Energy-saving refrigeration method and system

Abstract

The embodiment of the invention provides an energy-saving refrigeration method and system.The method is applied to a first subsystem and comprises the steps that an operation strategy for refrigeration equipment is generated through an AI model on the basis of the tail end cold demand of a refrigeration system, feedback data and the operation boundary of the refrigeration equipment in the refrigeration system; wherein the AI model comprises a first model of the refrigeration system and a second model corresponding to the refrigeration equipment; and the operation strategy is issued to a second subsystem, so that the second subsystem sends the operation strategy to the refrigeration equipment under the condition that the operation strategy meets an issuing condition, and the refrigeration equipment refrigerates according to the operation strategy. Two-stage subsystem hierarchical control is used, so that double safety guarantee is realized; after the AI is deployed, the AI can be continuously evolved according to operation feedback data of the corresponding water refrigeration system, global optimal energy-saving optimization strategy guidance and working condition adjusting scheme support are provided for the refrigeration system, and the efficiency of the refrigeration system is effectively improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of refrigeration optimization, and particularly to an energy-saving refrigeration method and system. Background Art

[0002] With the booming development of water-cooled source refrigeration systems, they are widely applied in multiple fields. To ensure the safe operation of the system, relatively large margins and conservative operation indicators are usually selected. Each independent device forms an operation adjustment island, and the refrigeration system often operates in a high-energy-consumption and low-efficiency range for a long time. Therefore, there is a common waste problem of imbalance between cold production and demand in water-cooled source refrigeration systems. How to quickly and effectively improve the overall operation indicators of the refrigeration system, reduce the waste of cold production, and achieve long-term stable, safe and energy-saving refrigeration of the refrigeration system has become particularly urgent.

[0003] In the related art, operation and maintenance personnel adjust the water-cooled system based on experience and automatic control adjustment schemes. However, this method not only has high requirements for operation and maintenance personnel, but also has limited optimization effects, and in many cases, energy conservation of the system cannot be achieved. Summary of the Invention

[0004] The embodiments of the present application provide an energy-saving refrigeration method and system, which helps to improve the efficiency of the refrigeration system.

[0005] In a first aspect, the embodiments of the present application provide an energy-saving refrigeration method, which is applied to a first subsystem. The method includes:

[0006] Based on the end cooling demand of the refrigeration system, feedback data, and the operation boundary of the refrigeration equipment in the refrigeration system, an operation strategy for the refrigeration equipment is generated through an Artificial Intelligence (AI) model; wherein, the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment;

[0007] The operation strategy is sent to a second subsystem, so that when the operation strategy meets the sending condition, the second subsystem sends the operation strategy to the refrigeration equipment, so that the refrigeration equipment refrigerates according to the operation strategy.

[0008] In the above solution, two-level subsystem hierarchical control is used to achieve double safety guarantees; after the AI model is deployed, it continuously self-iterates and updates. There is a clear physical logic association and strong interpretability between the models. After the AI is deployed, it can continuously self-evolve according to the operation feedback data of the corresponding water refrigeration system; based on the AI model technical route, the system equipment and the overall operation relationship are built, which can permanently provide the water refrigeration system with a globally optimal energy-saving optimization strategy guidance and working condition adjustment scheme support, effectively improving the efficiency of the refrigeration system, making the refrigeration system operate more efficiently, safely and energy-savingly.

[0009] In some alternative embodiments, both the first model and the second model include white-box mechanism models.

[0010] In the above solution, the first subsystem adopts white-box AI + mechanism model, and the energy-saving effect can be expected. The speed of going online, using and adjusting is fast. With the characteristics of being guided by the refrigeration logic mechanism of system equipment, the model has strong universality in use, high running logic correlation, the AI white-box mechanism model has interpretability, the computing power requirement for the optimization process of the system is small, the speed of generating energy-saving optimization strategies is fast, the accuracy of the system model is relatively higher, and the credibility of the energy-saving strategy is better.

[0011] In some alternative embodiments, it further includes:

[0012] Based on the operation strategy, obtain the predicted performance and the predicted operation parameters corresponding to the operation strategy;

[0013] Wherein, the predicted operation parameters are the operation parameters of the refrigeration equipment when the refrigeration equipment executes the operation strategy; the predicted performance is the performance of the refrigeration equipment when the refrigeration equipment executes the operation strategy.

[0014] In the above solution, through the prediction and analysis of the first subsystem, the iterative optimization of the AI energy consumption model is realized, as well as the double screening of the operation strategy corresponding to the abnormal data by the second subsystem.

[0015] In some alternative embodiments, it further includes:

[0016] Train the AI model according to the actual performance of the refrigeration equipment, the cooling demand at the end of the refrigeration system, the feedback data, and the operation boundary of the refrigeration equipment in the refrigeration system to obtain an updated AI model.

[0017] In the above solution, the AI model is continuously self-iterated and updated after being deployed to optimize the AI model.

[0018] In some alternative embodiments, it further includes:

[0019] If the error between the predicted performance and the actual performance of any refrigeration equipment is greater than the first error; and / or the error between the predicted operation parameters and the actual operation parameters is greater than the second error, then update the AI model.

[0020] In the above solution, through the AI prediction calculation of the first subsystem, the iterative optimization of the AI energy consumption model is realized, enabling the AI to continuously self-evolve and improve the prediction effect.

[0021] In some alternative embodiments, the first subsystem is specifically used for:

[0022] Iterate successively from the AI initial boundary to the safety boundary corresponding to the refrigeration equipment at a preset step size.

[0023] In the above solution, the first subsystem gradually iterates to the process safety boundary corresponding to the refrigeration equipment, tracks the global optimal energy-saving regulation strategy of the system, and guides the system to continuously and reliably operate in the energy-saving condition.

[0024] In some alternative embodiments, it further includes:

[0025] In response to an addition instruction for the first refrigeration equipment, add a second model corresponding to the first refrigeration equipment in the AI model; wherein, the first refrigeration equipment is a newly added refrigeration equipment in the refrigeration system; or

[0026] In response to a deletion instruction for the second refrigeration equipment, delete the second model corresponding to the second refrigeration equipment in the AI model; wherein, the second refrigeration equipment is an existing refrigeration equipment in the refrigeration system; or

[0027] In response to a modification instruction for the third refrigeration equipment, modify the second model corresponding to the third refrigeration equipment in the AI model; wherein, the third refrigeration equipment is an existing refrigeration equipment in the refrigeration system.

[0028] In the above solution, since AI models corresponding to each refrigeration equipment are set, therefore, when there are drastic changes in load or refrigeration structure at the deployment site, only the corresponding model settings need to be adjusted, and the system configuration parameters are updated to complete the corresponding upgrade. Without redeploying the entire AI system, the structural changes of the refrigeration system can be flexibly responded to.

[0029] In some alternative embodiments, the first subsystem is carried on the infrastructure management platform;

[0030] Sending the operation strategy to the second subsystem includes:

[0031] Send the operation strategy to the second subsystem through the infrastructure management platform;

[0032] The method further includes:

[0033] Receive the feedback data uploaded by the second subsystem through the infrastructure management platform.

[0034] In some alternative embodiments, the infrastructure management platform further includes a visualization interface;

[0035] Among them, the visualization interface includes an AI interface and a device interface; the AI interface includes some or all of an AI main interface, an AI energy-saving optimization strategy interface, an AI cold source parameter optimization interface, an AI cockpit management interface, an alarm management interface, and an AI model training management interface; the device interface includes some or all of a 3D view of the refrigeration equipment and an operating parameter interface.

[0036] In some alternative embodiments, the method further includes:

[0037] Performing alarm management through the infrastructure management platform.

[0038] In a second aspect, an embodiment of the present application further provides another energy-saving refrigeration method, which is applied to a second subsystem. The method includes:

[0039] Receiving operation strategies for each refrigeration equipment in the refrigeration system; wherein, the operation strategies are generated by an artificial intelligence (AI) model by a first subsystem based on the end cooling demand of the refrigeration system, feedback data, and the operation boundaries of the refrigeration equipment in the refrigeration system; the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment;

[0040] Determining whether the operation strategies meet the distribution conditions, and when the operation strategies meet the distribution conditions, sending the operation strategies to the refrigeration equipment so that the refrigeration equipment refrigerates according to the operation strategies.

[0041] In the above solution, two-level subsystem hierarchical control is used to achieve double safety guarantees; after the AI model is deployed, it continuously self-iterates and updates. There is a clear physical logic association and strong interpretability between the models. After the AI is deployed, it can continuously self-evolve according to the operation feedback data of the corresponding water refrigeration system; based on the AI model technical route, the system equipment and the overall operation relationship are built, which can permanently provide the water refrigeration system with a globally optimal energy-saving optimization strategy guidance and a working condition adjustment plan support, effectively improving the efficiency of the refrigeration system, making the refrigeration system operate more efficiently, more safely, and more energy-saving.

