Water-cooling refrigeration station system energy-saving optimization control method based on deep reinforcement learning

Abstract

The invention relates to the technical field of energy-saving control, in particular to a water-cooling refrigeration station system energy-saving optimization control method based on deep reinforcement learning, and the method comprises the steps: collecting the environment data of a water-cooling refrigeration station system and the operation data of refrigeration equipment; the energy efficiency score of the refrigeration equipment is calculated, and the load rate of the refrigeration equipment is calculated; calculating a load recommendation rate of the refrigeration equipment based on the energy efficiency score and the load rate of the refrigeration equipment; importing the energy efficiency score and the load recommendation rate of the refrigeration equipment into a clustering analysis algorithm, performing clustering analysis on the refrigeration equipment of the water-cooling refrigeration station system, and dividing the refrigeration equipment into a high-efficiency operation mode, a balanced operation mode and a low-efficiency operation mode; and the refrigeration equipment in the low-efficiency operation mode is extracted, and the corresponding water pump frequency and the cooling tower fan rotating speed are adjusted.

Description

Background Art

[0002] Water-cooled refrigeration station systems are widely used in large buildings, data centers, industrial cooling and other fields. They are energy-intensive facilities, and their energy consumption usually accounts for 30%-50% of the total building energy consumption. Reducing the energy consumption of the refrigeration station has become a key link in energy conservation and emission reduction.

[0003] Disadvantages of the prior art: Traditional control methods are mostly based on fixed rules or simple feedback mechanisms (such as adjusting the start and stop of equipment according to the set temperature), which are difficult to cope with dynamic environmental changes and complex coupling relationships of equipment; In the case of partial load, refrigeration equipment may operate inefficiently for a long time, resulting in energy waste. Traditional methods do not quantitatively evaluate the energy efficiency and load status of refrigeration equipment, and cannot distinguish between high-efficiency and low-efficiency operation modes of equipment, resulting in "one-size-fits-all" control, for example, all equipment synchronously adjusts frequency or speed, ignoring individual performance differences; Existing control strategies are based on fixed design parameters, but in actual operation, environmental temperature and load demand change frequently, and static strategies cannot optimize equipment combinations and operation parameters in real time, resulting in some equipment being in an inefficient interval for a long time; The start-stop scheduling and parameter adjustment of equipment rely on the experience of operation and maintenance personnel, lacking an automated energy efficiency optimization mechanism, resulting in lagging response and energy waste. Summary of the Invention

[0004] The main purpose of the present invention is to provide an energy-saving optimization control method for a water-cooled refrigeration station system based on deep reinforcement learning, which accurately depicts the actual energy efficiency of equipment through an energy efficiency scoring formula; introduces the load rate and load recommendation rate to quantify the equipment load status and optimization potential, providing data support for refined control; adopts the K-Means clustering algorithm to divide equipment into three operation modes based on the energy efficiency score and load recommendation rate, formulates differentiated strategies for different modes, makes full use of high-cost-effective equipment, accurately adjusts the pump frequency and the speed of the cooling tower fan, and specifically improves energy efficiency, solving the defect of "undifferentiated control" of traditional methods and realizing the optimal allocation of resources; The whole process control is closed-loop, without manual intervention, and can respond to environmental and load changes in real time, adapting to complex working conditions.

[0005] The technical solution of the present invention is as follows: In the first aspect, an energy-saving optimization control method for a water-cooled refrigeration station system based on deep reinforcement learning is proposed, and the method includes the following steps: S1. Collect environmental data and refrigeration equipment operation data of the water-cooled refrigeration station system; S2. Calculate the energy efficiency score of the refrigeration equipment and calculate the load rate of the refrigeration equipment; S3. Calculate the load recommendation rate of the refrigeration equipment based on the energy efficiency score and load rate of the refrigeration equipment; S4. Import the energy efficiency score and load recommendation rate of the refrigeration equipment into the clustering analysis algorithm, perform clustering analysis on the refrigeration equipment in the water-cooled refrigeration station system, and divide the refrigeration equipment into high-efficiency operation mode, balanced operation mode, and low-efficiency operation mode; S5. Extract the refrigeration equipment in the low-efficiency operation mode and adjust the corresponding water pump frequency and cooling tower fan speed.

[0006] Preferably, the environmental data in S1 includes the cooling water inlet temperature and the cooling water outlet temperature; the refrigeration equipment operation data includes the refrigeration capacity of the refrigeration equipment, the power consumption of the refrigeration equipment, the water pump frequency, and the cooling tower fan speed.

