure diagram of a control device according to an embodiment of this specification is shown. The control device 120 can be... Figure 1 The control equipment in the device executes the air conditioning control method described in this manual.

[0033] like Figure 2 As shown, the control device 120 may include at least one storage medium 230 and at least one processor 220. In some embodiments, the control device 120 may also include a communication port 250 and an internal communication bus 210. The control device 120 may also include I / O components 260.

[0034] The internal communication bus 210 can connect to different system components. For example, the internal communication bus 210 can connect to storage medium 230, processor 220, communication port 250, and I / O component 260, etc.

[0035] I / O component 260 supports input / output between control device 120 and other components.

[0036] Communication port 250 is used for data communication between control device 120 and the outside world. For example, communication port 250 can be used for data communication between control device 120 and network 140. Communication port 250 can be a wired communication port or a wireless communication port.

[0037] Storage medium 230 may include a data storage device. The data storage device may be a non-transitory storage medium or a temporary storage medium. For example, the data storage device may include one or more of a disk 232, a read-only storage medium (ROM) 234, or a random access storage medium (RAM) 236. Storage medium 230 also includes at least one instruction set stored in the data storage device. The instruction set may include computer program code, which may include programs, routines, objects, components, data structures, procedures, modules, etc.

[0038] At least one processor 220 may be communicatively connected to at least one storage medium 230. When the control device 120 is running, at least one processor 220 reads the at least one instruction set and executes the air conditioning control method provided in this specification according to the instructions of the at least one instruction set. The processor 220 may execute the steps included in the air conditioning control method. The processor 220 may be in the form of one or more processors. In some embodiments, the processor 220 may include one or more hardware processors, such as a microcontroller, microprocessor, reduced instruction set computer (RISC), application-specific integrated circuit (ASIC), application-specific instruction set processor (ASIP), central processing unit (CPU), graphics processing unit (GPU), physical processing unit (PPU), microcontroller unit, digital signal processor (DSP), field-programmable gate array (FPGA), advanced RISC machine (ARM), programmable logic device (PLD), any circuit or processor capable of performing one or more functions, or any combination thereof.

[0039] For illustrative purposes only, the control device 120 in the accompanying drawings shows only one processor 220. However, it should be noted that the control device 120 in this specification may also include multiple processors. Therefore, the operation and / or method steps disclosed in this specification may be executed by one processor or by multiple processors in combination. For example, if the processor 220 of the control device 120 described in this specification executes steps A and B, it should be understood that steps A and B may also be executed jointly or separately by two different processors 220 (e.g., the first processor executes step A, the second processor executes step B, or the first and second processors jointly execute steps A and B).

[0040] For air conditioning systems deployed in buildings, manual control and terminal adaptive control are typically employed. This method relies on fixed schedules or operator experience to manually control the start-up and shutdown of the air conditioning system and its operating parameters; terminal devices adjust parameters such as supply air temperature and air volume based on indoor temperature. This method is difficult to accurately match the building's dynamic load demands, leading to unnecessary energy consumption at times and insufficient indoor comfort at others, resulting in low energy efficiency of the air conditioning system.

[0041] Furthermore, air conditioning systems in individual buildings are typically controlled on a per-building basis, without interconnecting them across different buildings. For example, when considering an air conditioning system deployed in a single building, the goal is to achieve a pre-set, fixed temperature while minimizing energy consumption. This approach determines the control strategy based on achieving the desired temperature and minimizing energy consumption. However, this method struggles to meet the actual operational needs of a building. If a significant event requires further temperature reduction while still maintaining adequate energy efficiency, this approach may not be suitable for determining a definitive control strategy.

[0042] This specification provides an air conditioning control method, control device, and storage medium. It treats multiple air conditioning systems deployed in different buildings within a target area as a whole, considering the impact of personnel migration and / or distribution on the target load of each building's air conditioning system. Based on the target load of each air conditioning system, it performs coordinated control. This allows for a more accurate determination of the target load of each air conditioning system, leading to the development of a target control strategy that matches current needs. Coordinated control of the air conditioning systems based on this target control strategy improves the accuracy of control over each system and enhances the overall energy efficiency of the air conditioning systems within the target area.

[0043] The air conditioning control method provided in this manual will be described in detail below with reference to the accompanying drawings. Figure 3 A flowchart of an air conditioning control method according to an embodiment of this specification is shown. As previously described, the control device 120 can execute the air conditioning control method P300 of this specification. Specifically, the processor 220 in the control device 120 can read the instruction set stored in its local storage medium, and then execute the air conditioning control method P300 of this specification according to the provisions of the instruction set. This method is used for the coordinated control of multiple air conditioning systems deployed in different buildings in a target park, such as... Figure 3 As shown, method P300 may include steps 310 to 340 as described below.

[0044] Step 310: Obtain the predicted pedestrian flow trend data for the target park in the future time period. The predicted pedestrian flow trend data represents the migration and / or distribution of people between different buildings.

[0045] For multiple air conditioning systems deployed in different buildings within a target campus, the control device can treat them as a whole and control them collaboratively. In some embodiments, the control device can control the air conditioning system in time periods. The time periods mentioned later in this specification refer to the time periods on which the air conditioning system is controlled, which can also be considered the operating periods of the air conditioning system. Before the start of each time period, the control device can determine how the air conditioning system should operate during that time period, obtaining a target control strategy. At the start of that time period, based on the target control strategy, the control device issues corresponding control commands to the air conditioning system to instruct each device in the air conditioning system on the required operating state during that time period, that is, according to what operating parameters each device should operate during that time period. The operating state of each device in the air conditioning system can remain unchanged during a time period, and then be adjusted according to the corresponding needs in the next time period. For example, the duration of this time period can be one hour, half an hour, or other durations.

[0046] For example, the control device can determine, within the current time period, the target control strategy for coordinated control of multiple air conditioning systems in a future time period, which is the period following the current time period. The control device can also determine the target control strategy for the next time period as the current time period is about to end.

[0047] The control equipment determines the target control strategy and controls the air conditioning system at different times. The required operations are similar, and the operations performed by the control equipment at different times can be referenced each other. The following is an example of the operations performed by the control equipment in any given time period.

[0048] Personnel will migrate between different buildings within the target park, resulting in changes in the distribution of personnel in each building (e.g., foot traffic). The migration and distribution of personnel in different buildings are corresponding and interchangeable. Changes in foot traffic within buildings will affect the load on the air conditioning systems deployed in those buildings (i.e., the cooling or heating capacity required by the building). Personnel in buildings are the demanders of ambient temperature, and their bodies are heat sources. In scenarios requiring cooling, if the foot traffic in a building is high, the air conditioning system needs to output a large amount of cooling capacity to achieve a comfortable temperature for humans, resulting in a high cooling load. If the foot traffic in a building is low, the cooling load of the air conditioning system can be relatively small. In fact, when the foot traffic in a building is below a certain level, the air conditioning system can be shut down, eliminating the need to regulate the ambient temperature.

[0049] For example, on a university campus, the cafeteria experiences peak foot traffic half an hour before the first class in the morning, at which point the cafeteria's air conditioning system reaches its peak load. When class begins, people move from the cafeteria to the teaching buildings in large numbers, causing the cafeteria's air conditioning system load to decrease while the teaching buildings' air conditioning system reaches its peak load.

[0050] Control equipment needs to control the air conditioning system based on its load. The load of the air conditioning system in each building within the target area is related to the migration and / or distribution of people between different buildings. Therefore, in order to control the air conditioning system in the future, the control equipment can first determine data representing the migration and / or distribution of people between different buildings in the future (i.e., the predicted pedestrian flow trend data for the target area). Based on this data, the target load of the air conditioning system can be determined more accurately, and then the air conditioning system can be controlled more precisely based on the obtained target load.

[0051] In some embodiments, the control device may utilize a pedestrian flow trend prediction model to determine pedestrian flow trend prediction data for a target area in a future period. The pedestrian flow trend prediction model may be trained based on historical data related to at least one dimension of pedestrian flow in the target area (i.e., second observation data). Accordingly, step 310 may include: obtaining second observation data associated with the target area, the second observation data including data related to at least one dimension of pedestrian flow; and inputting the second observation data into the pre-trained pedestrian flow trend prediction model to obtain pedestrian flow trend prediction data.

[0052] For example, the second observation data includes at least one of the following: pedestrian flow data of different buildings in the target area during historical periods, event scheduling data of different buildings in the target area, meteorological data of the area where the target area is located, or traffic status data of the location of the target area. All of these data can affect the flow of people in the buildings of the target area.

[0053] The historical time period in question refers to the historical period for which the air conditioning control strategy needs to be determined in the future. The pedestrian traffic data for this historical time period includes pedestrian traffic data from at least one historical air conditioning work cycle that coincides with the future period. An air conditioning work cycle refers to the time between the first and second startups of multiple air conditioning systems in the target area. Since the functions of each building in the target area are fixed, the migration and distribution of people between different buildings will follow certain patterns, and the changes in pedestrian traffic data for the same period within similar air conditioning work cycles are usually small. Therefore, the control equipment can acquire pedestrian traffic data from the same period of historical air conditioning work cycles to serve as reference data for the corresponding period in the current air conditioning work cycle.

[0054] Assuming the air conditioning work cycle is one day and a time period is one hour, and the current time period for determining the air conditioning control strategy is the period from 9 am to 10 am, then the pedestrian traffic data of different buildings in the corresponding second observation data can include: pedestrian traffic data from the period from 9 am to 10 am in the previous at least one day (such as the previous three days).

[0055] The current period is also part of the historical period relative to the future period. Since the people in each building gradually migrate and change, the pedestrian flow data in one period will affect the pedestrian flow data in the next period. Therefore, the control equipment can obtain the pedestrian flow data of each building in the current period to determine the pedestrian flow data for the future period based on the pedestrian flow data.

[0056] The events and activities held in various buildings within the target campus significantly influence the movement and / or distribution of people within those buildings. For example, for a university campus, event schedule data might include at least one of the following: student course schedules, exam schedules, lecture schedules, competition schedules, or other event information. For a science and technology park, event schedule data might include at least one of the following: conference bookings, competition schedules, or other event information. When events are scheduled in a particular building (such as a teaching building or an office building), more people will migrate to that building. Therefore, the control equipment can acquire event schedule data from different buildings to determine future pedestrian traffic data. This event schedule data can be uploaded to the control equipment by the event organizer using a terminal, or obtained by the control equipment from the target campus's event management platform.