[0042] In some alternative embodiments, for the second subsystem to determine whether the operation strategies meet the distribution conditions, it includes:

[0043] For the refrigeration equipment, determining whether the operation strategy of the refrigeration equipment is within the safety boundary range corresponding to the refrigeration equipment, and determining whether the most recent prediction data of the refrigeration equipment is normal; the prediction data is the data corresponding to the operation strategy obtained by the first subsystem based on the operation strategy.

[0044] If the operation strategies of the refrigeration equipment are all within the corresponding safety boundaries and the predicted data of the refrigeration equipment are all normal, it is determined that the operation strategies meet the issuance conditions; otherwise, it is determined that the operation strategies do not meet the issuance conditions.

[0045] In the above solution, the second subsystem determines whether the operation strategy is within the corresponding safety boundary and whether the predicted data of the refrigeration equipment are normal; and only when the operation strategies of all refrigeration equipment are within the corresponding safety boundaries and the predicted data of all refrigeration equipment are normal, the operation strategy that conforms to the safety logic will be issued to the corresponding control points, realizing the second screening of the operation strategy and ensuring the safety of the operation strategy.

[0046] In some optional embodiments, it further includes:

[0047] If an exit instruction is received or it is determined that the control of the first subsystem is abnormal, the refrigeration equipment is controlled to operate based on the Building Automation (BA) control strategy;

[0048] In response to the access instruction, re-associate with the first subsystem.

[0049] In the above solution, by triggering the AI switching instruction or monitoring the abnormal operation of the first subsystem control module, AI switching is performed, and the refrigeration equipment is controlled to operate through the BA control strategy, reducing the occurrence of the situation of the AI platform malfunctioning.

[0050] In a third aspect, an embodiment of the present application further provides an energy-saving refrigeration system, including: a first subsystem and a second subsystem;

[0051] The first subsystem is used to generate an operation strategy for the refrigeration equipment based on the cooling demand at the end of the refrigeration system, feedback data, and the operation boundaries of the refrigeration equipment in the refrigeration system through an artificial intelligence (AI) model; wherein, the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment;

[0052] The second subsystem is used to receive the operation strategy; and when the operation strategy meets the issuance conditions, send the operation strategy to the refrigeration equipment so that the refrigeration equipment cools according to the operation strategy.

[0053] In some optional embodiments, both the first model and the second model include a white box mechanism model.

[0054] In some optional embodiments, the second subsystem is specifically used for:

[0055] For the refrigeration equipment, determine whether the operation strategy of the refrigeration equipment is within the corresponding safety boundary of the refrigeration equipment, and determine whether the most recent predicted data of the refrigeration equipment is normal; the predicted data is the data corresponding to the operation strategy obtained by the first subsystem based on the operation strategy.

[0056] If the operation strategies of the refrigeration equipment are all within the corresponding safety boundaries, and the predicted data of the refrigeration equipment is all normal, it is determined that the operation strategy meets the distribution condition; otherwise, it is determined that the operation strategy does not meet the distribution condition.

[0057] In some alternative embodiments, the first subsystem is further configured to:

[0058] Based on the operation strategy, obtain the predicted performance and the predicted operation parameters corresponding to the operation strategy; wherein, the predicted operation parameters are the operation parameters of the refrigeration equipment when the refrigeration equipment executes the operation strategy; the predicted performance is the performance of the refrigeration equipment when the refrigeration equipment executes the operation strategy.

[0059] In some alternative embodiments, the first subsystem is further configured to:

[0060] Train the AI model according to the actual performance of the refrigeration equipment, the cooling demand at the end of the refrigeration system, the feedback data, and the operation boundary of the refrigeration equipment in the refrigeration system to obtain an updated AI model.

[0061] In some alternative embodiments, the first subsystem is further configured to:

[0062] If the error between the predicted performance and the actual performance of any refrigeration equipment is greater than the first error; and / or the error between the predicted operation parameters and the actual operation parameters is greater than the second error, update the AI model.

[0063] In some alternative embodiments, the second subsystem is further configured to:

[0064] If an exit instruction is received, or it is determined that the control of the first subsystem is abnormal, control the operation of the refrigeration equipment based on the BA control strategy;

[0065] In response to the access instruction, re-associate with the first subsystem.

[0066] In some alternative embodiments, the first subsystem is specifically configured to:

[0067] Iteratively step by step from the AI initial boundary to the corresponding safety boundary of the refrigeration equipment.

[0068] In some alternative embodiments, the first subsystem is further configured to:

[0069] In response to an addition instruction for a first refrigeration device, add a second model corresponding to the first refrigeration device to the AI model; wherein the first refrigeration device is a newly added refrigeration device in the refrigeration system; or

[0070] In response to a removal instruction for a second refrigeration device, delete the second model corresponding to the second refrigeration device from the AI model; wherein the second refrigeration device is an existing refrigeration device in the refrigeration system; or

[0071] In response to a modification instruction for a third refrigeration device, modify the second model corresponding to the third refrigeration device in the AI model; wherein the third refrigeration device is an existing refrigeration device in the refrigeration system.

[0072] In some alternative embodiments, the first subsystem is carried on an infrastructure management platform;

[0073] The infrastructure management platform is configured to send the feedback data uploaded by the second subsystem to the first subsystem; and send the operation strategy generated by the first subsystem to the second subsystem.

[0074] In some alternative embodiments, the infrastructure management platform further includes a visualization interface;

[0075] Wherein, the visualization interface includes an AI interface and a device interface; the AI interface includes some or all of an AI main interface, an AI energy-saving optimization strategy interface, an AI cold source parameter optimization interface, an AI cockpit management interface, an alarm management interface, and an AI model training management interface; the device interface includes some or all of a 3D view of the refrigeration device and an operation parameter interface.

[0076] In some alternative embodiments, the infrastructure management platform is further configured to perform alarm management. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] To more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0078] Figure 1 It is a system architecture diagram of the first energy-saving refrigeration system provided by the embodiments of the present application;

[0079] Figure 2An interaction flowchart of an energy-saving refrigeration method provided by an embodiment of the present application;

[0080] Figure 3 A working flowchart of a first subsystem provided by an embodiment of the present application;

[0081] Figure 4 A physical model of a chiller provided by an embodiment of the present application;

[0082] Figure 5 A system architecture diagram of a second energy-saving refrigeration system provided by an embodiment of the present application;

[0083] Figure 6 A schematic diagram of an AI interface provided by an embodiment of the present application;

[0084] Figure 7 A schematic diagram of an AI energy-saving optimization strategy interface provided by an embodiment of the present application;

[0085] Figure 8 A schematic diagram of an AI cockpit management interface provided by an embodiment of the present application;

[0086] Figure 9 A schematic diagram of an alarm management interface provided by an embodiment of the present application;

[0087] Figure 10 A schematic flowchart of a first energy-saving refrigeration method provided by an embodiment of the present application;

[0088] Figure 11 A schematic flowchart of a second energy-saving refrigeration method provided by an embodiment of the present application;

[0089] Figure 12 A schematic structural diagram of a first energy-saving refrigeration device provided by an embodiment of the present application;

[0090] Figure 13 A schematic structural diagram of a second energy-saving refrigeration device provided by an embodiment of the present application;

[0091] Figure 14 A schematic structural diagram of a first subsystem provided by an embodiment of the present application. Detailed implementation manners

[0092] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the scope of protection of the present application.

[0093] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0094] With the vigorous development of water-cooled refrigeration systems, they are widely used in many fields. In order to ensure the safe operation of the system, a larger margin and conservative operating indicators are usually selected, and each independent device forms an operating adjustment island. The refrigeration system often runs in a high-energy consumption and low-efficiency range for a long time. Therefore, there is a common waste problem of unbalanced relationship between cold preparation and demand in water-cooled refrigeration systems. How to quickly and effectively improve the overall operating indicators of the refrigeration system, reduce the waste of cold preparation, and achieve long-term, stable and safe refrigeration of the refrigeration system has become particularly urgent.

[0095] In the related technology, operation and maintenance personnel adjust the water cooling system based on experience and automatic adjustment solutions. However, this method not only has high requirements on operation and maintenance personnel, but also has limited optimization effects, and in many cases cannot achieve system energy saving.

