Control method and device of machine room refrigeration system and related equipment

Abstract

The invention provides a control method and device for a machine room refrigeration system and related equipment. The method comprises the steps that N pieces of first user data matched with a target machine room are obtained; calculating the N pieces of first user data to obtain a load prediction value of the target machine room; the load prediction value, the environment variable corresponding to the target machine room, the operation time parameter corresponding to the target machine room and the switch parameter corresponding to the target machine room are input into a target deep learning model for prediction, and a refrigeration control strategy is obtained; and controlling a refrigeration system of the target machine room based on the refrigeration control strategy. According to the method, after the N first user data matched with the target machine room are obtained, the load prediction value of the target machine room is calculated according to the N first user data, the load prediction value, the environment variable, the operation time parameter and the switch parameter are input into the target deep learning model for prediction, and the refrigeration control strategy is obtained; therefore, the control effect of the refrigeration system control strategy on the refrigeration system is improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a control method, device and related equipment for a computer room refrigeration system. Background Art

[0002] Current energy-saving solutions for communications equipment rooms primarily utilize AI algorithms to build a power load model based on historical data such as the power load of each device within the room, the number of cooling equipment operations, historical meteorological data, and historical temperature data. This model then establishes the optimal air conditioning and cooling strategy. Existing solutions typically build on this foundation by collecting additional data, including temperature data around the cabinets and cooling equipment operation data, for more focused optimization.

[0003] However, in the prior art, the refrigeration system control strategy is passively issued according to the operating status and heat generation of the equipment in the computer room, which leads to the problem that the refrigeration system control strategy in the prior art has a poor control effect on the refrigeration system. Summary of the Invention

[0004] The present invention provides a control method, device and related equipment for a computer room refrigeration system, which solves the problem of poor control effect of the refrigeration system in the prior art.

[0005] To solve the above problems, the present invention is achieved as follows:

[0006] In a first aspect, an embodiment of the present application provides a method for controlling a computer room cooling system, the method comprising:

[0007] Obtaining N first user data matching the target computer room, where the first user data is Internet behavior data generated by the corresponding user based on the user account during the process of using broadband, where N is a positive integer;

[0008] Calculating the N first user data to obtain a load prediction value of the target computer room, where the load prediction value is the cooling load required in the target computer room;

[0009] Input the load prediction value, the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room, and the switching parameters corresponding to the target computer room into the target deep learning model for prediction to obtain a cooling control strategy, wherein the environmental variables include the temperature and humidity parameters corresponding to the target computer room, and the cooling control strategy is used to adjust the operating parameters corresponding to the cooling system of the target computer room, the operating time parameter is the service operation time of the target computer room, and the switching parameter is the number of times the cooling system of the target computer room is turned on and off during the service operation time;

[0010] The cooling system of the target computer room is controlled based on the cooling control strategy.

[0011] Optionally, obtaining N first user data matching the target computer room includes:

[0012] Obtaining M pieces of second user data, where the M pieces of second user data correspond one-to-one to M user accounts, the second user data being online behavior data generated by the corresponding user based on the user account during broadband use, where M is greater than or equal to N and is a positive integer;

[0013] Parsing the M second user data to obtain M parsing results, where the M parsing results correspond one-to-one to the M second user data, and the parsing results are used to indicate whether the computer room information corresponding to the second user data matches the computer room information corresponding to the target computer room;

[0014] The N first user data are determined from the M second user data according to the M parsing results, and the computer room information indicated by the parsing results corresponding to the first user data matches the computer room information corresponding to the target computer room.

[0015] Optionally, parsing the M second user data to obtain M parsing results includes:

[0016] performing data cleaning and data parsing on the M second user data to obtain the M parsing results, wherein each of the M parsing results includes a first field and a second field, the first field being used to indicate user access line identification information of the corresponding second user data, and the second field being used to indicate user access location information of the corresponding second user data, the user access line identification information being used to indicate a device identification of a user accessing a network, and the user access location information being used to indicate a geographical location of a user accessing a network, and when the device identification and the geographical location are the same as the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result matches the computer room information corresponding to the target computer room; and when the device identification and the geographical location are different from the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result does not match the computer room information corresponding to the target computer room;

[0017] Determining the N first user data from the M second user data according to the M parsing results includes:

[0018] For each piece of second user data, generating computer room information corresponding to each piece of second user data according to the first field and the second field corresponding to the second user data, the computer room information including the device identifier and the geographic location;

[0019] The N first user data are determined from the M second user data according to the computer room information corresponding to each second user data.