[0057] Meteorological data for the target park area can include current temperature, humidity, solar radiation, and rainfall. This meteorological data can be obtained based on weather forecasts. The meteorological data for the target park area reflects the environmental conditions outside the buildings within the park and also affects the movement and / or distribution of people within those buildings. When outdoor environmental comfort is low, such as when outdoor temperatures are low or high or during rainfall, people tend to stay indoors, resulting in less movement between buildings and a higher number of people inside. Conversely, when outdoor environmental comfort is high, more people are willing to spend time outdoors, leading to a relatively lower number of people inside buildings.

[0058] Traffic status data for the target park's location can include information reflecting the traffic flow near the park. This traffic status data affects the flow of people from outside the target park into the park. When traffic is congested, there are relatively fewer outsiders in the target park, and consequently, fewer people in the buildings within the park.

[0059] In some embodiments, the building where the air conditioning system is deployed may be equipped with various infrastructures for data acquisition (hereinafter referred to as data acquisition components), and the equipment in the air conditioning system may also have data acquisition or monitoring functions. The control equipment can collect the data required to determine the control strategy of the air conditioning system, i.e., data affecting the operation of the air conditioning system, based on these data acquisition components and the equipment in the air conditioning system. For example, data acquisition components such as pedestrian flow sensors, cameras, turnstiles, and card readers may be installed in the building, and the control equipment can obtain pedestrian flow data in the building based on these data acquisition components. Data inherent to the building itself, data related to human settings (such as transaction scheduling data within the building), and data from outside the building (such as meteorological data and traffic status data) can be uploaded to the control equipment by staff, or obtained by the control equipment from the Internet or certain information management platforms.

[0060] Changes in pedestrian flow are related to various factors and cannot be simply represented by a fixed pattern; they exhibit non-linear characteristics. In some embodiments, non-linear models are used to construct pedestrian flow trend prediction models, such as using machine learning or deep learning methods to fit the pedestrian flow trend prediction model.

[0061] For example, for a target area, data related to pedestrian flow can be accumulated in advance for a period of time. During this data accumulation phase, the control equipment stores the second observation data for each time period in each air conditioning control cycle. After sufficient data accumulation, the model training equipment can use the second observation data for each time period and the actual pedestrian flow data to train the model and obtain a pedestrian flow trend prediction model. The actual pedestrian flow data represents the actual pedestrian migration and / or distribution within the corresponding time period, such as including the actual pedestrian flow at each device in the target area during that time period.

[0062] The model training device can be the aforementioned control device, or it can be another device besides the control device. If the model training device and the control device are different, after the pedestrian flow trend prediction model is trained, the model can be deployed on the model training device. The model training device provides an interface to the control device, which can then invoke the model when needed. Alternatively, the pedestrian flow trend prediction model can also be deployed on the control device, allowing the control device to directly utilize it.

[0063] In some embodiments, the pedestrian flow trend prediction model is a graph neural network (GNN) based model. A graph neural network is a neural network that operates directly on graph-structured data to process the feature information of nodes, edges, or the entire graph. Graph neural networks are used to update the representation of the current node by aggregating the feature information of neighboring nodes, thereby capturing the dependencies and topological features between nodes in the graph. Accordingly, the control device can determine the pedestrian flow trend prediction data in the following manner: the control device can determine the historical pedestrian flow information between different buildings in the target area based on second observation data; construct edges using buildings in the target area as nodes and the historical pedestrian flow information between different buildings to generate a spatiotemporal graph of pedestrian flow; and input the spatiotemporal graph of pedestrian flow into the pedestrian flow trend prediction model to obtain the pedestrian flow trend prediction data.

[0064] The control device needs to determine the migration and distribution information of people among buildings in the future. This information can be reflected by the relationships between buildings related to personnel flow. The target park is naturally a graph structure, where each building can be a node in the graph, and the relationships between buildings can be edges. These relationships may include those arising from personnel flow. In some embodiments, these relationships may also include those formed due to geographical proximity or functional relevance. The control device utilizes a graph neural network to process the graph structure data corresponding to the target park effectively, efficiently inferring the relationships between buildings. Based on these relationships, it can predict the migration and distribution information of people among buildings in the future, obtaining predicted data on future personnel flow trends.

[0065] For example, the data of nodes in the graph structure can include the static attributes of the corresponding buildings, such as the building's functional positioning (office building or laboratory), the maximum number of people the building can accommodate, and the building's geographical location. The control device can determine the historical personnel flow information between different buildings in the target area based on second observation data related to historical personnel flow, and then use this historical personnel flow information as the edge of the graph structure corresponding to the target area, that is, the edge between two nodes. This historical personnel flow information can include the number of people who have historically flowed between any two buildings in the target area. If the relationships between nodes also include relationships formed due to geographical proximity or functional association, the control device can construct three initial edges for each of these three aspects of association: personnel flow, geographical proximity, and function, and then merge these three initial edges to obtain the edges between the nodes.

[0066] In this way, the control equipment can generate a spatiotemporal map of pedestrian flow by constructing edges using buildings in the target area as nodes and historical pedestrian flow information between different buildings. (Personnel flow spatiotemporal map) It can depict the physical topology and functional relationships of the target campus. Here, V represents the node set. Each node Let N represent a building in the target area, and E represent the total number of nodes and the edge set. This spatiotemporal graph of personnel flow is the data of the graph structure corresponding to the target area. This spatiotemporal graph of personnel flow can reflect the relationships between buildings in the target area regarding personnel flow.

[0067] The above scheme considers the strong spatiotemporal coupling caused by the regular flow of people between buildings in the target park. The control equipment utilizes the aforementioned pedestrian flow trend prediction model to model and analyze the entire park as a unified system. This allows for relatively accurate capture of the correlation between buildings and the patterns of pedestrian flow changes between them, thereby obtaining pedestrian flow trend prediction data for future periods and predicting the pedestrian flow in each building during those future periods.

[0068] In some embodiments, the pedestrian flow trend prediction data for a target area within a future time period determined by the control device may include pedestrian flow data for each building in the target area within the future time period. In some embodiments, the pedestrian flow trend prediction data may include pedestrian flow data between buildings within the future time period. The control device may combine this pedestrian flow data with the current pedestrian flow data in each building to obtain the pedestrian flow data for each building within the future time period.

[0069] Step 320: Based on the predicted pedestrian flow data, determine the target load of each of the multiple air conditioning systems deployed in different buildings.

[0070] The load of an air conditioning system in a building is related to the pedestrian traffic within that building. After determining the predicted pedestrian traffic trend data for the target area in the future, the control equipment can determine the pedestrian traffic data in each building based on this predicted pedestrian traffic trend data, and then determine the target load of the air conditioning system in each building based on this pedestrian traffic data.

[0071] In some embodiments, the control device can utilize a load training model to determine the target load of the air conditioning system in each building. The load training model can be trained based on historical load-related data of the air conditioning system in the target area, including pedestrian traffic data and at least one other dimension of data related to the air conditioning load (i.e., first observation data). Accordingly, in step 320, the control device can perform the following steps for each air conditioning system: based on pedestrian traffic trend prediction data, determine the pedestrian traffic data of the building where the air conditioning system is located in the future time period; obtain the first observation data associated with the air conditioning system, the first observation data including data of at least one dimension related to the air conditioning load; and input the pedestrian traffic data and the first observation data into the pre-trained load prediction model to obtain the target load of the air conditioning system.

[0072] In some embodiments, the predicted pedestrian flow trend data for future time periods includes pedestrian flow data for each building. Accordingly, the control device directly searches for pedestrian flow data for the corresponding building within the predicted pedestrian flow trend data for each air conditioning system. In some embodiments, the predicted pedestrian flow trend data is other forms of data characterizing pedestrian movement. The control device can perform certain processing on the predicted pedestrian flow trend data to obtain pedestrian flow data for each building. For example, the predicted pedestrian flow trend data may include 100 people migrating from building A to building B. The control device can subtract 100 from the number of people in building A in the current time period to obtain the pedestrian flow data for building A in the future time period. The control device can also add 100 to the number of people in building B in the current time period to obtain the pedestrian flow data for building B in the future time period.

[0073] Besides the factor of heat dissipation from people (corresponding to pedestrian traffic), the load on an air conditioning system is also affected by other factors. For example, the load on an air conditioning system is also affected by ambient temperature, solar radiation, building ventilation and infiltration, and heat dissipation from electrical equipment in the building. Control equipment can acquire data on these other influencing factors (such as first observation data) to jointly predict the target load of the air conditioning system based on pedestrian traffic data for each building and data on these other influencing factors, thereby improving the accuracy of target load prediction.

[0074] In some embodiments, the first observation data includes at least one of the following: the load of the air conditioning system during a historical period, environmental data of the building where the air conditioning system is located, operating data of the air conditioning system, or meteorological data of the region where the air conditioning system is located. Regarding the meteorological data of the region where the air conditioning system is located, please refer to the relevant introduction to meteorological data in the foregoing description of the second observation data; it will not be repeated here.

[0075] This historical period refers to the historical period for which the air conditioning control strategy needs to be determined in the future. The load of the air conditioning system in the historical period includes the load of the air conditioning system during the same period of the future period in at least one historical air conditioning duty cycle. Please refer to the relevant content in the above introduction to the second observation data for information on this historical period and air conditioning duty cycle. Since the changes in the building's demand for cooling and heating during the same period of similar air conditioning duty cycles are usually small, and these historical data also contain the influence of the building's thermal characteristics and other factors that are not directly measured, the control equipment can obtain the load of the air conditioning system during the same period of the historical air conditioning duty cycle as reference data for the load of the air conditioning system in the corresponding period of the current duty cycle. Assuming that the air conditioning duty cycle is one day, and the duration of a period is one hour, and the future period for which the air conditioning control strategy needs to be determined is the period from 9:00 to 10:00, then the load of the air conditioning system in the historical period in the corresponding first observation data can include: the load during the period from 9:00 to 10:00 in at least one previous day (e.g., the previous three days).

[0076] The current period is also part of the historical timeframe relative to the future. The required temperature of each building typically changes gradually, and there are certain relationships between loads in adjacent periods; for example, the loads in adjacent periods may not differ significantly. Therefore, control equipment can obtain the actual load of the air conditioning system in the current period and use this load as a reference to predict the target load for future periods.

[0077] In some embodiments, the control device can use a first formula to determine the actual load of the air conditioning system at any given time period. Taking a water-cooled air conditioning system as an example, the actual load of the air conditioning system within a time period can be obtained based on the integral of the chilled water inlet and outlet temperature difference and the chilled water flow rate within that time period. The corresponding first formula is: It represents the amount of heat exchanged between time t1 and time t2 (i.e., within any given time period), and is used to represent the load; The mass flow rate at time t is the mass of chilled water passing through the cross-section of the chilled water pipe per unit time. It can be calculated from the flow rate of the chilled water and the density of the water. This indicates the specific heat capacity of water; and These represent the outlet and inlet temperatures of the chilled water at time t, respectively.