[0096] The adjustment of operating parameters at the system level requires very high HVAC calculations and operating experience based on actual on-site conditions. At the same time, it is necessary to use automated implementation methods to control and balance the operating logic associations and effect interlocking effects between equipment to complete the complete AI energy-saving operation logic of collection-solution-strategy-judgment-regulation-feedback at the overall refrigeration system level. It is necessary to have precise control over the energy-saving control points of the specific water refrigeration system, clearly define the operating strategy, safety strategy, adjustment strategy and their corresponding range boundaries, and have a solid understanding of the judgment point self-control characteristics and refrigeration equipment characteristics.

[0097] Based on this, the embodiments of the present application provide an energy-saving refrigeration method and system to effectively improve the efficiency of a refrigeration system.

[0098] See also Figure 1 As shown, the first energy-saving refrigeration system provided in the embodiment of the present application includes two-level control subsystems: a first subsystem and a second subsystem;

[0099] The first subsystem is the management layer, including the server and the client; the second subsystem includes the operation layer and the control layer; the second subsystem is for one or more refrigeration systems, Figure 1 Take n refrigeration systems as an example; each refrigeration system includes multiple refrigeration devices.

[0100] The first subsystem is configured to generate an operation strategy for the refrigeration equipment based on the end cooling demand of the refrigeration system, feedback data, and the operation boundary of the refrigeration equipment in the refrigeration system through an artificial intelligence (AI) model. The AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment.

[0101] The second subsystem is configured to receive the operation strategy and, when the operation strategy meets the sending condition, send the operation strategy to the refrigeration equipment so that the refrigeration equipment refrigerates according to the operation strategy.

[0102] In the above solution, two-level subsystem hierarchical control is used to achieve dual safety guarantees. After the AI model is deployed, it continuously self-iterates and updates. There is a clear physical logic association and strong interpretability between the models. After the AI is deployed, it can continuously self-evolve according to the operation feedback data of the corresponding water refrigeration system. Based on the AI model technology route, the system equipment and the overall operation relationship are built, which can permanently provide the water refrigeration system with global optimal energy-saving optimization strategy guidance and operating condition adjustment plan support, effectively improving the efficiency of the refrigeration system and making the refrigeration system operate more efficiently, safely, and energy-savingly.

[0103] Next, in combination with the accompanying drawings and specific embodiments, the technical solution of the present application and how the technical solution of the present application solves the above technical problems will be described in detail. These specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0104] An embodiment of the present application provides an energy-saving refrigeration method, as Figure 2 shown. The method includes:

[0105] Step S201: The first subsystem generates an operation strategy for the refrigeration equipment based on the end cooling demand of the refrigeration system, feedback data, and the operation boundary of the refrigeration equipment in the refrigeration system through an artificial intelligence (AI) model.

[0106] Wherein, the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment.

[0107] Step S202: The first subsystem sends the operation strategy to the second subsystem.

[0108] Step S203: The second subsystem determines whether the operation strategy meets the sending condition, and when the operation strategy meets the sending condition, sends the operation strategy to the refrigeration equipment.

[0109] The refrigeration equipment refrigerates according to the operation strategy after receiving the operation strategy.

[0110] In this embodiment, a combined control strategy of the first subsystem + the second subsystem is adopted. The first subsystem is responsible for parameter calculation and optimization strategy generation, and the second subsystem realizes basic control and process safety boundary management and control.

[0111] Specifically, AI optimization calculation gives the optimal parameter combination of each refrigeration equipment control point in the refrigeration system, generates an optimized control strategy (operation strategy), and realizes the search for an energy-saving optimization plan. After completing the energy-saving parameter modulation, the operation data collected by each sensor distributed in the refrigeration system is collected and uploaded through the second subsystem, and fed back to the first subsystem for judging the adjustment effect of the operation strategy. And continuously realize the self-update and iteration of the AI model through the cycle of collection - calculation - distribution - feedback, continuously improve the accuracy of the AI model, infinitely fit the characteristics of specific on-site operation conditions, and achieve the purpose of long-term credibility and high energy efficiency after AI deployment. That is to say, the first subsystem can achieve a certain range of self-evolution along with the operation of the refrigeration system. When some non-mutating / violent changes occur in the refrigeration equipment or the terminal side, the first subsystem has good self-correction and new environment adaptability, and can fit the new refrigeration operation conditions after several iteration cycles without the need to retrain the AI model or redeploy the system solution, only need to adjust the original settings and logical associations of the product according to the new scenario requirements. The first subsystem can continuously optimize the refrigeration station control strategy with the changes of the Internet Technology (IT) load in the computer room and the external climate, achieve optimal control, perform AI management and control on the cold station of commercial buildings, and realize AI strategy optimization on the basis of group control. In this embodiment, it can cooperate with the energy-saving optimization construction of commercial buildings (such as office buildings, shopping malls, data centers), further optimize the operation of the chilled water equipment in commercial buildings, and reduce the energy consumption of the cold station. Different energy-saving improvement plans will be matched for different refrigeration system scenarios to improve the operation efficiency of each device, reduce the imbalance problem of cold quantity supply and demand, improve the economy of refrigeration operation, and make the device performance reach the design value.

[0112] The first subsystem takes the demand of the terminal evaporation coil as the starting point of optimization, reflects the cooling capacity demand of the cooled object for the chilled water, and serves as the target value on the cold source side. Its basic principle is to calculate the operating condition set point with the lowest energy consumption through the equipment performance models such as the chiller, system water pump, and cooling tower under the constraint conditions of the operating safety boundary conditions on the premise of meeting the cooling demand of the terminal, and use it as the optimal control parameter. By issuing and commanding the parameter adjustment of each control point, the whole system is maintained to operate in the most energy-saving interval, and the energy-saving goal of the water-cooled source side is realized.

[0113] This embodiment does not make specific limitations on the first subsystem. Refer to Figure 3 As shown, in a cycle, through system operation data feature engineering (data processing) - energy consumption prediction and safety guarantee model - dynamic iteration of overall parameter optimization.

[0114] The specific content of the operation strategy in this embodiment is not limited. Different types of refrigeration equipment will be set in different refrigeration systems. The operation strategy may include information such as operation parameters and operation modes. For example: the control of the number of chillers, the optimized setting of the chilled water supply temperature of the chiller; the control of the number of pumps, the variable frequency control of pumps, the optimized setting of the chilled water pressure difference and temperature difference; the control of the number of cooling towers, the variable frequency control of the cooling tower fans, the optimized setting of the inlet and outlet water temperatures of the cooling tower; the full-automatic addition and subtraction control of the chiller station, etc.

[0115] In the above solution, two-level subsystem hierarchical control is used to achieve double safety guarantees; after the AI model is deployed, it continuously self-iterates and updates. There is a clear physical logic association and strong interpretability among the models. After the AI is deployed, it can continuously self-evolve according to the operation feedback data of the corresponding water refrigeration system; based on the AI model technical route, the system equipment and the overall operation relationship are built, which can permanently provide the global optimal energy-saving optimization strategy guidance and operating condition adjustment scheme support for the water refrigeration system, effectively improving the efficiency of the refrigeration system and making the refrigeration system operate more efficiently, more safely and more energy-saving.

[0116] In some alternative embodiments, both the first model and the second model include white-box mechanism models.

[0117] Since the black-box type AI completely eliminates the dimensionalized digital driving method and directly ignores the interpretability of system calculations, it often ignores a large number of physical association logics and invalidates them. There are often under-coupled or over-coupled models in the AI model, and its credibility is a major hidden danger. The black-box solution is only applicable to building a comprehensive large model of the whole system and does not have descriptive and integrative power for each specific operating equipment in the system. If sub-models of equipment are established step by step, it is also impossible to support the operation optimization meaning calculation logic of the whole system model by integrating and associating these sub-models. In addition, the black-box solution also has disadvantages such as huge computational resource requirements, too long on-site deployment and online effectiveness cycle, no self-evolution ability after the model is established, poor adaptability to changes in the on-site refrigeration system and the cooled object, and weak ability to support user scenarios that require rapid production delivery. In addition, for some users with data confidentiality requirements, the cloud deployment calculation method of some black-box solutions is not secure and compliant, which also becomes a difficult point restricting the implementation of the black-box solution.

[0118] Based on this, in this embodiment, the first subsystem deploys a white-box mechanism model of the refrigeration system and white-box mechanism models corresponding to each refrigeration device, that is, a combined strategy of white-box mechanism modeling plus data-driven AI machine learning is adopted. The first subsystem takes the demand of the terminal evaporation coil (cooling demand at the terminal) as the starting point for optimization, reflects the cooling demand of the cooled object for the chilled water, and serves as the target value on the cold source side. On the premise of meeting the cooling demand at the terminal, through the refrigeration device model, under the constraint of the operating boundary conditions, the operating condition set point with the lowest energy consumption is calculated through the AI optimization algorithm and used as the optimal control parameter (i.e., the operating strategy of the refrigeration device).