[0007] Preferably, the calculation formula for the energy efficiency score in S2 is: ; Where is the energy efficiency score of the i-th refrigeration equipment, is the refrigeration capacity of the i-th refrigeration equipment, is the power consumption of the i-th refrigeration equipment, is the cooling water inlet temperature of the i-th refrigeration equipment, is the cooling water outlet temperature of the i-th refrigeration equipment, is the designed optimal temperature difference of the refrigeration equipment, , are weight factors respectively, and .

[0008] Preferably, the calculation formula for the load rate in S2 is: ; Where is the load rate of the i-th refrigeration equipment, is the maximum refrigeration capacity of the i-th refrigeration equipment.

[0009] Preferably, the calculation formula for the load recommendation rate in S3 is: ; Where is the load recommendation rate of the i-th refrigeration equipment, n is the total number of refrigeration equipment, , are weight factors respectively, and .

[0010] Preferably, S4 includes the following specific steps: S41. Extract the energy efficiency score and the load recommendation rate of the i-th refrigeration equipment and combine them into a feature vector Randomly select 3 eigenvectors from the set of all eigenvectors as the initial cluster centers of the clustering analysis algorithm, calculate the Euclidean distance from each eigenvector to each initial cluster center, and assign it to the cluster corresponding to the initial cluster center with the minimum distance; S42. Calculate the mean vector of all eigenvectors within a single cluster, update the initial cluster center, and reassign each eigenvector until the preset number of iterations is reached and then stop the operation to complete the clustering analysis. Each cluster represents an operation mode. Define 3 operation modes according to the clustering analysis results. The operation modes include an efficient operation mode, a balanced operation mode, and an inefficient operation mode; the clustering analysis algorithm is the K-Means clustering algorithm, and K = 3.

[0011] Preferably, the specific content of defining 3 operation modes according to the clustering analysis results in S42 is as follows: If the average energy efficiency score of different refrigeration devices within a single cluster is greater than the preset energy efficiency score threshold, and the average load recommendation rate of different refrigeration devices within a single cluster is greater than the preset load recommendation rate threshold, define the refrigeration devices within the single cluster as the efficient operation mode; if the average energy efficiency score of different refrigeration devices within a single cluster is not greater than the preset energy efficiency score threshold, and the average load recommendation rate of different refrigeration devices within a single cluster is not greater than the preset load recommendation rate threshold, define the refrigeration devices within the single cluster as the inefficient operation mode; if a single cluster does not meet the definitions of the efficient operation mode and the inefficient operation mode, define the refrigeration devices within the single cluster as the balanced operation mode.

[0012] Preferably, S5 includes the following specific steps: S51. Extract the refrigeration devices in the inefficient operation mode and adjust the corresponding water pump frequency. The adjustment formula for the water pump frequency is: ; where is the water pump frequency adjustment value of the j-th refrigeration device in the inefficient operation mode, is the water pump reference frequency, is the water pump frequency adjustment coefficient, , is the difference between the cooling water outlet temperature and the cooling water inlet temperature of the j-th refrigeration device in the inefficient operation mode, is the set temperature difference reference value; S52. Adjust the rotational speed of the cooling tower fan corresponding to the refrigeration device in the inefficient operation mode. The adjustment formula for the rotational speed of the cooling tower fan is: ; where is the rotational speed adjustment value of the cooling tower fan of the j-th refrigeration device in the inefficient operation mode, is the rotational speed reference value of the cooling tower fan, is the rotational speed adjustment coefficient of the cooling tower fan, .

[0013] In a second aspect, a computer-readable storage medium is proposed, on which a computer program is stored. When the computer program is executed by a processor, the energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning as described above is implemented.

[0014] In a third aspect, an electronic device is proposed, including a memory for storing instructions; and a processor for executing the instructions, such that the device implements the energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning as described above.