[0078] The load on an air conditioning system is influenced by the actual environmental data of the building in which it is located. This environmental data includes at least one of the following: ambient temperature, ambient humidity, or the temperature difference between the indoor temperature and the desired target temperature. For example, in a cooling scenario, the higher the ambient temperature in the building, the higher the required cooling capacity and the higher the load on the air conditioning system. This environmental data can be collected by various sensors (such as temperature sensors, humidity sensors, etc.) deployed in different rooms within the building.

[0079] The operating data of an air conditioning system can include feedback data from various devices within the system. This data reflects the current operating status of the system. For example, the feedback data may include at least one of the actual supply air temperature, actual return air temperature, and supply-return air temperature difference of the terminal devices. This temperature-related data can be acquired using temperature sensors installed on the terminal devices. The feedback data may also include a target supply air temperature set for the terminal devices, such as a temperature manually set by the user through a room temperature control panel, or a temperature set through other control modules.

[0080] For example, flow rate sensors and velocity sensors can be installed on chilled water pipes in air conditioning systems to collect data on chilled water flow rate and velocity. Pressure sensors, frequency sensors, and power sensors can also be installed on water pumps to collect data on water pressure, pump frequency, and power. Wind speed sensors and air volume sensors can also be installed on ventilation ducts to collect data on delivered wind speed and air volume. Valve opening sensors can also be installed on water valves and air valves to collect valve opening data and determine the settings for water and air volume. The data collected by these sensors also constitutes the operating data of the air conditioning system.

[0081] In some embodiments, the data streams collected by various sensors installed in the building can be transmitted to the control device separately, or a single module can integrate the data streams collected by each sensor and transmit them together to the control device.

[0082] In some embodiments, the first observation data may further include time information, such as the current month, time period, whether it is a holiday, and whether it is the first day after a holiday. This time information can reflect the periodic patterns of human activity and building use, which can also affect the load on the air conditioning system. Essentially, this time information influences pedestrian flow. In some embodiments, this time information can also be included in the second observation data to predict more accurate pedestrian flow trends, thereby determining the pedestrian flow data for each building in the future. Thus, the pedestrian flow data carries the influence of this time information, and the target load of the air conditioning system is determined based on this pedestrian flow data, achieving the goal of considering the influence of time information when determining the target load.

[0083] In some embodiments, the first observation data may also include fixed data such as equipment information of the air conditioning system and building data of each building. The equipment information of the air conditioning system may include the rated power of each device in the air conditioning system, such as the rated power of the water-cooled chiller, the rated power of the water pumps (including chilled water pumps and cooling water pumps), the rated power of the cooling tower, the rated power of the air handling unit, and the rated power of the fan coil unit. This equipment information may also include the cooling capacity provided by the water-cooled chiller, the number of water-cooled chillers, and the number of water pumps. Building data may include its structural data, shape coefficient, orientation of each room, and window-to-wall ratio, etc.

[0084] The load conditions of air conditioning systems are related to various factors such as weather, human activities, and economic activities, and cannot be simply characterized by a fixed law; they exhibit nonlinear characteristics. In some embodiments, nonlinear models can be used to construct load forecasting models, such as using machine learning or deep learning methods to fit the load forecasting model.

[0085] For example, for a target industrial park, load-related data can be accumulated in advance for a period of time. During this data accumulation phase, the first observation data, actual pedestrian flow data, and actual load for each time period in each air conditioning control cycle are stored. After sufficient data accumulation, the model training equipment can use the first observation data, actual pedestrian flow data, and actual load for each time period to train the model and obtain a load prediction model. For details on training and deploying the load prediction model, please refer to the aforementioned introduction to pedestrian flow trend prediction models; further details will not be elaborated here.

[0086] In some embodiments, since the functions, structures, configurations, and requirements of different buildings may differ, the required heating and cooling loads of different buildings may also differ under the same observation data. Therefore, corresponding load forecasting models can be constructed for the air conditioning systems of each building in the target park. For each air conditioning system, its corresponding load forecasting model is used to predict the load for future periods.

[0087] In some embodiments, a unified load forecasting model can be constructed for different buildings. In this approach, load-related data from a large number of buildings can be acquired for model training to obtain the load forecasting model. When forecasting the load of the air conditioning system in any building, the data input to the load forecasting model includes not only the aforementioned pedestrian traffic data and the first observation data, but also data about the building itself. This data may include information about the building's function, structure, configuration, and requirements.

[0088] The load changes of various buildings within a park are not independent, especially those caused by heat dissipation from people, exhibiting strong spatiotemporal coupling. Traditional solutions determine the load only for the air conditioning system of a single building, ignoring the correlation of pedestrian flow changes between different buildings in the park. This makes it impossible to predict sudden changes in pedestrian flow across buildings, resulting in low load determination accuracy and consequently, poor control strategies for the air conditioning system. In the above embodiment, the control equipment uses a pedestrian flow trend prediction model to mine the migration patterns of pedestrian flow between buildings in the target park, determining the predicted pedestrian flow trend data for the target park in the future. Then, based on this pedestrian flow trend prediction data, the load of the air conditioning system in the future is inferred using a load prediction model; that is, the energy required from the air conditioning system to achieve the desired indoor temperature in the current state. Thus, this load prediction considers the dynamic characteristics of the load caused by pedestrian flow at the park level, resulting in higher accuracy in predicting the load for the future period.

[0089] In the above embodiments, the changes in pedestrian flow trends within the park are predicted first, followed by the prediction of the air conditioning system load corresponding to different buildings. Instead of directly simulating and determining the load on the buildings within the park, this approach improves modeling efficiency and prediction accuracy, and is easier to implement. Alternatively, based on actual needs, the pedestrian flow trend prediction model and the load prediction model can be used separately.

[0090] Step 330: Based on the target load of each of the multiple air conditioning systems, and taking the multiple air conditioning systems as a whole, jointly plan the operating parameters of the multiple air conditioning systems to obtain the target control strategy for coordinated control of the multiple air conditioning systems.

[0091] The control equipment determines the target load for each air conditioning system, which represents the demand for the operation of the air conditioning system. Based on the target load, the control equipment can then determine the operating parameters of each air conditioning system, i.e., determine how to control each system, thus obtaining a corresponding target control strategy. The control equipment treats multiple air conditioning systems in the target area as a whole, jointly planning their operating parameters. This means that when determining the operating parameters of each air conditioning system, the operating parameters of other air conditioning systems are also considered to arrive at a more optimal target control strategy for the entire target area. Based on this target control strategy, the control equipment performs coordinated control of these multiple air conditioning systems, improving the overall energy efficiency of the air conditioning systems in the target area.

[0092] The control equipment must ensure that each air conditioning system operates to meet the minimum environmental comfort requirements (i.e., the output energy of each air conditioning system is not less than the target load). Based on this, the control equipment considers how to set the operating parameters of each air conditioning system to optimize its operation. The output energy of the air conditioning system can include the output energy of the main unit. In cooling mode, the output energy of the main unit is the cooling capacity; in heating mode, the output energy is the heating capacity. If the air conditioning system also includes heat dissipation equipment (such as a cooling tower), then the output energy of the air conditioning system can also include the heat dissipation of that equipment.

[0093] In some embodiments, the control device can consider how to minimize the overall energy consumption of multiple air conditioning systems in a target area while maximizing the overall environmental comfort of each building (corresponding to a higher overall energy output from the multiple air conditioning systems). This allows for joint planning of the operating parameters of the multiple air conditioning systems, resulting in a target control strategy. For example, for buildings with low foot traffic, the demand for environmental comfort is lower. The control device can reduce the energy output of the air conditioning systems in these buildings, while allowing the air conditioning systems in buildings with high foot traffic to output more energy. Thus, coordinated control of the air conditioning systems can minimize the overall energy consumption of the air conditioning systems in the area while ensuring greater environmental comfort for more people, thereby improving overall energy efficiency.

[0094] In some embodiments, the target control strategy includes at least one of the following: the start-up time, runtime, or equipment operating parameters of multiple air conditioning systems. In this approach, the control device also uses the start-up time and runtime of the air conditioning systems in each building as variables for joint optimization. Each air conditioning system can start at its corresponding start-up time according to the needs of its respective building, rather than all starting at the same fixed start-up time. This reduces the risk of some buildings requiring the target temperature later, leading to energy waste; and reduces the risk of some buildings requiring the target temperature earlier, but the air conditioning systems starting later, resulting in substandard user comfort. This approach improves the start-up flexibility of the air conditioning systems, enhances the energy efficiency of the air conditioning systems in the target area during start-up, and allows the air conditioning systems to better meet the building's needs.

[0095] For example, the control device searches among multiple possible control strategies based on the previously determined target load of each air conditioning system in the future time period to obtain a target control strategy. Each control strategy includes a combination of operating parameters for each air conditioning system, which may include at least one of the following: the start-up time, operating duration, or equipment operating parameters of each air conditioning system. The target control strategy is the control strategy that minimizes the overall energy consumption of the multiple air conditioning systems while ensuring that the output energy of each air conditioning system is not less than the corresponding target load, among the multiple possible control strategies.

[0096] The aforementioned runtime can refer to the total duration of the air conditioning system from the start-up time to the shutdown time. In some embodiments, the air conditioning system runs continuously from the start-up time to the shutdown time, and the shutdown time is fixed (e.g., a technology park stipulates that the air conditioning system will be shut down uniformly at 9 PM). In this case, the start-up time and runtime of the air conditioning system in the target control strategy have essentially the same meaning, and the target control strategy may include only one of the start-up time and runtime. If the target control strategy includes the start-up time, the runtime can be obtained based on the start-up time and the fixed shutdown time. If the target control strategy includes the runtime, the start-up time can be obtained based on the runtime and the fixed shutdown time. In some embodiments, the shutdown time of the air conditioning system is not fixed. In this case, the target control strategy determined by the control device may include both the start-up time and the runtime of the air conditioning system.

[0097] An air conditioning system comprises multiple devices, such as the main unit, cooling system, and water pump. The operating status of each device in the air conditioning system can remain constant during a given period. For example, devices that need to be turned on will remain on throughout the period, and devices that need to be turned off will remain off. Accordingly, the previously determined start-up time and operating duration of the air conditioning system also correspond to the start-up time and operating duration of the devices that need to be turned on within the system.