[0119] The white-box mechanism model includes at least a white-box principle model and also a mechanism model.

[0120] In a possible embodiment, the first model is a white-box principle model and the second model is a mechanism model; in a second possible embodiment, both the first model and the second model are white-box principle models.

[0121] Since different types of refrigeration devices are set in different refrigeration systems, the AI models will also be different. The following takes several specific refrigeration devices as examples for illustration:

[0122] For chillers, the chiller's loading and unloading and unit addition and subtraction are reasonably allocated according to the heat load. According to the actual equipment, a physical model of the chiller is established, which can be referred to Figure 4 as shown. The physical model of the chiller reflects the basic operating characteristics of the actual equipment, conforms to the unique operating curve of the chiller, and the coefficient of performance (COP) of the host can be calculated from this model under various operating conditions. According to the principles of meeting the process design, cooling demand, and global optimization of the central air-conditioning system, the chilled water outlet temperature of the optimized chiller is dynamically set, and reasonable unit addition and subtraction judgment is made dynamically to reduce power consumption. According to the COP characteristic curve obtained by AI machine learning, the optimal chilled water temperature at different load rates is predicted, and the highest point of the chiller energy efficiency pursues the boundary value of the energy efficiency curve.

[0123] For cooling towers, according to the actual equipment, a physical model of the cooling tower is established. And the energy consumption of the cooling tower under various operating conditions can be calculated from this model. According to the principles of meeting the system heat rejection demand and global optimization of the central air-conditioning system, the optimal outlet water temperature under the current operating condition is determined, and the optimal number of operating fans is automatically selected according to this temperature, and the operating frequency of the cooling fan is dynamically adjusted. According to the physical model of the cooling tower, the optimal outlet water temperature of the cooling tower is predicted to achieve the best regulation of the cooling tower fan; within the heat exchange capacity range of the cooling tower, the cooling water pump is adjusted according to the prediction model to achieve the best control of the cooling water volume. The key point for adjusting the cooling tower efficiency is to adjust the air-water ratio, and the cooling tower's loading and unloading and unit addition and subtraction are reasonably allocated according to the cooling demand.

[0124] For water pumps, the optimal regulation is given according to the water pump performance model. According to theoretical calculations, the relationship between the water pump frequency and power is cubic. For every 1 Hz decrease in the water pump frequency, the power will decrease significantly. Based on the actual equipment, physical models of chilled water pumps and cooling water pumps are established, and the energy consumption of the water pumps under various operating conditions can be calculated from this model. According to the principle of meeting the total cooling capacity demand of the system and the global optimization of the central air-conditioning system, and considering the changes in the chilled water supply / return water temperature and pressure difference, the optimal operating frequency and number of chilled water pumps are determined. The operating frequency of the water pump is coordinated to ensure the supply and return water pressure difference at the most unfavorable end of the chilled water loop system, meet the chilled water flow demand of the air-conditioning terminal, and dynamically adjust the chilled water frequency.

[0125] For air conditioners, sensors in the area of the end cooling object monitor the environmental conditions in real time according to the actual load changes and feed the data back to the first subsystem to calculate the cooling capacity required to be adapted to the corresponding end area, generate the corresponding system operation strategy, pass the safety boundary condition review, and issue it to all control points to execute new adjustment actions and operating states. The cooling capacity prepared by the chiller is distributed to the control end as needed through multiple regulation couplings such as water pump flow rate changes, valve opening adjustments, and pipeline pressure difference adjustments. And through the continuous monitoring and feedback of the sensors, the cooling capacity of demand and supply is always in a dynamically balanced range. The system predicts and calculates the expected air-conditioning load. On the premise of meeting the total cooling capacity demand of the system, with the lowest overall energy consumption of the central air-conditioning system as the control target, through control optimization measures, the joint operation of each device in the cold station and the air system is coordinated. The control model realizes the dynamic tracking and real-time online control of the air-conditioning system load according to the characteristics of air-conditioning load changes, reasonably adjusts the control parameters and states of each device, enables the chilled water to be dynamically adjusted following the load changes, enables the system to have a high strain capacity, analyzes the optimal operating condition points of the water system chiller, chilled water pump, cooling water pump, cooling tower fan, electric butterfly valve, electric two-way valve, electric pressure difference bypass valve, modular air handling unit, and air valve, so as to match the energy supply with the energy demand and minimize the total energy consumption of the air-conditioning system to the greatest extent, and make the operation efficiency of the entire central air-conditioning system optimal.

[0126] In the above solution, the first subsystem adopts a white-box AI + mechanism model, and the energy-saving effect can be expected. The speed of going online, using, and adjusting to take effect is fast; due to the characteristics of being guided by the refrigeration logic mechanism of system equipment, the model has strong universality in use, high correlation of operating logic, the AI white-box mechanism model is interpretable, the computing power demand for the system optimization process is small, the speed of generating energy-saving optimization strategies is fast, the accuracy of the system model is relatively higher, and the credibility of the energy-saving strategy is better.

[0127] In some optional implementation manners, the first subsystem further performs the following steps:

[0128] Based on the operation strategy, obtain the predicted performance and the predicted operation parameters corresponding to the operation strategy;

[0129] Among them, the predicted operation parameters are the operation parameters of the refrigeration equipment when the refrigeration equipment executes the operation strategy predicted; the predicted performance is the performance of the refrigeration equipment when the refrigeration equipment executes the operation strategy predicted.

[0130] In this embodiment, the first subsystem has a prediction and analysis function, and can perform energy efficiency simulation calculations on the refrigeration system under different loads, different working conditions, and different control strategy conditions, and diagnose the on-site energy consumption and efficiency conditions. The AI energy consumption model can be updated according to the configuration of the refrigeration equipment and relevant data, and the performance and operation can be predicted.

[0131] In the above solution, through the simulation and analysis of the first subsystem, the iterative optimization of the AI energy consumption model and the double screening of the operation strategy corresponding to the abnormal data by the second subsystem are realized.

[0132] In some optional embodiments, the first subsystem further performs the following steps:

[0133] Train the AI model according to the actual performance of the refrigeration equipment, the cooling demand at the end of the refrigeration system, the feedback data, and the operation boundary of the refrigeration equipment in the refrigeration system to obtain an updated AI model.

[0134] In the above solution, the AI model continues to self-iterate and update continuously after being deployed to optimize the AI model.

[0135] In some optional embodiments, the first subsystem further performs the following steps:

[0136] If the error between the predicted performance and the actual performance of any refrigeration equipment is greater than the first error; and / or the error between the predicted operation parameters and the actual operation parameters is greater than the second error, then update the AI model.

[0137] In this embodiment, in order to update the AI model more pertinently, the predicted performance is compared with the actual performance, and the predicted operation parameters are compared with the actual operation parameters. If the error is small, the credibility verification of the AI model passes. If the error is large, the credibility verification of the AI model fails, and the current AI model needs to be iterated into the next round of calculation. In this embodiment, the time interval for triggering the comparison is not specifically limited, such as 5 minutes, and the above first error and second error can be the same or different.

[0138] Exemplarily, the minimum granularity of the AI prediction is 5 minutes. When the AI clock cycle reaches the prediction point, the actual working conditions are retrieved and compared with the prediction calculation results (the prediction efficiency is compared with the actual efficiency, and the predicted operating parameters are compared with the actual operating parameters). If the prediction calculation accuracy error is less than 2%, the credibility verification of the AI model passes, and the energy-saving strategy is sent to the second subsystem; if it is greater than 2%, the deviation of the AI model credibility record and the convergence warning mark of the calculation results are recorded, and the current round of strategy is not sent, and it remains in the previous round of credible energy-saving strategy operation. At the same time, the current AI model is iterated into the next round of calculation until the new AI energy-saving optimization strategy passes the credibility verification before allowing the execution of the sending process. At the same time, the new AI model that passes the process continues to perform verification, prediction, and evolution tasks.

[0139] In the above solution, the AI prediction calculation is performed by the first subsystem to realize the iterative optimization of the AI energy consumption model, enabling the AI to continuously self-evolve and improve the prediction effect.

[0140] In some alternative embodiments, the above step S201 can be implemented by but not limited to the following methods:

[0141] Iteratively step by step from the AI initial boundary to the safety boundary corresponding to the refrigeration equipment at a preset step size.

[0142] In this embodiment, preset step sizes corresponding to different models are set. The AI gradually iterates from the initial boundary to the safety boundary corresponding to the BA with an "optimization mechanism of running in small steps" to track the optimal energy-saving adjustment strategy and guide the system to continuously and reliably operate in an energy-saving condition. In this embodiment, the safety boundary corresponding to the refrigeration equipment is set. To ensure the safe operation of the equipment, the safety boundary corresponding to the refrigeration equipment has a safety margin relative to the absolute safety boundary.