[0015] The technical effects of the present invention are as follows: An energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning is constructed. The actual energy efficiency of the equipment is accurately characterized by the energy efficiency scoring formula; the load rate and load recommendation rate are introduced to quantify the load state and optimization potential of the equipment, providing data support for refined control; the K-Means clustering algorithm is used to divide the equipment into three operating modes based on the energy efficiency score and load recommendation rate, and differential strategies are formulated for different modes, making full use of high-cost-performance equipment, accurately adjusting the pump frequency and the rotational speed of the cooling tower fan, and specifically improving the energy efficiency, solving the defect of "undifferentiated control" of the traditional method, and realizing the optimal allocation of resources; the whole process control loop is closed, without manual intervention, and can respond to environmental and load changes in real time, adapting to complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] By reading the detailed description of the non-limiting embodiments with reference to the following drawings, other features, objectives, and advantages of the present invention will become more apparent: Figure 1 It is a schematic flowchart of the energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning according to Embodiment 1 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Embodiment 1 This embodiment proposes an energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning. The actual energy efficiency of the equipment is accurately characterized by the energy efficiency scoring formula; the load rate and load recommendation rate are introduced to quantify the load state and optimization potential of the equipment, providing data support for refined control; the K-Means clustering algorithm is used to divide the equipment into three operating modes based on the energy efficiency score and load recommendation rate, and differential strategies are formulated for different modes, making full use of high-cost-performance equipment, accurately adjusting the pump frequency and the rotational speed of the cooling tower fan, and specifically improving the energy efficiency, solving the defect of "undifferentiated control" of the traditional method, and realizing the optimal allocation of resources; the whole process control loop is closed, without manual intervention, and can respond to environmental and load changes in real time, adapting to complex working conditions. Specifically, as Figure 1 shown, the adaptive propulsion control method for the water-air amphibious aircraft based on the intelligent algorithm proposed in this embodiment includes the following specific steps: S1. Collect the environmental data and the operation data of the refrigeration equipment in the water-cooled refrigeration station system; S2. Calculate the energy efficiency score of the refrigeration equipment and calculate the load rate of the refrigeration equipment; S3. Based on the energy efficiency score and the load rate of the refrigeration equipment, calculate the load recommendation rate of the refrigeration equipment; S4. Import the energy efficiency score and the load recommendation rate of the refrigeration equipment into the clustering analysis algorithm, conduct clustering analysis on the refrigeration equipment in the water-cooled refrigeration station system, and divide the refrigeration equipment into high-efficiency operation mode, balanced operation mode and low-efficiency operation mode; S5. Extract the refrigeration equipment in the low-efficiency operation mode and adjust the corresponding water pump frequency and the rotating speed of the cooling tower fan.

[0018] In this embodiment, the environmental data in S1 includes the inlet temperature of the cooling water and the outlet temperature of the cooling water; the operation data of the refrigeration equipment includes the refrigeration capacity of the refrigeration equipment, the power consumption of the refrigeration equipment, the water pump frequency and the rotating speed of the cooling tower fan.

[0019] In this embodiment, the calculation formula for the energy efficiency score in S2 is: ; Where is the energy efficiency score of the i-th refrigeration equipment, is the refrigeration capacity of the i-th refrigeration equipment, is the power consumption of the i-th refrigeration equipment, is the inlet temperature of the cooling water of the i-th refrigeration equipment, is the outlet temperature of the cooling water of the i-th refrigeration equipment, is the designed optimal temperature difference of the refrigeration equipment, , are weight factors respectively, and .

[0020] In this embodiment, the calculation formula for the load rate in S2 is: ; Where is the load rate of the i-th refrigeration equipment, is the maximum refrigeration capacity of the i-th refrigeration equipment.

[0021] In this embodiment, the calculation formula for the load recommendation rate in S3 is: ; Where is the load recommendation rate of the i-th refrigeration equipment, n is the total number of refrigeration equipment, , are weight factors respectively, and 。

[0022] In this embodiment, S4 includes the following specific steps: S41. Extract the energy efficiency score and the load recommendation rate of the i-th refrigeration device and combine them into a feature vector ; randomly select 3 feature vectors from the set of all feature vectors as the initial cluster centers of the clustering analysis algorithm, calculate the Euclidean distance from each feature vector to each initial cluster center, and assign it to the cluster corresponding to the initial cluster center with the minimum distance; S42. Calculate the mean vector of all feature vectors within a single cluster, update the initial cluster centers, and reassign each feature vector until the operation stops after reaching the preset number of iterations, completing the clustering analysis. Each cluster represents an operating mode. Define 3 operating modes according to the clustering analysis results. The operating modes include an efficient operating mode, a balanced operating mode, and an inefficient operating mode; the clustering analysis algorithm is the K-Means clustering algorithm, and K = 3.