[0098] The operating parameters of an air conditioning system can include parameters that indicate how multiple devices in the system should operate. Taking a water-cooled air conditioning system as an example, the operating parameters can include the on / off status (i.e., whether it is on) of each water-cooled unit, the outlet temperature of the chilled water in the water-cooled unit, the number of cooling towers in operation, and the on / off status and frequency of each water pump.

[0099] In some embodiments, the on / off state of each device in the air conditioning system can change during a period of time. For example, a device that has just been turned on may be turned off at a certain moment in the middle of the period, and a device that has just been turned off may be turned on at a certain moment in the middle of the period. In this approach, the target control strategy may also include the start-up time and running duration of each device in the air conditioning system, so as to achieve more refined control over each device in the air conditioning system.

[0100] Traditional air conditioning systems rely on manual experience or fixed schedules to start and stop, ensuring the system is turned on at fixed times. The operating parameters during the steady-state operation are then optimized, without considering the start-up time and runtime as optimization variables. This approach lacks foresight regarding future loads and offers low control precision.

[0101] In the above embodiments, the target load of each air conditioning system in the target park is predicted in the future time period, and the start-up time and runtime of the air conditioning system are included in the optimization variables. Based on the target load of each air conditioning system in the target park, the operation of each air conditioning system is optimized collaboratively to determine the appropriate start-up time, runtime, and equipment operating parameters for each air conditioning system. That is, it determines when and with which equipment combination each air conditioning system should be turned on to accurately match the dynamic load demand of the building. In this way, there is no need to set a fixed start-up time. The entire process of the air conditioning system from shutdown to meeting the temperature requirements is optimized as a whole, guiding each air conditioning system to start at the optimal time. This method can reduce the risk of resource waste caused by starting too early or insufficient environmental comfort caused by starting too late, can precisely meet the dynamic needs of the system, and improve energy saving and park management efficiency.

[0102] In some embodiments, in step 330, the control device can obtain the air conditioning usage requirements of the target park, including the target temperature that each building needs to reach at the target time; and with the constraint that the output energy of each air conditioning system is not lower than its respective target load, and with the goal that the output energy of each air conditioning system supports the corresponding building to reach the target temperature at the target time, the operating parameters of multiple air conditioning systems are jointly planned to obtain the target control strategy.

[0103] The target temperatures required for different buildings within the target area, and the target times for achieving those temperatures, may differ. For example, on a certain day, people in the target area might need to work in Building A at 9:00 AM and move from Building A to Building B for an event at 10:00 AM. This scenario might require the indoor temperature in Building A to reach 26 degrees Celsius at 9:00 AM and the indoor temperature in Building B to reach 26 degrees Celsius at 10:00 AM. Control equipment within the target area can acquire the air conditioning usage requirements of the target area to obtain information about the target temperature and target time. This air conditioning usage requirement can be uploaded to the control equipment by staff in the target area, or it can be inferred by the control equipment based on the activity schedule, current environment, or historical air conditioning usage requirements within the target area.

[0104] Due to structural limitations, air conditioning systems have a maximum output energy. The actual output energy of an air conditioning system is related to its operating conditions (such as operating parameters and start-up time), and may also be related to its own structure and the actual environment. Control equipment can determine the output energy that the air conditioning system can provide under different operating parameters. The relationship between the output energy and its operating parameters may differ for different air conditioning systems. In some embodiments, a corresponding output energy prediction model is pre-built for each air conditioning system. The output energy model can be a nonlinear model such as machine learning or deep learning. The output energy prediction model can be trained based on the operating parameters, environmental data, and actual output energy during the historical operating periods of the air conditioning system. The control equipment uses this output energy prediction model to predict the output energy of the air conditioning system operating with certain parameters in future periods.

[0105] In the above embodiments, for each air conditioning system, its output energy must not be lower than its target load, and its operation must meet the air conditioning needs of the target area. In other words, the output energy of each air conditioning system should support the corresponding building to reach the target temperature at the target time. To achieve this goal, the control equipment plans the operating parameters of the air conditioning system, determining how it should operate to ensure the building's ambient temperature reaches the target temperature precisely at the target time. This ensures the air conditioning system operates precisely to meet the air conditioning needs, reducing the risk of energy waste due to prematurely meeting the needs, and reducing the risk of insufficient environmental comfort due to delayed meeting the needs, where the building's ambient temperature has not yet reached the target temperature. This improves the energy efficiency of the air conditioning system.

[0106] In this planning process, the control equipment can take all air conditioning systems into account holistically. For example, while ensuring the operation of the air conditioning systems meets the aforementioned target load and air conditioning usage requirements, the control equipment should strive to minimize the overall energy consumption of each air conditioning system and maximize the overall environmental comfort in all buildings within the target area. Based on this, the operating parameters of each air conditioning system can be determined, leading to the target control strategy.

[0107] Based on the air conditioning usage requirements, the target control strategy determined by the control equipment can include at least one of the following: the start-up time or the operating duration of the air conditioning system. The following example assumes the air conditioning system's shutdown time is fixed, and the target control strategy only includes the start-up time. Starting the air conditioning system at this time ensures that the ambient temperature in the building reaches the target temperature precisely at the desired time. This approach ensures the air conditioning system operates precisely to meet the usage requirements, reducing the risk of energy waste due to starting too early and reaching the target temperature prematurely, and reducing the risk of insufficient environmental comfort due to starting too late and the ambient temperature not reaching the target temperature by the desired time. This improves the energy efficiency of the air conditioning system.

[0108] The actual air conditioning usage needs of a target park at different times are related to the actual activities of people within the park and are dynamically changing information, making them difficult to predict in advance. In some scenarios, air conditioning usage needs are vague and do not include precise target temperature or energy consumption requirements. Air conditioning usage needs also vary depending on the park's operational scenarios. For example, when hosting important events, it is necessary to prioritize maintaining a lower indoor temperature (requiring a larger output of the air conditioning system). In this case, minimizing energy consumption is no longer the sole objective; rather, the goal is to minimize both energy consumption and indoor temperature. When electricity prices are low, park managers may be willing to accept slightly higher energy consumption for a better environmental comfort within the park. For these situations, using traditional single-objective optimization methods, such as focusing solely on minimizing energy consumption, will fail to determine a suitable control strategy.

[0109] In some embodiments, the control device can set multiple objectives for multiple air conditioning systems in a target area, and determine better candidate control strategies that balance these objectives under different air conditioning usage demands. Then, based on the air conditioning usage demands in future time periods and these candidate control strategies, a matching target control strategy is determined. For example, one objective might be to maximize the total output energy of the multiple air conditioning systems, while another objective might be to minimize the total energy consumption of the multiple air conditioning systems.

[0110] Accordingly, in step 330, the control device, constrained by ensuring that the output energy of each air conditioning system is not lower than its respective target load, and aiming to maximize the total output energy of multiple air conditioning systems and minimize the total energy consumption of multiple air conditioning systems, jointly plans the operating parameters of multiple air conditioning systems to obtain multiple candidate control strategies. Each candidate control strategy includes: a total output energy, a total energy consumption, and control instructions that need to be executed for multiple air conditioning systems to achieve the total output energy and total energy consumption; and, based on the target park's demand information for total output energy and total energy consumption in the future time period, determines the target control strategy from the multiple candidate control strategies. The total output energy and total energy consumption in each candidate control strategy jointly represent a possible air conditioning usage demand. The control instructions in the candidate control strategy represent how each air conditioning system should operate to meet the air conditioning usage demand; these control instructions may include the operating parameters of each device in each air conditioning system.

[0111] The output energy of an air conditioning system corresponds to the temperature achievable within a building, thus affecting the environmental comfort. In some embodiments, the control device can convert the output energy of the air conditioning system into the achievable temperature within the building, and determine a target control strategy based on the overall temperature demand and total energy consumption of the target area. In the above scheme, the control device considers the total output energy and total energy consumption of the air conditioning systems in the target area as a whole (or, in other words, the overall environmental comfort and total energy consumption of the target area), thereby considering multiple air conditioning systems in the target area as a whole. In the above embodiments, the control device can determine better candidate control strategies for different combinations of total output energy and total energy consumption requirements for multiple air conditioning systems in the target area. Based on actual needs in the future, a suitable target control strategy can be selected from these multiple candidate control strategies to achieve diversified control of the air conditioning system under different air conditioning usage requirements.

[0112] The process of determining candidate control strategies by the control equipment can be abstracted as an optimization problem. The optimization variable in this problem can be a vector containing the start-up times and operating parameters of each air conditioning system in the target area. This optimization problem requires optimization for multiple objectives, and correspondingly uses multiple objective functions, each corresponding to an optimization objective. The optimization problem needs to balance the total output energy (corresponding to the overall comfort of the area) and total energy consumption of each air conditioning system to maximize the total output energy and minimize the total energy consumption. Accordingly, two objective functions can be used: one to determine the total output energy of each air conditioning system, and the other to determine the total energy consumption of each air conditioning system. The constraint in this optimization problem is that the output energy of each air conditioning system must meet the building's energy demand; that is, the output energy of each air conditioning system must be greater than or equal to the corresponding target load. By solving this optimization problem, the control equipment obtains a Pareto solution set that balances the total output energy and total energy consumption of each air conditioning system.

[0113] Air conditioning systems, due to their structural limitations, have a maximum energy consumption. The actual energy consumption of an air conditioning system is related to its operating parameters, and may also be related to its own structure and the actual environment. Control equipment can determine the energy consumption of the air conditioning system under different operating conditions. The relationship between the energy consumption and operating conditions may differ for different air conditioning systems. In some embodiments, corresponding energy consumption prediction models are pre-built for each air conditioning system in the target area. The energy consumption model can be a nonlinear model such as machine learning or deep learning. The energy consumption prediction model can be trained based on the operating parameters, environmental data, and actual energy consumption of the air conditioning system during historical working periods. The actual energy consumption in each working period can be derived based on the changes in electricity consumption measured during that working period. By using this energy consumption prediction model, the control equipment can predict the energy consumption required for the air conditioning system to operate according to certain operating parameters in future periods.

[0114] In some embodiments, the control device can predict the output energy and energy consumption of each air conditioning system operating under a possible operating condition (including operating parameters and start-up time) in the target area, using an output energy prediction model and an energy consumption prediction model built for that air conditioning system. Summing the output energy of each air conditioning system in the target area yields the total output energy of each system; summing the energy consumption of each system yields the total energy consumption of each system. This total output energy and total energy consumption are predicted; for ease of description, the total output energy and total energy consumption of each air conditioning system based on the predictions will be referred to below as the total predicted output energy and total predicted energy consumption, respectively.