[0143] In some alternative embodiments, the first subsystem further performs the following steps:

[0144] In response to an addition instruction for the first refrigeration equipment, add a second model corresponding to the first refrigeration equipment to the AI model; wherein, the first refrigeration equipment is a newly added refrigeration equipment in the refrigeration system; or

[0145] In response to a deletion instruction for the second refrigeration equipment, delete the second model corresponding to the second refrigeration equipment from the AI model; wherein, the second refrigeration equipment is an existing refrigeration equipment in the refrigeration system; or

[0146] In response to a modification instruction for the third refrigeration equipment, modify the second model corresponding to the third refrigeration equipment in the AI model; wherein, the third refrigeration equipment is an existing refrigeration equipment in the refrigeration system.

[0147] In this embodiment, AI models corresponding to each refrigeration device are set. Therefore, when there are drastic changes in load or refrigeration structure at the deployment site, only the corresponding model settings need to be adjusted, and the system configuration parameters are updated to complete the corresponding upgrade. Without redeploying the entire AI system, the structural changes of the refrigeration system can be flexibly handled.

[0148] In some alternative embodiments, determining whether the operation strategy meets the issuance condition in step S203 above can be achieved by but not limited to the following methods:

[0149] For the refrigeration device, determine whether the operation strategy of the refrigeration device is within the safety boundary range corresponding to the refrigeration device, and determine whether the most recent prediction data of the refrigeration device is normal; the prediction data is the data corresponding to the operation strategy obtained by the first subsystem based on the operation strategy.

[0150] If the operation strategies of all refrigeration devices are within the corresponding safety boundary ranges, and the prediction data of all refrigeration devices is normal, it is determined that the operation strategy meets the issuance condition; otherwise, it is determined that the operation strategy does not meet the issuance condition.

[0151] In this embodiment, a dual safeguard mechanism of the first subsystem + the second subsystem is set. The first layer is set in the first subsystem (AI software layer control), and the other layer is set in the second subsystem (BA group control layer control). The second subsystem performs a second screening of the operation strategy based on whether the operation strategy is within the corresponding safety boundary range and whether the prediction data of the refrigeration device is normal; if the operation strategies of all refrigeration devices are within the corresponding safety boundary ranges, and the prediction data of all refrigeration devices is normal, it is determined that the operation strategy meets the issuance condition; the second subsystem only issues operation strategies that conform to the safety logic to the corresponding control points.

[0152] This embodiment does not make specific limitations on the determination of whether the prediction data is normal, such as whether there are mutations, etc. The above prediction data can be issued by the first subsystem. The prediction data may include the above prediction efficiency and prediction operation parameters.

[0153] In the above solution, the second subsystem determines whether the operation strategy is within the corresponding safety boundary range and whether the prediction data of the refrigeration device is normal; and only when the operation strategies of all refrigeration devices are within the corresponding safety boundary ranges and the prediction data of all refrigeration devices is normal, will it issue operation strategies that conform to the safety logic to the corresponding control points, realizing a second screening of the operation strategy and ensuring the safety of the operation strategy.

[0154] In some alternative embodiments, the second subsystem is further configured to:

[0155] If an exit instruction is received, or if it is determined that the control of the first subsystem is abnormal, the refrigeration equipment is controlled based on the BA control strategy;

[0156] In response to the access instruction, re-associate with the first subsystem.

[0157] In this embodiment, the second subsystem can achieve independent control, that is, AI switching can be performed.

[0158] In implementation, the user can input / exit the control of the first subsystem as needed. When it is necessary to exit the control of the first subsystem, an AI exit instruction is triggered. The second subsystem exits the association with the first subsystem in response to this AI exit instruction. At this time, the second subsystem will not execute the operation strategy from the first subsystem and can control the refrigeration equipment based on the BA control strategy of the second subsystem;

[0159] When it is necessary to input the control of the first subsystem, an AI access instruction is triggered, and the second subsystem re-associates with the first subsystem. At this time, the second subsystem will receive the operation strategy from the first subsystem again.

[0160] To ensure the safe operation of the equipment, automatic switching of the first subsystem control can also be performed. When an abnormality occurs in the first subsystem control (such as large fluctuations in the first subsystem or AI software failure), the first subsystem abnormality is triggered, and the second subsystem will also exit the association with the first subsystem.

[0161] In implementation, during the AI switching process, the first subsystem and the second subsystem need to track each other to reduce the system disturbance caused by large changes in the set value during the control mode switching.

[0162] In the above solution, by triggering the AI switching instruction or monitoring the abnormality of the first subsystem control, AI switching is performed to reduce the occurrence of AI malfunction operation.

[0163] In some optional implementation manners, the first subsystem is carried on the infrastructure management platform;

[0164] Refer to Figure 5 As shown, it is the system architecture diagram of the second energy-saving refrigeration system provided by the embodiment of the present application, Figure 5 Taking the infrastructure management platform as SiteWeb6 (a comprehensive energy management platform) as an example, other infrastructure management platforms can also be used in implementation.

[0165] Exemplarily, the AI management layer includes software and hardware structures such as SiteWeb6 software, AI Building software, one AI energy-saving server, and AI Station (an interactive screen operated by users). The AI energy-saving transformation can be implemented step by step to improve the flexibility of solution deployment. For the architecture part, the basic conditions for realizing AI energy saving can be completed by first establishing the system regulation and control capabilities and building a data lake to achieve basic control. Then, the SiteWeb6 platform and the construction of the AI Building execution model operation and optimization capabilities can be deployed to adapt to the on-site conditions of different configuration conditions.

[0166] The infrastructure management platform is used to send the feedback data uploaded by the second subsystem to the first subsystem; and, send the operation strategy generated by the first subsystem to the second subsystem.

[0167] In this embodiment, the second subsystem collects and uploads operation data, which needs to be fed back to the first subsystem to judge the adjustment effect of the operation strategy. Refer to Figure 5 As shown, the second subsystem uploads the feedback data to SiteWeb6 through the data communication API (program interface), and the first subsystem receives the feedback data; the second subsystem receives the operation strategy transmitted by SiteWeb6 through the AI access interface.

[0168] In some optional embodiments, the infrastructure management platform is further used to provide a visual interface; wherein, the visual interface includes an AI interface and a device interface.

[0169] The AI interface includes some or all of the AI main interface, AI energy-saving optimization strategy interface, AI cold source parameter optimization interface, AI cockpit management interface, alarm management interface, and AI model training management interface;

[0170] Refer to Figure 6 As shown, taking the AI interface including the AI main interface, AI energy-saving optimization strategy interface, AI cold source parameter optimization interface, AI cockpit management interface, alarm management interface, and AI model training management interface as an example.

[0171] The above Figure 6 is only an exemplary illustration. In implementation, some of the above interfaces or other AI interfaces can be set. In addition, the system background configuration function can configure and manage the AI interface.

[0172] Refer to Figure 7 As shown, the AI energy-saving optimization strategy interface can include three parts: energy consumption statistics, dynamic energy-saving strategy, and dynamic process of energy-saving strategy optimization model (power change curves of different refrigeration equipment); Figure 7Taking the energy consumption statistics including Power Usage Effectiveness (PUE), cumulative power consumption, and cumulative power saving, and the dynamic energy-saving strategies including control commands, operating conditions, chilled water outlet temperature, number of chillers, chilled water return temperature, chilled water pumps, chilled water temperature difference, and cooling water pumps as an example. During implementation, other information related to the strategies can also be displayed.

[0173] Refer to Figure 8 As shown, the AI cockpit management interface can include several pieces of information such as the reward value curve, energy efficiency index curve, current day's energy efficiency index, strategy history, temperature, and air conditioner operating status; among them, Figure 8 Taking the current day's energy efficiency index including PUE, Cooling Load Factor (CLF), and IT power consumption, the strategy history including Strategies 1 - 3, the temperature including Temperatures 1 - 3, and the air conditioner operating status including Statuses 1 - 3 as an example. During implementation, other information related to AI management can also be displayed.

[0174] Refer to Figure 9 As shown, the alarm management interface can include information such as location, device type, device name, alarm base class, alarm level, whether the event is confirmed, whether the event is over, and convergence events; Figure 9 Taking the display of the alarm level as an example, during implementation, after selecting other information, the relevant content of the other information will be displayed.

[0175] The device interface includes a 3D view of the refrigeration equipment and some or all of the operating parameter interfaces.