[0023] In this embodiment, the specific content of defining 3 operating modes according to the clustering analysis results in S42 is as follows: If the average energy efficiency score of different refrigeration devices within a single cluster is greater than the preset energy efficiency score threshold, and the average load recommendation rate of different refrigeration devices within a single cluster is greater than the preset load recommendation rate threshold, define the refrigeration devices within the single cluster as the efficient operating mode; if the average energy efficiency score of different refrigeration devices within a single cluster is not greater than the preset energy efficiency score threshold, and the average load recommendation rate of different refrigeration devices within a single cluster is not greater than the preset load recommendation rate threshold, define the refrigeration devices within the single cluster as the inefficient operating mode; if a single cluster does not meet the definitions of the efficient operating mode and the inefficient operating mode, define the refrigeration devices within the single cluster as the balanced operating mode.

[0024] In this embodiment, S5 includes the following specific steps: S51. Extract the refrigeration devices in the inefficient operating mode and adjust the corresponding water pump frequency. The adjustment formula for the water pump frequency is: ; where is the water pump frequency adjustment value of the j-th refrigeration device in the inefficient operating mode, is the water pump reference frequency, is the water pump frequency adjustment coefficient, , is the difference between the cooling water outlet temperature and the cooling water inlet temperature of the j-th refrigeration device in the inefficient operating mode, is the set temperature difference reference value; S52. Adjust the rotational speed of the cooling tower fan corresponding to the refrigeration equipment operating in an inefficient mode. The adjustment formula for the rotational speed of the cooling tower fan is as follows: ; where is the adjustment value of the rotational speed of the cooling tower fan for the j-th refrigeration equipment operating in an inefficient mode, is the reference value of the rotational speed of the cooling tower fan, is the adjustment coefficient of the rotational speed of the cooling tower fan, .

[0025] The setting of the threshold and weight can be based on the default settings of the present invention or can be set by the operator himself.

[0026] Embodiment 2 This embodiment provides an electronic device, including: a processor and a memory. Among them, a computer program that can be called by the processor is stored in the memory; the processor executes the above-mentioned energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning by calling the computer program stored in the memory.

[0027] This electronic device may have relatively large differences due to different configurations or performances, and can include one or more processors (Central Processing Units, CPUs) and one or more memories. Among them, at least one computer program is stored in the memory, and this computer program is loaded and executed by the processor to implement the energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning provided by the above method embodiment. This electronic device can also include other components for realizing the functions of the device. For example, this electronic device can also have components such as wired or wireless network interfaces and input / output interfaces for data input and output. This embodiment will not be elaborated here.

[0028] Those skilled in the art of the relevant technology know that the present invention can be implemented as a system, a method, or a computer program product. Therefore, the present disclosure can be specifically implemented in the following forms: it can be completely hardware, can also be completely software (including firmware, resident software, microcode, etc.), and can also be in the form of a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" in this article. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable media contains computer-readable program code.

[0029] Any combination of one or more computer-readable media may be employed. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present document, a computer-readable storage medium may be any tangible medium that contains or stores a program which can be used by or in connection with an instruction execution system, apparatus, or device.

[0030] The present invention is described with reference to the flowcharts and block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow or block in the flowcharts and block diagrams, and combinations of flows and blocks in the flowcharts or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to produce a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices produce means for implementing the functions specified in Figure 1 one flow or multiple flows and blocks one block or multiple blocks.

[0031] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and blocks Figure 1 Figure 1 one block or multiple blocks.

[0032] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope protected by the present invention and the claims. All of these are within the protection scope of the present invention.

Claims

1. An energy-saving optimization control method for a water-cooled refrigeration station system based on deep reinforcement learning, characterized in that: It includes the following specific steps: S1. Collect the environmental data and the operation data of the refrigeration equipment of the water-cooled refrigeration station system; S2. Calculate the energy efficiency score of the refrigeration equipment and calculate the load rate of the refrigeration equipment; S3. Based on the energy efficiency score and the load rate of the refrigeration equipment, calculate the load recommendation rate of the refrigeration equipment; S4. Import the energy efficiency score and the load recommendation rate of the refrigeration equipment into the clustering analysis algorithm, perform clustering analysis on the refrigeration equipment of the water-cooled refrigeration station system, and divide the refrigeration equipment into a high-efficiency operation mode, a balanced operation mode, and a low-efficiency operation mode; S5. Extract the refrigeration equipment in the low-efficiency operation mode and adjust the corresponding water pump frequency and the rotational speed of the cooling tower fan.