[0115] In some embodiments, the control device may also determine the overall indoor temperature achievable by the target area based on the total predicted output energy, and use this overall indoor temperature instead of the total predicted output energy for subsequent processing. For example, the control device may determine the indoor temperature that the air conditioning system can achieve in the building based on the predicted output energy of each air conditioning system and the environmental data of the building where the air conditioning system is located. Then, the overall indoor temperature of the target area is obtained by directly averaging or weighted averaging the indoor temperatures achievable by each building.

[0116] In some embodiments, the control device determines the multiple candidate control strategies by generating a Pareto solution set. For example, the control device uses the control commands of multiple air conditioning systems in a target area as variables, with the constraint that the output energy of each air conditioning system is not less than its respective target load, and the objectives being to maximize the total output energy of the multiple air conditioning systems and minimize the total energy consumption of the multiple air conditioning systems. It then determines a Pareto solution set that balances total output energy and total energy consumption, with each solution in the Pareto solution set corresponding to a candidate control strategy. For instance, the control device can determine this Pareto solution set based on a multi-objective optimization framework using Non-Dominated Sorting Genetic Algorithm II (NSGA-II). NSGA-II is an efficient global optimization algorithm suitable for handling complex optimization problems with conflicting objectives.

[0117] In some embodiments, the control device may perform steps one through four below to determine a plurality of candidate control strategies.

[0118] Step 1: The control device can use the control commands randomly generated for multiple air conditioning systems as individuals in the population to obtain the initial parent population.

[0119] In genetic algorithms, an initial parent population is first randomly generated, consisting of multiple individuals (i.e., chromosomes). Each individual corresponds to a solution to the aforementioned optimization problem, and the initial parent population is thus an initial set of solutions. Each individual includes a control command for the multiple air conditioning systems. The control command included in the individual is obtained by combining the operating parameters of each air conditioning system. In some embodiments, each air conditioning system may include multiple devices (such as a main unit, a heat dissipation unit, and a water pump unit), and the operating parameters of each air conditioning system include the operating parameters of each of its devices. Accordingly, each individual includes the operating parameters of each device in each air conditioning system. By adjusting the individuals in this way, the operating parameters of multiple air conditioning systems can be adjusted as a whole, achieving joint planning of the operating parameters of multiple air conditioning systems.

[0120] The operating parameters of each air conditioning system can be encoded using a hybrid encoding method. For example, for each air conditioning system, binary bits represent the on / off status of the equipment, while real bits represent other specific parameter settings of the equipment. The control device can randomly generate an initial parent population P(0) containing N individuals. Let P(t) represent the current parent population, where t is the current iteration generation, and t=0 for the initial parent population.

[0121] Figure 4 A schematic diagram of the encoding structure of an individual according to an embodiment of this specification is shown. Figure 4 As shown, assume the target area comprises two air conditioning systems deployed in two separate buildings. One air conditioning system includes two main units (e.g., water-cooled chillers), two chilled water pumps, two cooling water pumps, and two cooling devices (e.g., cooling towers); the other air conditioning system includes two main units, two chilled water pumps, three cooling water pumps, and three cooling devices. The operating parameters for each main unit include its on / off status and outlet water temperature; the operating parameters for each water pump include its on / off status and operating frequency; and the operating parameters for each cooling device include its on / off status. Combining the operating parameters of each device yields an individual from the population.

[0122] Each device within each type of equipment can be labeled sequentially using identifiers such as "1#", "2#", etc. For example... Figure 4 As shown, the operating parameters of each device can be mathematically expressed as follows. The on / off state of each device is represented by 0 or 1, where 0 represents the device is off and 1 represents the device is on. Assume the outlet water temperature of the main unit is set in the range [7, 10] degrees Celsius. To reduce the solution space, the outlet water temperature can be converted into a discrete value, such as a step size of 0.1 degrees Celsius, or 0.2 degrees Celsius, or other step sizes. The operating frequency of the water pump can be set in the range [40, 50] hertz (Hz). To reduce the search space, the frequency is also converted into a discrete value, such as a step size of 1 hertz, or 2 hertz, or other step sizes.

[0123] Figure 4 This illustrates what an individual might contain, which consists of operating parameters from each device in two air conditioning systems. Figure 4Taking the sequential arrangement of the operating parameters of the main air conditioning units, chilled water pumps, cooling pumps, and cooling equipment in two air conditioning systems as an example, the operating parameters of the first air conditioning system corresponding to this unit include: Main air conditioning unit 1 is on, with an outlet water temperature of 8.5 degrees Celsius, and other main air conditioning units are off; chilled water pump 1 is on, operating at a frequency of 50 Hz, and other chilled water pumps are off; cooling pump 2 is on, operating at a frequency of 45 Hz, and other cooling pumps are off; and cooling equipment 1 is on, and other cooling equipment is off. The operating parameters of the second air conditioning system corresponding to this unit include: Main air conditioning unit 1 is on, with an outlet water temperature of 7.5 degrees Celsius, and other main air conditioning units are off; chilled water pump 1 is on, operating at a frequency of 47 Hz, and other chilled water pumps are off; cooling pump 1 is on, operating at a frequency of 40 Hz, and other cooling pumps are off; and cooling equipment 2 is on, and other cooling equipment is off.

[0124] In some embodiments, each individual also includes at least one of the start-up time or runtime of each air conditioning system. The start-up time or runtime can be represented by real numbers. For example, at least one of the start-up time or runtime of each air conditioning system can be added before the operating parameters of each device in the air conditioning system.

[0125] Step 2: Based on the target load set and output energy set corresponding to each individual in the initial parent population, determine the non-dominated level and congestion distance of each individual. The target load set includes the target load of each of the multiple air conditioning systems, and the output energy set includes the output energy of each of the multiple air conditioning systems.

[0126] In some embodiments, for each individual in the initial parent population, the control device can predict the output energy of the air conditioning system based on the operating parameters and operating time of each air conditioning system encoded by that individual, using the output energy prediction model of that air conditioning system, thereby obtaining the output energy of each of the multiple air conditioning systems corresponding to that individual.

[0127] The target load of an air conditioning system serves as the minimum constraint on its operation. The output energy of the air conditioning system characterizes its actual operation based on the coded operating parameters within each individual system. The actual operation of the air conditioning system must meet this minimum constraint to satisfy the operational requirements. If any air conditioning system operates based on the coded operating parameters within its individual system and its output energy is less than the target load, it indicates that the system does not meet the constraints. Operating an air conditioning system based on these coded operating parameters will fail to meet the building's air conditioning needs, and the system may be prohibited from operating based on these parameters.

[0128] For each individual and each corresponding air conditioning system, the control equipment can construct constraints based on the target load of that air conditioning system to calibrate the individual. For example, by comparing the output energy of each air conditioning system corresponding to the individual with its target load, it can determine whether the coded operating parameters of that air conditioning system within the individual are appropriate, and consequently, whether the individual is optimal. For instance, based on the calibration results, the non-dominated level and congestion distance of the individual can be determined to characterize the individual's quality, that is, the appropriateness of the operating parameters of each coded air conditioning system within the individual.

[0129] Undominated hierarchy and crowding distance are two concepts in multi-objective optimization problems. We will first explain undominated hierarchy and crowding distance.

[0130] For a multi-objective optimization problem, solution a is considered to dominate solution b if solution a is no worse than solution b on all objectives and solution a is strictly better than solution b on at least one objective. In other cases, if solution a is better than solution b on one objective and solution b is better than solution a on another objective, then solutions a and b are not mutually dominant and belong to a pair of non-dominated solutions. In step two, each individual is assigned a solution to the optimization problem, and the dominance relationship between individuals is described above in the section on the dominance relationship between solutions a and b.

[0131] A non-dominated level includes solutions that are not dominated by other solutions. Solutions in lower non-dominated levels (those partitioned later in the non-dominated hierarchy, with larger level numbers) are dominated by solutions in higher non-dominated levels (those partitioned earlier in the non-dominated hierarchy, with smaller level numbers). For example, a solution in the second non-dominated level is dominated by a solution in the first non-dominated level. When selecting solutions from multiple non-dominated levels, solutions in higher non-dominated levels are preferred, as these solutions are generally better than solutions in lower non-dominated levels across all objectives.

[0132] Crowding distance measures the degree of crowding between a solution and other solutions within the same non-dominated hierarchy; that is, the density of solutions surrounding the given solution. For a specific solution within a non-dominated hierarchy, its crowding distance is half the perimeter of a rectangle surrounding it. The solutions in this non-dominated hierarchy are ordered based on a certain objective function, and this rectangle is determined by two solutions adjacent to the specific solution and located in the same non-dominated hierarchy. When selecting a subset of solutions from the same non-dominated hierarchy, solutions with larger crowding distances are preferred. These solutions with larger crowding distances are located in sparser regions of the solution space, and selecting them helps reduce the risk of the optimization problem's solution getting trapped in a localized region of superiority, thus obtaining a more evenly distributed solution set.

[0133] In the above embodiments, the two objectives in the optimization problem are the total predicted output energy and the total predicted energy consumption for each individual. The determination of the total predicted output energy and total predicted energy consumption can be referred to the foregoing description, and will not be repeated here. In some embodiments, the control device performs a fast non-dominated sorting on all individuals in the initial parent population P(0) based on the total predicted output energy and total predicted energy consumption for each individual. Higher total predicted output energy is better, and lower total predicted energy consumption is better. For any two individuals (e.g., individuals c and d), if individual c is better at one objective and individual d is better at the other, then individuals c and d are not dominated by each other. If individual c is better at both objectives, then individual c dominates individual d. For example, if the total predicted output energy of individual c is greater than that of individual d, but the total predicted energy consumption of individual d is less than that of individual c, then individuals c and d are not dominated by each other. For example, if the total predicted output energy of individual c is greater than that of individual d, and the total predicted energy consumption of individual c is less than that of individual d, then individual c dominates individual d.

[0134] Based on the aforementioned fast non-dominated sorting, the control device first identifies all individuals (non-dominated solutions) in the initial parent population P(0) that are not dominated by any other individuals, and assigns them to the first non-dominated level. Then, among the remaining individuals, it again identifies all individuals that are not dominated by any other individuals and assigns them to the second non-dominated level. This process of identifying all individuals that are not dominated by any other individuals and assigning them to the next non-dominated level is repeated until all individuals in the initial parent population P(0) are assigned to the corresponding non-dominated level. The crowding distance of individuals in a non-dominated level can be determined using the method described above, and will not be described in detail here.