[0176] In this embodiment, a 3D visualization display of refrigeration equipment such as control cabinets, chiller hosts, water pumps, and cooling towers in the refrigeration system can be provided to intuitively calculate the energy consumption of the cold source system; the operating parameter information of devices such as chillers, cooling pumps, chilled water pumps, cooling towers, electric butterfly valves, electric two-way valves, electric differential pressure bypass valves, and sensors can also be monitored and displayed.

[0177] During implementation, the system background configuration function can configure and manage multiple interfaces, support the application expansion of the later operation of the refrigeration system, and meet the integration and access of other mechanical and electrical equipment subsystems and operation information subsystems in this system to adapt to the construction and operation in the later use scenarios. For example, the 2D configuration tool realizes the presentation of the data center site, facilities, logical relationships, and key indicators through configuration. The 2D configuration tool provides room alarm overviews, server communication status, historical indicators, configuration routing, indicator ring meters, multi-indicator graphs, tables, picture icons, text, etc.

[0178] In some alternative implementation manners, the infrastructure management platform is further used for alarm management.

[0179] Exemplary. When problems occur during the operation of the device, the alarm system will issue an alarm and record the event; when the management personnel need to modify the necessary parameters, the system will record events such as the modification time (but set the modification permission) and the modifier, for future query.

[0180] This system can support the access capabilities of multiple remote sites according to customized requirements, and support multi-user remote access for browsing and monitoring.

[0181] In addition, the system has "user verification" and "user management" functions to manage user operators, prevent unauthorized operations by irrelevant personnel, and ensure the security of system operation management. "User verification" is used to verify the identity of operators, and only after the verification of their user names and passwords can they operate on the system equipment. "User management" is used to manage user operators, such as adding users, modifying users, and deleting users, etc.

[0182] In some optional embodiments, the infrastructure management platform also provides a reporting function, which can perform real-time analysis and statistical analysis on real-time data, alarm data, historical data, operation policies, power consumption, PUE, etc., can draw corresponding data feedback diagrams, and can manually or automatically generate various data reports that meet customer requirements.

[0183] The embodiment of the present application provides a first energy-saving refrigeration method, which is applied to the above-mentioned first subsystem. Refer to Figure 10 As shown, this method includes:

[0184] Step S1001: Based on the end cooling demand of the refrigeration system, feedback data, and the operation boundary of the refrigeration equipment in the refrigeration system, generate an operation strategy for the refrigeration equipment through an artificial intelligence AI model; wherein, the AI model includes the first model of the refrigeration system and the second model corresponding to the refrigeration equipment;

[0185] Step S1002: Send the operation strategy to the second subsystem, so that when the operation strategy meets the sending condition, the second subsystem sends the operation strategy to the refrigeration equipment, so that the refrigeration equipment refrigerates according to the operation strategy.

[0186] In some optional embodiments, both the first model and the second model include white-box mechanism models.

[0187] In some optional embodiments, it further includes:

[0188] Based on the operation strategy, obtain the predicted efficiency and the predicted operation parameters corresponding to the operation strategy;

[0189] Wherein, the predicted operating parameters are the operating parameters of the refrigeration equipment when the refrigeration equipment executes the operating strategy, and the predicted efficiency is the efficiency of the refrigeration equipment when the refrigeration equipment executes the operating strategy.

[0190] In some optional embodiments, it further includes:

[0191] Training the AI model according to the actual efficiency of the refrigeration equipment, the cooling demand at the end of the refrigeration system, the feedback data, and the operating boundary of the refrigeration equipment in the refrigeration system to obtain an updated AI model.

[0192] In some optional embodiments, it further includes:

[0193] If the error between the predicted efficiency and the actual efficiency of any refrigeration equipment is greater than the first error; and / or the error between the predicted operating parameters and the actual operating parameters is greater than the second error, then update the AI model.

[0194] In some optional embodiments, it further includes:

[0195] Iteratively move from the AI initial boundary to the safety boundary corresponding to the refrigeration equipment step by step with a preset step size.

[0196] In some optional embodiments, it further includes:

[0197] In response to an addition instruction for a first refrigeration equipment, adding a second model corresponding to the first refrigeration equipment to the AI model; wherein, the first refrigeration equipment is a newly added refrigeration equipment in the refrigeration system; or

[0198] In response to a deletion instruction for a second refrigeration equipment, deleting the second model corresponding to the second refrigeration equipment from the AI model; wherein, the second refrigeration equipment is an existing refrigeration equipment in the refrigeration system; or

[0199] In response to a modification instruction for a third refrigeration equipment, modifying the second model corresponding to the third refrigeration equipment in the AI model; wherein, the third refrigeration equipment is an existing refrigeration equipment in the refrigeration system.

[0200] In some optional embodiments, the first subsystem is carried on an infrastructure management platform;

[0201] Issuing the operating strategy to the second subsystem includes:

[0202] Sending the operating strategy to the second subsystem through the infrastructure management platform;

[0203] The method further includes:

[0204] Receive the feedback data uploaded by the second subsystem through the infrastructure management platform.

[0205] In some alternative embodiments, the infrastructure management platform further includes a visualization interface;

[0206] wherein, the visualization interface includes an AI interface and a device interface; the AI interface includes some or all of an AI main interface, an AI energy-saving optimization strategy interface, an AI cold source parameter optimization interface, an AI cockpit management interface, an alarm management interface, and an AI model training management interface; the device interface includes some or all of a 3D view of the refrigeration equipment and an operating parameter interface.

[0207] In some alternative embodiments, the method further includes:

[0208] Perform alarm management through the infrastructure management platform.

[0209] The embodiment of the present application provides a second energy-saving refrigeration method, which is applied to the above-mentioned second subsystem. Refer to Figure 11 As shown, the method includes:

[0210] Step S1101: Receive the operation strategies for each refrigeration equipment in the refrigeration system; wherein, the operation strategies are generated by the first subsystem through an artificial intelligence (AI) model based on the end cooling demand of the refrigeration system, the feedback data, and the operation boundaries of the refrigeration equipment in the refrigeration system; the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment.

[0211] Step S1102: Determine whether the operation strategies meet the issuance conditions, and when the operation strategies meet the issuance conditions, send the operation strategies to the refrigeration equipment so that the refrigeration equipment cools according to the operation strategies.

[0212] In some alternative embodiments, determining whether the operation strategies meet the issuance conditions includes:

[0213] For the refrigeration equipment, determine whether the operation strategies of the refrigeration equipment are within the corresponding safety boundary range of the refrigeration equipment, and determine whether the most recent prediction data of the refrigeration equipment is normal; the prediction data is the data corresponding to the operation strategies obtained by the first subsystem based on the operation strategies.

[0214] If the operation strategies of the refrigeration equipment are all within the corresponding safety boundary ranges, and the prediction data of the refrigeration equipment are all normal, it is determined that the operation strategies meet the issuance conditions; otherwise, it is determined that the operation strategies do not meet the issuance conditions.

[0215] In some optional embodiments, it further includes:

[0216] If an exit instruction is received, or it is determined that the control of the first subsystem is abnormal, the refrigeration equipment is controlled to operate based on the BA control strategy;

[0217] In response to the access instruction, re - associate with the first subsystem.

[0218] Based on the same inventive concept as the above - mentioned energy - saving refrigeration method on the first - subsystem side, an embodiment of the present application provides an energy - saving refrigeration device. Referring to Figure 12 as shown, the energy - saving refrigeration device 1200 includes:

[0219] An optimization module 1201, configured to generate an operation strategy for the refrigeration equipment through an AI model based on the end cooling demand, feedback data, and the operation boundaries of each refrigeration equipment in the refrigeration system; wherein, the AI model includes a white - box principle model of the refrigeration system and a second model corresponding to each refrigeration equipment;

[0220] A strategy issuing module 1202, configured to issue the operation strategy to the second subsystem, so that when the operation strategy meets the issuing condition, the second subsystem sends the operation strategy to the refrigeration equipment, so that the refrigeration equipment refrigerates according to the operation strategy.

[0221] In some optional embodiments, both the first model and the second model include a white - box mechanism model.

[0222] In some optional embodiments, it further includes a simulation and prediction module 1203, configured to:

[0223] Based on the operation strategy, obtain the predicted efficiency and the predicted operation parameters corresponding to the operation strategy;

[0224] Wherein, the predicted operation parameters are the operation parameters of the refrigeration equipment when the refrigeration equipment executes the operation strategy; the predicted efficiency is the efficiency of the refrigeration equipment when the refrigeration equipment executes the operation strategy.

[0225] In some optional embodiments, the optimization module 1201 is further configured to:

[0226] Train the AI model according to the actual efficiency of the refrigeration equipment, and the end cooling demand, feedback data of the refrigeration system, and the operation boundaries of the refrigeration equipment in the refrigeration system to obtain an updated AI model.