2. The energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning according to claim 1, wherein: The environmental data in S1 includes the inlet temperature of the cooling water and the outlet temperature of the cooling water; the operation data of the refrigeration equipment includes the refrigerating capacity of the refrigeration equipment, the power consumption of the refrigeration equipment, the water pump frequency, and the rotational speed of the cooling tower fan.

3. The energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning according to claim 2, wherein: The calculation formula for the energy efficiency score in S2 is: ； Among them, is the energy efficiency score of the i-th refrigeration device, is the refrigerating capacity of the i-th refrigeration device, is the power consumption of the i-th refrigeration device, is the inlet temperature of the cooling water of the i-th refrigeration device, is the outlet temperature of the cooling water of the i-th refrigeration device, is the design optimal temperature difference of the refrigeration device, 、 are the weight factors respectively, and .

4. The energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning according to claim 3, wherein: The calculation formula for the load rate in S2 is: ； wherein, is the load rate of the i-th refrigeration device, is the maximum refrigerating capacity of the i-th refrigeration device.

5. The energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning according to claim 4, characterized in that: The calculation formula for the load recommendation rate in S3 is: ； Among them, is the load recommendation rate of the i-th refrigeration device, n is the total number of refrigeration devices, , are the weight factors respectively, and .

6. The energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning according to claim 5, characterized in that: S4 includes the following specific steps: S41. Extract the energy efficiency score of the i-th refrigeration device and the load recommendation rate and combine them into a feature vector ; randomly select 3 feature vectors from the set of all feature vectors as the initial cluster centers of the clustering analysis algorithm, calculate the Euclidean distance from each feature vector to each initial cluster center, and assign it to the cluster corresponding to the initial cluster center with the smallest distance; S42. Calculate the mean vector of all feature vectors within a single cluster, update the initial cluster center, and reassign each feature vector until the preset number of iterations is reached and then stop the operation to complete the clustering analysis. Each cluster represents an operation mode. Define 3 operation modes according to the clustering analysis result. The operation modes include a high-efficiency operation mode, a balanced operation mode, and a low-efficiency operation mode; the clustering analysis algorithm is the K-Means clustering algorithm, and K = 3.

7. The energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning according to claim 6, characterized in that: The specific content of defining 3 operation modes according to the clustering analysis result in S42 is: If the average energy efficiency score of different refrigeration equipment within a single cluster is greater than the preset energy efficiency score threshold, and the average load recommendation rate of different refrigeration equipment within a single cluster is greater than the preset load recommendation rate threshold, define the refrigeration equipment within the single cluster as the high-efficiency operation mode; If the average energy efficiency score of different refrigeration equipment within a single cluster is not greater than the preset energy efficiency score threshold, and the average load recommendation rate of different refrigeration equipment within a single cluster is not greater than the preset load recommendation rate threshold, define the refrigeration equipment within the single cluster as the low-efficiency operation mode; If a single cluster does not meet the definitions of the high-efficiency operation mode and the low-efficiency operation mode, define the refrigeration equipment within the single cluster as the balanced operation mode.

8. The energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning according to claim 7, characterized in that: S5 includes the following specific steps: S51. Extract the refrigeration equipment operating in the inefficient mode and adjust the corresponding pump frequency. The adjustment formula for the pump frequency is as follows: ; where is the pump frequency adjustment value of the j-th refrigeration equipment in the inefficient mode, is the pump reference frequency, is the pump frequency adjustment coefficient, , is the difference between the cooling water outlet temperature and the cooling water inlet temperature of the j-th refrigeration equipment in the inefficient mode, is the set temperature difference reference value; S52. Adjust the rotational speed of the cooling tower fan corresponding to the refrigeration equipment operating in an inefficient mode. The adjustment formula for the rotational speed of the cooling tower fan is as follows: ; where is the adjustment value of the rotational speed of the cooling tower fan for the j-th refrigeration equipment operating in an inefficient mode, is the reference value of the rotational speed of the cooling tower fan, is the adjustment coefficient of the rotational speed of the cooling tower fan, .

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning as described in any one of claims 1-8.

10. An electronic device, characterized in that, It includes a memory for storing instructions; a processor for executing the instructions, so that the device executes the energy-saving optimization control method for the water-cooled refrigeration station system based on deep reinforcement learning as described in any one of claims 1 to 8.

Images