[0135] The above provides an example of determining the non-dominated hierarchy and crowding distance of an individual in the initial parent population P(0) without considering constraint verification. For ease of distinction, the non-dominated hierarchy and crowding distance determined in this case will be referred to as the initial non-dominated hierarchy and the initial crowding distance, respectively.

[0136] The following section describes how the control equipment verifies an individual and determines the individual's final non-dominated level and crowding distance based on the verification results.

[0137] The control equipment verifies the output energy of each individual's corresponding air conditioning system using its target load. The result falls into two categories: If the individual fails the verification, the output energy of at least one of the individual's corresponding air conditioning systems is less than its target load. If the individual passes the verification, the output energy of all the individual's corresponding air conditioning systems is not less than its target load. For these two cases, the control equipment can use different methods to determine the individual's non-dominated level and congestion distance.

[0138] In some embodiments, if the output energy of any air conditioning system corresponding to an individual is less than its target load, the control device can predict the total predicted output energy and total predicted energy consumption corresponding to that individual, and determine the individual's non-dominated level and congestion distance based on the total predicted output energy, the total predicted energy consumption, and a preset penalty term. The method for determining the total predicted output energy and total predicted energy consumption can be referred to in the preceding related descriptions, and will not be repeated here. To distinguish it from other cases, the non-dominated level and congestion distance determined in this case will be referred to as the penalized non-dominated level and penalized congestion distance, respectively.

[0139] If the output energy of any air conditioning system corresponding to an individual is less than its target load, it indicates that the individual does not meet the constraints and fails the verification. In this case, the control device applies a penalty to the individual based on a preset penalty term, resulting in a very low non-dominated level and a very low crowding distance for that individual. Subsequently, in the evolutionary process, this individual is eliminated based on the penalty non-dominated level and the penalty crowding distance, guiding the population to evolve within a feasible range.

[0140] In some embodiments, the control device can predict the total predicted output energy and total predicted energy consumption of multiple air conditioning systems corresponding to each individual in the initial parent population. Based on the total predicted output energy and total predicted energy consumption of each individual, the initial non-dominated level and initial congestion distance of each individual are determined. Then, each individual is verified based on the output energy and target load of each air conditioning system corresponding to each individual. For individuals that fail the verification, the control device adjusts the initial non-dominated level and initial congestion distance of that individual based on a preset penalty term to obtain a penalized non-dominated level and penalized congestion distance. The method for determining the initial non-dominated level and initial congestion distance can be referred to the relevant description above, and will not be repeated here.

[0141] For example, the preset penalty items include a preset penalty level number and a preset penalty distance (such as a large distance value). The control device can reduce the preset penalty level number based on the individual's initial non-dominated level (e.g., reducing the first non-dominated level to the third non-dominated level) to obtain the individual's penalty non-dominated level. The control device can subtract the preset penalty distance from the individual's initial crowding distance to obtain the individual's penalty crowding distance.

[0142] In some embodiments, the preset penalty items include a preset non-dominated level (a lower level) and a preset congestion distance (a smaller distance). The control device can directly use the preset non-dominated level as the penalty non-dominated level for an individual and the preset congestion distance as the penalty congestion distance for that individual. In this embodiment, the control device can first verify each individual based on the output energy and target load of each air conditioning system corresponding to each individual. For individuals that fail the verification, it is not necessary to determine their total predicted output energy and total predicted energy consumption; the corresponding penalty non-dominated level and penalty congestion distance are directly determined based on the preset penalty items. For individuals that pass the verification, non-dominated ranking and congestion calculation are then performed based on the total predicted output energy and total predicted energy consumption to determine the corresponding penalty non-dominated level and penalty congestion distance.

[0143] In some embodiments, if the output energy of multiple air conditioning systems corresponding to an individual is not less than their respective target loads, the total predicted output energy and total predicted energy consumption of the individual are predicted, and the non-dominated level and congestion distance of the individual are determined based on the total predicted output energy and total predicted energy consumption. The fact that the output energy of multiple air conditioning systems corresponding to an individual is not less than their respective target loads indicates that the individual meets the constraints, i.e., the individual passes the verification. Furthermore, the control device can use the individual's own non-dominated level and congestion distance (i.e., the aforementioned initial non-dominated level and initial congestion distance) as the individual's final non-dominated level and congestion distance. For the method of predicting the total predicted output energy and total predicted energy consumption of an individual, and determining the individual's non-dominated level and congestion distance based on the total predicted output energy and total predicted energy consumption, please refer to the aforementioned relevant descriptions, which will not be repeated here.

[0144] Step 3: Based on the non-dominated hierarchy and crowding distance of each individual, the initial parent population is multi-generationally reproduced to obtain the target parent population.

[0145] In some embodiments, the process of each generation of reproduction includes the following steps s1 to s3.

[0146] Step s1: Based on the non-dominated hierarchy and crowding distance of each individual in the current parent population, select parent individuals multiple times from the current parent population, and perform at least one of the crossover or mutation operations on each selected parent individual to generate multiple offspring individuals, thus obtaining the current offspring population.

[0147] The control device selects a parent individual from the current parent population and performs crossover and / or mutation operations to generate at least one offspring individual. Multiple selections and executions of crossover (i.e., combination) and / or mutation (i.e., fine-tuning) operations can generate the desired number of offspring individuals. For example, a target number of offspring individuals can be preset, and the process of selecting and performing crossover and / or mutation operations can be repeated multiple times based on this target number until the target number of offspring individuals is obtained. In some embodiments, the target number can be the same as the number of individuals in the parent population.

[0148] In some embodiments, the control device employs a binary tournament selection method, selecting superior parent individuals from the current parent population based on each individual's non-dominated level and crowding distance to generate offspring individuals. The control device preferentially selects individuals with higher non-dominated levels. For individuals with the same non-dominated level, since individuals with larger crowding distances contribute to better population diversity, the control device preferentially selects individuals with larger crowding distances.

[0149] For example, the control device can randomly select two individuals from the current parent population and compare them based on their non-dominated level and crowding distance, selecting the superior individual as the parent individual for generating offspring. During the comparison, the control device first compares the non-dominated levels of the two individuals; the individual with the higher non-dominated level is superior to the individual with the lower non-dominated level. If the non-dominated levels of the two individuals are the same, the control device compares their crowding distance; the individual with the larger crowding distance is superior to the individual with the smaller crowding distance. Finally, the control device selects the superior individual determined by the comparison. In the next parent individual selection, the control device again performs the above process, that is, randomly selecting two individuals from the current parent population for comparison. This random selection of individuals is a selection with replacement; the same individual in the current parent population may be selected multiple times. If the individual is superior, it can be selected multiple times as the parent individual for generating offspring. Using athletes as an analogy for individuals in a population, the same individual is repeatedly selected as the parent individual for generating offspring, similar to an excellent athlete participating in multiple auditions and advancing multiple times.

[0150] In some embodiments, each air conditioning system includes multiple devices, and each individual includes the operating parameters of each device in each air conditioning system. Cross-operations performed on parent individuals include exchanging the operating parameters of the same devices in different parent individuals.

[0151] Each individual entity includes operating parameters for multiple air conditioning systems, and the operating parameters of each air conditioning system include operating parameters for multiple devices within that system. Different air conditioning systems may contain different devices, and even within the same air conditioning system, different devices may have different types of operating parameters; therefore, operating parameters from different devices cannot be used interchangeably. Thus, cross-operation must be performed on the operating parameters of the same devices, that is, exchanging the operating parameters of the same devices in different parent entities. In some embodiments, the control device adjusts the parent entity based on cross-operation, enabling the exchange of operating parameters at the granularity of a single air conditioning system corresponding to the individual entity; or at the granularity of the devices included in a single air conditioning system corresponding to the individual entity.

[0152] Assuming, Figure 4 The individual shown serves as the parent individual used to generate offspring individuals. The control device needs to perform cross-operations on this parent individual. For example, the control device can exchange the operating parameters of the first air conditioning system in this parent individual with the operating parameters of the first air conditioning system in another parent individual. The control device can also exchange the operating parameters of the main unit in the first air conditioning system of this parent individual with the operating parameters of the main unit in the first air conditioning system of another parent individual. Similarly, the control device can exchange the operating parameters of the chilled pump and cooling pump in the second air conditioning system of this parent individual with the operating parameters of the chilled pump and cooling pump in the second air conditioning system of another parent individual. Other methods can also be used to perform this cross-operation, which will not be listed here.

[0153] In some embodiments, the mutation operation performed on the parent individual includes: randomly changing at least some of the operating parameters of a single individual. The control device may randomly change the operating parameters of at least one device in the parent individual. For example, for... Figure 4 The control device can randomly change the operating parameters of any one or more devices in the first air conditioning system, or randomly change the operating parameters of any one or more devices in the second air conditioning system, in order to achieve the mutation operation of the individual.

[0154] The control device performs at least one of the crossover or mutation operations on the parent individuals selected multiple times in the current parent population P(t) to generate multiple offspring individuals, thus obtaining the current offspring population Q(t). If the current parent population is the initial parent population P(0), the control device performs operations such as selection, crossover, and mutation on the initial parent population P(0) to obtain the initial offspring population Q(0).

[0155] Step s2: Merge the current parent population with the current child population to obtain the current temporary population, and determine the non-dominated order and crowding distance of each individual in the current temporary population.

[0156] The control device can merge the current parent population P(t) with the current offspring population Q(t) to obtain the current temporary population R(t). The determination of the non-dominated order and crowding distance of each individual in the current temporary population R(t) is described in step two above, and will not be repeated here.

[0157] Step s3: Based on the non-dominant level and crowding distance of each individual in the current temporary population, select multiple target individuals from the current temporary population to form the next parent population.

[0158] By controlling the non-dominated level and crowding distance of each individual in the control device, the top N better individuals are selected from the current temporary population R(t) as target individuals to form the next parent population P(t+1). In the process of selecting target individuals, individuals with higher non-dominated levels can be given priority; for individuals at the same non-dominated level, individuals with larger crowding distances are given priority.

[0159] For example, the individuals in the current temporary population R(t) belong to three non-dominated levels: the first non-dominated level contains 10 individuals, the second non-dominated level contains 15 individuals, and the third non-dominated level contains 10 individuals. Assuming N=20, the control device can identify all 10 individuals in the first non-dominated level as target individuals, and then select the 10 individuals with the highest crowding distance in the second non-dominated level as target individuals as well.

[0160] Step 4: Based on individuals in the highest non-dominant hierarchy of the target parent population, identify multiple candidate control strategies.