[0227] In some optional embodiments, the optimization module 1201 is further configured to:

[0228] If the error between the predicted efficiency and the actual efficiency of any refrigeration device is greater than the first error; and / or the error between the predicted operating parameters and the actual operating parameters is greater than the second error, then update the AI model.

[0229] In some alternative embodiments, the optimization module 1201 is specifically configured to:

[0230] Iteratively step by step from the AI initial boundary to the safety boundary corresponding to the refrigeration device.

[0231] In some alternative embodiments, it further includes an adjustment module 1204, configured to:

[0232] In response to an addition instruction for a first refrigeration device, add a second model corresponding to the first refrigeration device in the AI model; wherein, the first refrigeration device is a newly added refrigeration device in the refrigeration system; or

[0233] In response to an addition instruction for a second refrigeration device, delete the second model corresponding to the second refrigeration device in the AI model; wherein, the second refrigeration device is an existing refrigeration device in the refrigeration system; or

[0234] In response to a modification instruction for a third refrigeration device, modify the second model corresponding to the third refrigeration device in the AI model; wherein, the third refrigeration device is an existing refrigeration device in the refrigeration system.

[0235] In some alternative embodiments, the first subsystem is carried on an infrastructure management platform;

[0236] The policy distribution module 1202 is specifically configured to:

[0237] Send the operation policy to the second subsystem through the infrastructure management platform;

[0238] It further includes a receiving module 1205, configured to:

[0239] Receive the feedback data uploaded by the second subsystem through the infrastructure management platform.

[0240] In some alternative embodiments, the infrastructure management platform further includes a visualization interface;

[0241] Wherein, the visualization interface includes an AI interface and a device interface; the AI interface includes some or all of an AI main interface, an AI energy-saving optimization strategy interface, an AI cold source parameter optimization interface, an AI cockpit management interface, an alarm management interface, and an AI model training management interface; the device interface includes some or all of a 3D view of the refrigeration device and an operating parameter interface.

[0242] In some optional embodiments, an alarm module 1206 is further included, which is used for:

[0243] Performing alarm management through the infrastructure management platform.

[0244] Since this device is the device in the energy-saving refrigeration method in the embodiments of the present application, and the principle of the device to solve problems is similar to that of this method, the implementation of this device can refer to the implementation of this method, and the repeated parts will not be described again.

[0245] Based on the same inventive concept as the energy-saving refrigeration method on the second subsystem side above, an energy-saving refrigeration device is provided in the embodiments of the present application. Referring to Figure 13 As shown, the energy-saving refrigeration device 1300 includes:

[0246] A policy receiving module 1301, which is used to receive the operation policies for each refrigeration device in the refrigeration system; wherein, the operation policies are generated by the first subsystem through an artificial intelligence AI model based on the end cooling demand of the refrigeration system, the feedback data, and the operation boundaries of the refrigeration devices in the refrigeration system; the AI model includes the first model of the refrigeration system and the second model corresponding to the refrigeration device;

[0247] A policy filtering module 1302, which is used to determine whether the operation policies meet the distribution conditions, and when the operation policies meet the distribution conditions, send the operation policies to the refrigeration devices, so that the refrigeration devices refrigerate according to the operation policies.

[0248] In some optional embodiments, the policy filtering module 1302 is specifically used for:

[0249] For the refrigeration device, determine whether the operation policy of the refrigeration device is within the safety boundary range corresponding to the refrigeration device, and determine whether the most recent prediction data of the refrigeration device is normal; the prediction data is the data corresponding to the operation policy obtained by the first subsystem based on the operation policy.

[0250] If the operation policies of the refrigeration devices are all within the corresponding safety boundary ranges, and the prediction data of the refrigeration devices are all normal, it is determined that the operation policies meet the distribution conditions; otherwise, it is determined that the operation policies do not meet the distribution conditions.

[0251] In some optional embodiments, a switching module 1303 is further included, which is used for:

[0252] If an exit instruction is received, or it is determined that the control of the first subsystem is abnormal, control the operation of the refrigeration device based on the BA control strategy;

[0253] In response to the access instruction, re-associate with the first subsystem.

[0254] Since this device is the device in the energy-saving refrigeration method in the embodiments of the present application, and the principle of the device to solve problems is similar to that of the method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be elaborated.

[0255] Based on the same technical concept, the embodiments of the present application also provide a first subsystem 1400, as Figure 14 shown, including at least one processor 1401 and a memory 1402 connected to the at least one processor. In the embodiments of the present application, the specific connection medium between the processor 1401 and the memory 1402 is not limited. Figure 14 Taking the connection between the processor 1401 and the memory 1402 through a bus 1403 as an example. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 14 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0256] Among them, the processor 1401 is the control center of the first subsystem, and can connect various parts of the first subsystem through various interfaces and lines. By running or executing the instructions stored in the memory 1402 and calling the data stored in the memory 1402, data processing can be achieved. Optionally, the processor 1401 may include one or more processing units. The processor 1401 may integrate an application processor and a modem processor. Among them, the application processor mainly processes the operating system, user interface, application programs, etc., and the modem processor mainly processes the issued instructions. It can be understood that the above-mentioned modem processor may not be integrated into the processor 1401. In some embodiments, the processor 1401 and the memory 1402 can be implemented on the same chip, and in some embodiments, they can also be separately implemented on independent chips.

[0257] The processor 1401 can be a general-purpose processor, such as a central processing unit (CPU), a digital signal processor, an application specific integrated circuit (ASIC), a field programmable gate array or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the energy-saving refrigeration method can be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0258] The memory 1402, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. The memory 1402 may include at least one type of storage medium, for example, it may include flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic memory, magnetic disk, optical disc, and so on. The memory 1402 is any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 1402 in the embodiments of the present application may also be a circuit or any other device capable of implementing a storage function, for storing program instructions and / or data.

[0259] In the embodiments of the present application, the memory 1402 stores a computer program, and when the program is executed by the processor 1401, the processor 1401 is caused to execute:

[0260] Based on the end cooling demand, feedback data of the refrigeration system, and the operating boundaries of the refrigeration equipment in the refrigeration system, generate an operating strategy for the refrigeration equipment through an artificial intelligence AI model; wherein, the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment;

[0261] Send the operating strategy to the second subsystem, so that when the operating strategy meets the sending condition, the second subsystem sends the operating strategy to the refrigeration equipment, so that the refrigeration equipment refrigerates according to the operating strategy.

[0262] In some optional embodiments, both the first model and the second model include a white-box mechanism model.

[0263] In some optional embodiments, the processor 1401 further executes:

[0264] Based on the operating strategy, obtain the predicted performance and the predicted operating parameters corresponding to the operating strategy;

[0265] Wherein, the predicted operating parameter is the operating parameter of the refrigeration device when the refrigeration device executes the operating strategy; the predicted efficiency is the efficiency of the refrigeration device when the refrigeration device executes the operating strategy.

[0266] In some alternative embodiments, the processor 1401 further executes:

[0267] Train the AI model according to the actual efficiency of the refrigeration device, the cooling demand at the end of the refrigeration system, the feedback data, and the operating boundary of the refrigeration device in the refrigeration system to obtain an updated AI model.

[0268] In some alternative embodiments, the processor 1401 further executes:

[0269] If the error between the predicted efficiency and the actual efficiency of any refrigeration device is greater than the first error; and / or the error between the predicted operating parameter and the actual operating parameter is greater than the second error, then update the AI model.

[0270] In some alternative embodiments, the processor 1401 further executes:

[0271] Iteratively step by step from the AI initial boundary to the safety boundary corresponding to the refrigeration device.

[0272] In some alternative embodiments, the processor 1401 further executes:

[0273] In response to an addition instruction for a first refrigeration device, add a second model corresponding to the first refrigeration device to the AI model; wherein, the first refrigeration device is a newly added refrigeration device in the refrigeration system; or

[0274] In response to a deletion instruction for a second refrigeration device, delete the second model corresponding to the second refrigeration device from the AI model; wherein, the second refrigeration device is an existing refrigeration device in the refrigeration system; or

[0275] In response to a modification instruction for a third refrigeration device, modify the second model corresponding to the third refrigeration device in the AI model; wherein, the third refrigeration device is an existing refrigeration device in the refrigeration system.

[0276] In some alternative embodiments, the first subsystem is carried on an infrastructure management platform;

[0277] The processor 1401 specifically executes:

[0278] Send the operating strategy to the second subsystem through the infrastructure management platform;

[0279] The processor 1401 further executes:

[0280] Receive the feedback data uploaded by the second subsystem through the infrastructure management platform.