[0161] The target parent population refers to the current parent population that meets the reproduction termination condition. This termination condition may include at least one of the following: the number of generations of reproduction exceeds a threshold, or the change in the parent population obtained through reproduction is less than a threshold. Multiple individuals in the highest non-dominated level (i.e., the first non-dominated level) of the target parent population are considered better individuals. Based on these individuals, a Pareto solution set for the multi-objective optimization problem can be constructed, with each individual corresponding to a solution in the Pareto solution set. This Pareto solution set contains a series of better candidate control strategies that balance the total output energy and total energy consumption of multiple air conditioning systems in the target area.

[0162] Each candidate control strategy may include: a total output energy, a total energy consumption, and control instructions that need to be executed for multiple air conditioning systems to achieve the total output energy and total energy consumption. These control instructions are the control instructions included in the corresponding individual system. The total output energy in the candidate control strategy can be the total predicted output energy corresponding to that individual system, and the total energy consumption in the candidate control strategy can be the total predicted energy consumption corresponding to that individual system.

[0163] In some embodiments, candidate control strategies may also weigh the overall indoor temperature and total energy consumption of the target area. The control device can convert the total output energy of each air conditioning system in the target area into an overall indoor temperature that the target area can achieve, based on the current actual temperature data of the target area, and then characterize the strategy based on the temperature dimension.

[0164] Figure 5 A schematic diagram illustrating a candidate control strategy provided according to an embodiment of this specification is shown, and Figure 5 Let's take the example of representing a strategy in the temperature dimension as an illustration. Figure 5 As shown, the control device can display a graph showing the correspondence between control commands and demand groups in each candidate control strategy. The demand groups corresponding to each candidate control strategy include the overall indoor temperature of a park and the total energy consumption of all air conditioning equipment. This graph can be a curve in a target coordinate system, where the horizontal and vertical axes are constructed based on the two objectives to be balanced. For example, the horizontal axis represents total energy consumption, and the vertical axis represents the overall indoor temperature of the target park. The curve in this target coordinate system represents a Pareto curve composed of Pareto solution sets. Each point on the curve corresponds to a candidate control strategy; the horizontal axis of that point represents the total energy consumption corresponding to that candidate control strategy, and the vertical axis represents the overall indoor temperature of the target park corresponding to that candidate control strategy.

[0165] For example, Figure 5The highest point in the curves shown represents the candidate control strategy under the economic model. Using this candidate control strategy to control the air conditioning systems in the target area requires lower total energy consumption, but also results in lower user comfort levels in the target area. Figure 5 The lowest point in the curves shown represents a candidate control strategy in comfort mode. Using this candidate control strategy to control the air conditioning systems in the target area results in higher user comfort, but also higher total energy consumption.

[0166] In some embodiments, the control device can acquire information on the target area's demand for total output energy and total energy consumption over a future period. This demand information can be manually uploaded or derived by the control device itself. Based on this demand information, the control device can determine the target control strategy that best matches the demand information from among various candidate control strategies. This target control strategy includes the operating parameters of multiple air conditioning systems, and correspondingly, coordinated control of multiple air conditioning systems can be achieved based on this target control strategy.

[0167] In some embodiments, the target control strategy can be manually selected by the management personnel of the target area. For example, the control device can output multiple candidate control strategies (e.g., displayed in a chart), and the management personnel can select the target control strategy based on the target area's future demand for total output energy and total energy consumption. In this case, the total output energy of each air conditioning system can be converted into the overall indoor temperature that the total output energy can achieve in the target area and displayed, making it easier and more intuitive for management personnel to determine the required target control strategy.

[0168] For example, if there is an important event in the target park on a certain day, and the overall indoor temperature in the target park needs to be low, some energy consumption can be sacrificed accordingly. Multiple candidate control strategies are identified for this purpose. Figure 5 The Pareto curve shown has multiple points marked on it, and this curve is determined based on several candidate control strategies. The management of the target park can select a target control strategy from these candidates that offers acceptable temperature and energy consumption based on actual needs. The Pareto curve shows that from a certain point onwards, total energy consumption increases rapidly, but the overall indoor temperature of the park does not decrease significantly. Candidate control strategies after this point are relatively less cost-effective, while candidate control strategies before this point, although having lower total energy consumption, also have higher temperatures. Considering all factors, this point represents a good balance between total energy consumption and temperature. Management can choose the candidate control strategy closest to this point (e.g., ...). Figure 5The candidate control strategy corresponding to point W in the diagram is used as the target control strategy to maximize the balance between energy consumption and temperature. Alternatively, managers can completely disregard energy consumption and directly select the control strategy that minimizes the overall indoor temperature, such as... Figure 5 The candidate control strategy in the comfort mode corresponding to the last point in the middle.

[0169] In the above embodiments, the control device does not consider a single air conditioning system to determine the optimal strategy for that individual system. Instead, it considers the multiple air conditioning systems in the target area as a whole, taking into account the relationships between them (such as the load relationship based on personnel movement), and determines multiple candidate control strategies that are optimal for the entire area. This approach can improve the accuracy of control strategy determination, that is, improve the adaptability of the control strategy to the target area. Furthermore, these multiple candidate control strategies can meet the diverse needs of the target area under different operating scenarios, finding a better balance between the total energy consumption of each air conditioning system and user comfort under each requirement.

[0170] Step 340: Based on the target control strategy, perform control operations on the multiple air conditioning systems.

[0171] After determining the target control strategy, the control equipment can execute control operations on multiple air conditioning systems within the target area, based on the control instructions included in the target control strategy for coordinated control. For example, the control equipment can set the parameters of the corresponding devices to the specified operating parameters in the control instructions to achieve control over each device. Alternatively, the control equipment can send sub-control instructions to each device, each carrying the device's operating parameters; upon receiving the sub-control instructions, each device can parse the corresponding operating parameters and adjust its current operating parameters accordingly.

[0172] In some embodiments, the control device may determine the target control strategy using the above method only for the first working period of the air conditioning system's operating cycle (e.g., a day). For subsequent working periods, the control device may use other methods to determine the corresponding control strategies. For example, in subsequent working periods, a required temperature may be set for each building, and the control device may determine a control strategy that minimizes the energy consumption of the air conditioning system in that building based on the required temperature corresponding to each building, or determine a control strategy that minimizes the total energy consumption of all air conditioning systems in the target area.

[0173] In some embodiments, the control device may also employ the method of determining the target control strategy provided in the above embodiments for each working period of the day. The target control strategy corresponding to the first working period indicates the operating parameters of each air conditioning system, and also indicates the start-up time and / or running duration of each air conditioning system. The target control strategy corresponding to the working periods after the first working period may only indicate the operating parameters of each air conditioning system, and no longer indicate the start-up time and / or running duration of each air conditioning system; or, the target control strategy may indicate that the start-up time of each air conditioning system is the initial time of the working period, and the running duration is the total duration of the working period.

[0174] In some embodiments, the control device can comprehensively utilize the methods provided in the above embodiments to predict pedestrian flow trends in the target area and coordinately control the various air conditioning systems in the target area based on the prediction results. The control device can utilize data acquisition components installed in each building and each air conditioning system, and also combine other data acquisition methods, to acquire pedestrian flow data for each building, first observation data affecting the load of the air conditioning system (excluding pedestrian flow), and second observation data affecting pedestrian flow. Based on this historical data, a pedestrian flow trend prediction model for the target area and a load prediction model corresponding to each air conditioning system can be constructed.

[0175] Subsequently, during the control of each air conditioning system, the control equipment utilizes the constructed pedestrian flow trend prediction model and load prediction model. For example, before the start of an air conditioning work cycle, the control equipment can input the current pedestrian flow data and second observation data into the pedestrian flow trend prediction model to predict the pedestrian flow trend, which characterizes the migration and / or distribution of people in the target area during future periods (such as the first work period in the air conditioning work cycle). Before the start of an air conditioning work cycle, the current pedestrian flow can be considered negligible, and thus only the second observation data can be input into the pedestrian flow trend prediction model. For other work periods in the air conditioning work cycle, there is usually pedestrian flow in each building; accordingly, the observed actual pedestrian flow data and the second observation data can be input together into the pedestrian flow trend prediction model for prediction. Based on this pedestrian flow trend prediction data, the control equipment obtains the predicted pedestrian flow for each building in the future period. The first observation data corresponding to the predicted pedestrian flow for each building is then input into the load prediction model corresponding to that building to predict the target load of the air conditioning system in that building in the future period.

[0176] Subsequently, the control equipment can use the operating parameters and start-up times of each air conditioning system in the target area as variables, and the target load of each building as constraints. The goal is to maximize the total output energy of each air conditioning system (e.g., minimizing the overall indoor temperature of the target area or maximizing environmental comfort) and minimize the total energy consumption of each air conditioning system as two objectives. The NSGA-II method is used to simultaneously optimize these two conflicting objectives, resulting in a Pareto solution set that balances total energy consumption and environmental comfort. Each solution in this Pareto solution set can serve as a candidate control strategy, indicating when and with what operating parameters each air conditioning system should start up to achieve a given total output energy and total energy consumption. Based on this Pareto solution set, the managers of the target area can select appropriate target control strategies according to different operating scenarios to control the air conditioning systems in different buildings separately.

[0177] In the above embodiments, the control device captures the flow trends of people between multiple buildings in the target park, and predicts the target load of the air conditioning system in each building with relatively high accuracy. The control device integrates the ability to decide the start-up time of multiple air conditioning systems in the target park (e.g., air conditioning systems in different buildings can correspond to different start-up times) with multi-objective decision-making capabilities to obtain a target control strategy for coordinated control of each air conditioning system. This allows for more accurate matching of the dynamic load demand of different buildings in the target park, resulting in a target control strategy that achieves a good balance between user comfort and energy consumption costs, giving managers the ability to flexibly adjust strategies; furthermore, it enables flexible and intelligent control of the air conditioning system in the target park, improving the intelligence and flexibility of park management, and improving the overall energy efficiency of the air conditioning system, thereby increasing economic and social benefits.

[0178] In summary, the air conditioning control method provided in this specification, targeting a specific industrial park, determines the target load of each building's air conditioning system based on pedestrian flow trend prediction data characterizing the migration and / or distribution of people between different buildings in the future. Then, based on this target load, the operating parameters of multiple air conditioning systems are jointly planned as a whole to obtain a target control strategy for coordinated control of these systems, thereby executing control operations on each system. Thus, by considering the pedestrian flow relationships between buildings in the target industrial park, the target load of each building's air conditioning system can be determined more accurately. This allows for the determination of a target control strategy that matches the current demand for each air conditioning system as a whole. Coordinated control of each air conditioning system based on this target control strategy improves the control accuracy of each system and enhances the overall energy efficiency of the air conditioning systems in the target industrial park.