[0281] In some alternative embodiments, the infrastructure management platform further includes a visualization interface;

[0282] Wherein, the visualization interface includes an AI interface and a device interface; the AI interface includes some or all of an AI main interface, an AI energy-saving optimization strategy interface, an AI cold source parameter optimization interface, an AI cockpit management interface, an alarm management interface, and an AI model training management interface; the device interface includes some or all of a 3D view of the refrigeration equipment and an operating parameter interface.

[0283] In some alternative embodiments, the processor 1401 further executes:

[0284] Conduct alarm management through the infrastructure management platform.

[0285] Since the first subsystem is the first subsystem in the energy-saving refrigeration method in the embodiments of the present application, and the principle of the first subsystem to solve problems is similar to that of the method, the implementation of the first subsystem can refer to the implementation of the method, and the repeated parts will not be elaborated.

[0286] The embodiments of the present application further provide a second subsystem, including at least one processor and at least one memory, wherein the memory stores a computer program, and when the program is executed by the processor, the processor is caused to execute:

[0287] Receive the operation strategies for each refrigeration equipment in the refrigeration system; wherein, the operation strategies are generated by the first subsystem based on the end cooling demand of the refrigeration system, the feedback data, and the operation boundaries of the refrigeration equipment in the refrigeration system through an artificial intelligence AI model; the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment;

[0288] Determine whether the operation strategies meet the distribution conditions, and when the operation strategies meet the distribution conditions, send the operation strategies to the refrigeration equipment so that the refrigeration equipment refrigerates according to the operation strategies.

[0289] In some alternative embodiments, the processor specifically executes:

[0290] For the refrigeration equipment, determine whether the operation strategies of the refrigeration equipment are within the safety boundary range corresponding to the refrigeration equipment, and determine whether the most recent prediction data of the refrigeration equipment is normal; the prediction data is the data corresponding to the operation strategies obtained by the first subsystem based on the operation strategies.

[0291] If the operation strategies of the refrigeration equipment are all within the corresponding safety boundaries, and the predicted data of the refrigeration equipment are all normal, it is determined that the operation strategies meet the distribution conditions; otherwise, it is determined that the operation strategies do not meet the distribution conditions.

[0292] In some alternative embodiments, the processor further executes:

[0293] If an exit instruction is received, or it is determined that the first subsystem is abnormally controlled, the refrigeration equipment is controlled to operate based on the BA control strategy;

[0294] In response to the access instruction, re-associate with the first subsystem.

[0295] Since the second subsystem is the second subsystem in the energy-saving refrigeration method in the embodiments of the present application, and the principle of the first subsystem to solve problems is similar to that of this method, the implementation of the second subsystem can refer to the implementation of this method, and the repeated parts will not be elaborated.

[0296] Based on the same technical concept, the embodiments of the present application also provide a computer-readable storage medium, which stores a computer program executable by an electronic device. When the program runs on the electronic device, the electronic device is caused to execute the steps of the above energy-saving refrigeration method.

[0297] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0298] The present application is described with reference to the flowcharts and / or block diagrams of the method, device (system), and computer program product according to the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0299] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the processes Figure 1 or blocks Figure 1 specified in one or more of the blocks.

[0300] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the processes Figure 1 or blocks Figure 1 specified in one or more of the blocks.

[0301] Although the preferred embodiments of the present application have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present application.

[0302] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. An energy-saving refrigeration method, characterized in that, Applied to the first subsystem, the method includes: Based on the cooling demand at the end of the refrigeration system, feedback data, and the operating boundaries of the refrigeration equipment in the refrigeration system, an operating strategy for the refrigeration equipment is generated through an artificial intelligence (AI) model. The AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration equipment. The operating strategy is sent to the second subsystem, so that when the operating strategy meets the sending condition, the second subsystem sends the operating strategy to the refrigeration equipment, so that the refrigeration equipment cools according to the operating strategy.

2. The energy-saving refrigeration method according to claim 1, wherein Both the first model and the second model include a white-box mechanism model.

3. The energy-saving refrigeration method according to claim 1, characterized in that, It further includes: Based on the operating strategy, the predicted efficiency and the predicted operating parameters corresponding to the operating strategy are obtained. The predicted operating parameters are the operating parameters of the refrigeration equipment when the refrigeration equipment executes the operating strategy. The predicted efficiency is the efficiency of the refrigeration equipment when the refrigeration equipment executes the operating strategy.

4. The energy-saving refrigeration method according to claim 1, wherein It further includes: The AI model is trained according to the actual efficiency of the refrigeration equipment, the cooling demand at the end of the refrigeration system, feedback data, and the operating boundaries of the refrigeration equipment in the refrigeration system to obtain an updated AI model.

5. The energy-saving refrigeration method according to claim 3, wherein It further includes: If the error between the predicted efficiency and the actual efficiency of any refrigeration equipment is greater than the first error; and / or the error between the predicted operating parameters and the actual operating parameters is greater than the second error, the AI model is updated.

6. The energy-saving refrigeration method according to claim 1, wherein It further includes: Iteratively move from the AI initial boundary to the safety boundary corresponding to the refrigeration equipment step by step at a preset step size.

7. The energy-saving refrigeration method according to claim 1, characterized in that, It further includes: In response to an addition instruction for the first refrigeration equipment, a second model corresponding to the first refrigeration equipment is added to the AI model. The first refrigeration equipment is a newly added refrigeration equipment in the refrigeration system; or In response to a deletion instruction for the second refrigeration equipment, the second model corresponding to the second refrigeration equipment is deleted from the AI model. The second refrigeration equipment is an existing refrigeration equipment in the refrigeration system; or In response to a modification instruction for the third refrigeration equipment, the second model corresponding to the third refrigeration equipment is modified in the AI model. The third refrigeration equipment is an existing refrigeration equipment in the refrigeration system.

8. The energy-saving refrigeration method according to any one of claims 1 to 7, characterized in that, The first subsystem is carried on an infrastructure management platform; Sending the operating strategy to the second subsystem includes: Sending the operating strategy to the second subsystem through the infrastructure management platform; The method further includes: Receiving the feedback data uploaded by the second subsystem through the infrastructure management platform.

9. The energy-saving refrigeration method according to claim 8, wherein, The infrastructure management platform further includes a visualization interface; The visualization interface includes an AI interface and a device interface. The AI interface includes some or all of an AI main interface, an AI energy-saving optimization strategy interface, an AI cold source parameter optimization interface, an AI cockpit management interface, an alarm management interface, and an AI model training management interface. The device interface includes some or all of a 3D view of the refrigeration equipment and an operating parameter interface.

10. The energy-saving refrigeration method according to claim 8, characterized in that, The method further includes: Alarm management is performed through the infrastructure management platform.

11. An energy-saving refrigeration method, characterized in that, Applied to the second subsystem, the method includes: Receiving operation strategies for each refrigeration device in the refrigeration system; wherein, the operation strategies are generated by the first subsystem based on the end cooling demand of the refrigeration system, feedback data, and the operation boundaries of the refrigeration devices in the refrigeration system through an artificial intelligence AI model; the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration device; Determining whether the operation strategies meet the distribution conditions, and when the operation strategies meet the distribution conditions, sending the operation strategies to the refrigeration devices so that the refrigeration devices refrigerate according to the operation strategies.

12. The energy-saving refrigeration method according to claim 11, wherein Determining whether the operation strategies meet the distribution conditions includes: For the refrigeration device, determining whether the operation strategy of the refrigeration device is within the safety boundary range corresponding to the refrigeration device, and determining whether the most recent prediction data of the refrigeration device is normal; the prediction data is the data corresponding to the operation strategy obtained by the first subsystem based on the operation strategy; If the operation strategies of the refrigeration devices are all within the corresponding safety boundary ranges and the prediction data of the refrigeration devices are all normal, it is determined that the operation strategies meet the distribution conditions; otherwise, it is determined that the operation strategies do not meet the distribution conditions.

13. The energy-saving refrigeration method according to claim 11, wherein, It further includes: If an exit instruction is received, or it is determined that the control of the first subsystem is abnormal, controlling the operation of the refrigeration device based on the BA control strategy; In response to an access instruction, re-associating with the first subsystem.

14. An energy-saving refrigeration system, characterized in that, The energy-saving refrigeration system includes: a first subsystem and a second subsystem; The first subsystem is used to generate operation strategies for the refrigeration devices through an artificial intelligence AI model based on the end cooling demand of the refrigeration system, feedback data, and the operation boundaries of the refrigeration devices in the refrigeration system; wherein, the AI model includes a first model of the refrigeration system and a second model corresponding to the refrigeration device; The second subsystem is used to receive the operation strategies; and when the operation strategies meet the distribution conditions, sending the operation strategies to the refrigeration devices so that the refrigeration devices refrigerate according to the operation strategies.

Images