[0179] This specification, in another aspect, provides a computer-readable non-transitory storage medium storing at least one instruction set. When the at least one instruction set is executed by a processor, the at least one instruction set instructs the processor to implement the steps of the air conditioning control method P300 described in this specification. In some possible embodiments, various aspects of this specification can also be implemented as a program product comprising program code. When the program product is run on a control device 120, the program code is used to cause the control device 120 to perform the steps of the air conditioning control method P300 described in this specification. The program product for implementing the above method may employ a portable compact disc read-only memory (CD-ROM) containing program code and may run on the control device 120. However, the program product of this specification is not limited thereto. In this specification, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system. The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media include: portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. The computer-readable storage medium may include a data signal propagated as part of a carrier wave in baseband, carrying readable program code. This propagated data signal may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium may also be any readable medium other than a readable storage medium that can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof. Program code for performing operations of this specification may be written in any combination of one or more programming languages, including object-oriented programming languages—such as Java, C++, etc.—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the control device 120, partially on the control device 120, as a standalone software package, partially on the control device 120 and partially on a remote computing device, or entirely on a remote computing device.

[0180] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0181] In summary, after reading this detailed disclosure, those skilled in the art will understand that the foregoing detailed disclosure is presented by way of example only and is not restrictive. Although not explicitly stated herein, those skilled in the art will understand that this specification requires various reasonable changes, improvements, and modifications to the embodiments. These changes, improvements, and modifications are intended to be made by this specification and are within the spirit and scope of the exemplary embodiments described herein.

[0182] Furthermore, certain terms in this specification have been used to describe embodiments of this specification. For example, "an embodiment," "an embodiment," and / or "some embodiments" mean that a particular feature, structure, or characteristic described in connection with that embodiment may be included in at least one embodiment of this specification. Therefore, it is to be emphasized and understood that two or more references to "an embodiment" or "an embodiment" or "alternative embodiment" in various parts of this specification do not necessarily refer to the same embodiment. Moreover, specific features, structures, or characteristics may be suitably combined in one or more embodiments of this specification.

[0183] It should be understood that in the foregoing description of the embodiments in this specification, various features are combined in a single embodiment, drawing, or description for the purpose of simplifying the description and to aid in understanding a feature. However, this does not mean that the combination of these features is necessary, and those skilled in the art, upon reading this specification, may readily identify some of the devices as separate embodiments. That is, the embodiments in this specification can also be understood as an integration of multiple secondary embodiments. And the content of each secondary embodiment is valid even if it contains fewer than all the features of a single foregoing disclosed embodiment.

[0184] Every patent, patent application, publication of a patent application, and other material such as articles, books, specifications, publications, documents, articles, etc., cited herein, except for those inconsistent with or conflicting with this document, or those having a restrictive effect on the widest scope of the claims, may be incorporated herein by reference for all purposes now or hereafter associated with this document. Furthermore, in the event of any inconsistency or conflict between the description, definition, and / or use of relevant terms in any material and the description, definition, and / or use of relevant terms in this document, the terms in this document shall prevail.

[0185] Finally, it should be understood that the embodiments disclosed herein are illustrative of the principles of the embodiments described in this specification. Other modified embodiments are also within the scope of this specification. Therefore, the embodiments disclosed in this specification are merely examples and not limitations. Those skilled in the art can implement the applications described in this specification using alternative configurations based on the embodiments in this specification. Therefore, the embodiments in this specification are not limited to the embodiments precisely described in the applications.

Claims

1. An air conditioning control method for collaboratively controlling multiple air conditioning systems deployed in different buildings within a target industrial park, the method comprising: Obtain predicted pedestrian flow trends in the target park during a future time period, wherein the predicted pedestrian flow trends represent the migration and / or distribution of people between different buildings; Based on the predicted pedestrian flow data, the target load of each of the multiple air conditioning systems is determined; Based on the target load of each of the multiple air conditioning systems, and taking the multiple air conditioning systems as a whole, the operating parameters of the multiple air conditioning systems are jointly planned to obtain a target control strategy for the coordinated control of the multiple air conditioning systems. as well as Based on the target control strategy, control operations are performed on the multiple air conditioning systems.

2. The method according to claim 1, wherein, The step of determining the target load for each of the multiple air conditioning systems based on the predicted pedestrian flow data includes: Perform the following steps for each air conditioning system: Based on the predicted pedestrian flow data, determine the pedestrian flow data of the building where the air conditioning system is located during the future time period; Obtain first observation data associated with the air conditioning system, the first observation data including data of at least one dimension related to the air conditioning load; and The pedestrian flow data and the first observation data are input into a pre-trained load prediction model to obtain the target load of the air conditioning system.

3. The method according to claim 2, wherein, The first observation data includes at least one of the following: the load of the air conditioning system during a historical period, environmental data of the building where the air conditioning system is located, operating data of the air conditioning system, or meteorological data of the region where the air conditioning system is located.

4. The method according to claim 1, wherein, The process of obtaining the predicted pedestrian flow trend data for the target park in the future time period includes: Obtain second observation data associated with the target park, the second observation data including data of at least one dimension related to personnel flow; and The second observation data is input into a pre-trained pedestrian flow trend prediction model to obtain the pedestrian flow trend prediction data.

5. The method according to claim 4, wherein, The pedestrian flow trend prediction model is a graph neural network-based model. The step of inputting the second observation data into the pre-trained pedestrian flow trend prediction model to obtain the pedestrian flow trend prediction data includes: Based on the second observation data, historical personnel flow information between different buildings in the target park is determined; Using buildings in the target park as nodes and constructing edges based on historical personnel flow information between different buildings, a spatiotemporal graph of personnel flow is generated; and The spatiotemporal map of population flow is input into the population flow trend prediction model to obtain the population flow trend prediction data.

6. The method according to claim 4, wherein, The second observation data includes at least one of the following: pedestrian traffic data of different buildings in the target park during historical periods, transaction arrangement data of different buildings in the target park, meteorological data of the area where the target park is located, or traffic status data of the location of the target park.

7. The method according to claim 1, wherein, The target control strategy includes at least one of the following: The start-up time, running time, or equipment operating parameters of the multiple air conditioning systems.

8. The method according to claim 1, wherein, The method involves jointly planning the operating parameters of the multiple air conditioning systems based on their respective target loads, treating the multiple air conditioning systems as a whole, to obtain a target control strategy for coordinated control of the multiple air conditioning systems, including: Obtain the air conditioning usage requirements of the target campus, including the target temperature required for each building at a target time; and With the constraint that the output energy of each air conditioning system is not lower than its respective target load, and with the objective that the output energy of each air conditioning system supports the corresponding building to reach the target temperature at the target time, the operating parameters of the multiple air conditioning systems are jointly planned to obtain the target control strategy.

9. The method according to claim 1, wherein, The method involves jointly planning the operating parameters of the multiple air conditioning systems based on their respective target loads, treating the multiple air conditioning systems as a whole, to obtain a target control strategy for coordinated control of the multiple air conditioning systems, including: With the constraint that the output energy of each air conditioning system is not less than its respective target load, and with the objectives of maximizing the total output energy of the multiple air conditioning systems and minimizing the total energy consumption of the multiple air conditioning systems, the operating parameters of the multiple air conditioning systems are jointly planned to obtain multiple candidate control strategies. Each candidate control strategy includes: a total output energy, a total energy consumption, and control instructions to be executed for the multiple air conditioning systems to achieve the total output energy and the total energy consumption; and Based on the target park's demand information for total output energy and total energy consumption in the future time period, the target control strategy is determined from the multiple candidate control strategies.

10. The method according to claim 9, wherein, The method involves jointly planning the operating parameters of multiple air conditioning systems, with the constraint that the output energy of each air conditioning system is not less than its respective target load, and with the objectives of maximizing the total output energy of the multiple air conditioning systems and minimizing the total energy consumption of the multiple air conditioning systems, to obtain multiple candidate control strategies, including: The control commands randomly generated for the multiple air conditioning systems are used as individuals in the population to obtain the initial parent population. Based on the target load set and output energy set corresponding to each individual in the initial parent population, the non-dominated level and crowding distance of each individual are determined, wherein the target load set includes the target load of each of the multiple air conditioning systems, and the output energy set includes the output energy of each of the multiple air conditioning systems; Based on the non-dominated hierarchy and crowding distance of each individual, the initial parent population is multi-generationally reproduced to obtain the target parent population; and Based on individuals in the highest non-dominant hierarchy of the target parent population, multiple candidate control strategies are identified.

11. The method according to claim 10, wherein, For any individual in the initial parent population, based on the target load set and output energy set corresponding to that individual in the initial parent population, determine the non-dominated level and crowding distance of that individual, including: If the output energy of any sub-air conditioning system corresponding to the individual is less than its target load, then the total predicted output energy and total predicted energy consumption corresponding to the individual are predicted, and based on the total predicted output energy, the total predicted energy consumption, and a preset penalty term, the non-dominated level and congestion distance of the individual are determined; or If the output energy of the multiple air conditioning systems corresponding to the individual is not less than their respective target loads, then the total predicted output energy and total predicted energy consumption corresponding to the individual are predicted, and based on the total predicted output energy and total predicted energy consumption, the non-dominated level and congestion distance of the individual are determined.

12. The method according to claim 10, wherein, The process of reproduction in each generation includes: Based on the non-dominant level and crowding distance of each individual in the current parent population, parent individuals are selected multiple times from the current parent population, and at least one of the crossover or mutation operations is performed on the selected parent individuals each time to generate multiple offspring individuals, thus obtaining the current offspring population. The current parent population and the current child population are merged to obtain a current temporary population, and the non-dominant level and crowding distance of each individual in the current temporary population are determined; and Based on the non-dominant level and crowding distance of each individual in the current temporary population, multiple target individuals are selected from the current temporary population to form the next parent population.

13. The method according to claim 12, wherein, The air conditioning system includes multiple devices, and each individual device includes the operating parameters of each device in the air conditioning system. The crossover operation includes: exchanging the operating parameters of the same devices in different individuals; and / or, The mutation operation includes randomly changing at least some of the operating parameters of a single individual.

14. A control device, comprising: At least one storage medium storing at least one instruction set for controlling the air conditioning system; as well as At least one processor is communicatively connected to the at least one storage medium, wherein the at least one processor reads the at least one instruction set during operation and performs the method as described in any one of claims 1-13 according to the instructions of the at least one instruction set.

15. A computer-readable non-volatile storage medium, wherein, The computer-readable non-volatile storage medium stores at least one instruction set, which, when executed by at least one processor, implements the method as described in any one of claims 1-13.

Images