[0020] Optionally, calculating the N first user data to obtain a load prediction value of the target computer room includes:

[0021] Traversing the N first user data, obtaining Internet access information corresponding to each first user data, the Internet access information including the user's Internet access time period and the user's Internet access traffic;

[0022] Obtaining a target time period and a bandwidth capacity of the target computer room, wherein the target time period is a service start time period of the target computer room, and the bandwidth capacity of the computer room is a maximum bandwidth that the target computer room can provide;

[0023] Calculating the N pieces of Internet access information, the target time period, and the bandwidth capacity of the computer room to obtain a bandwidth utilization rate of the target computer room;

[0024] The bandwidth utilization is determined as the load prediction value of the target computer room.

[0025] Optionally, before inputting the load prediction value, the environmental variables corresponding to the target computer room, and other variables corresponding to the target computer room into a target deep learning model for prediction to obtain a cooling control strategy, the method further includes:

[0026] Obtaining a training data set, the training data set including historical load prediction values, historical environmental variables, historical operating time parameters, historical switch parameters, and training labels corresponding to the target computer room, the training labels being historical cooling control strategies corresponding to the target computer room, the historical load prediction values, the historical environmental variables, the historical operating time parameters, and the historical switch parameters being data within the same historical time period;

[0027] The deep learning model is trained based on the training data set to obtain the target deep learning model.

[0028] Optionally, the refrigeration control strategy includes at least one of the following: a refrigeration equipment supply / return air temperature control strategy, a refrigeration equipment fan speed adjustment strategy, a refrigeration equipment compressor speed adjustment strategy, a refrigeration equipment switch quantity control strategy, a humidification / dehumidification control strategy, a fluorine gas switching strategy, a fresh air switching strategy, a dual cold source switching strategy and a compressor load setting strategy.

[0029] Optionally, the environmental variables include at least one of the following: Internet power, total power of the computer room, ratio of energy consumed to energy consumed by Internet load, refrigeration equipment supply air temperature, refrigeration equipment return air temperature, cold channel temperature and humidity, hot channel temperature and humidity, outdoor temperature and humidity, refrigeration equipment start and stop status, fan start and stop status, compressor start and stop status, fan speed, compressor speed and outdoor temperature.

[0030] In a second aspect, an embodiment of the present application provides a control device for a computer room cooling system, the device comprising:

[0031] an acquisition module, configured to acquire N first user data matching the target computer room, wherein the first user data is online behavior data generated by the corresponding user based on the user account during the process of using broadband, and N is a positive integer;

[0032] a calculation module, configured to calculate the N first user data to obtain a load prediction value of the target computer room, where the load prediction value is the cooling load required in the target computer room;

[0033] a prediction module, configured to input the load prediction value, the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room, and the switching parameters corresponding to the target computer room into a target deep learning model for prediction to obtain a refrigeration control strategy, wherein the environmental variables include the temperature and humidity parameters corresponding to the target computer room; the refrigeration control strategy is configured to adjust the operating parameters corresponding to the refrigeration system of the target computer room; the operating time parameter is the service operation time of the target computer room; and the switching parameter is the number of times the refrigeration system of the target computer room is switched on and off during the service operation time;

[0034] A control module is used to control the refrigeration system of the target computer room based on the refrigeration control strategy.

[0035] In a third aspect, the present application also provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method described in the first aspect above.

[0036] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the method described in the first aspect above are implemented.

[0037] In a fifth aspect, the present application also provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps in the method described in the first aspect above.

[0038] The present application provides a control method, device and related equipment for a computer room cooling system, the method comprising: obtaining N first user data matching a target computer room; calculating the N first user data to obtain a load prediction value for the target computer room; inputting the load prediction value, the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room and the switch parameters corresponding to the target computer room into a target deep learning model for prediction to obtain a cooling control strategy; and controlling the cooling system of the target computer room based on the cooling control strategy. After obtaining N first user data matching a target computer room, the present application calculates the load prediction value for the target computer room based on the N first user data, inputs the load prediction value, the environmental variables, the operating time parameters and the switch parameters into a target deep learning model for prediction to obtain a cooling control strategy, thereby improving the control effect of the cooling system control strategy on the cooling system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for the description of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0040] Figure 1 A flow chart of a method for controlling a computer room cooling system provided in an embodiment of the present application;

[0041] Figure 2 A schematic diagram of the cooling control strategy provided in an embodiment of the present application;

[0042] Figure 3 Flowchart of the parsing algorithm of logicalportno provided in the embodiment of the present application;

[0043] Figure 4 A schematic diagram of the DCN model provided in this embodiment;

[0044] Figure 5 A schematic diagram of the structure of a control device for a computer room cooling system provided in an embodiment of the present application;

[0045] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0047] The terms "first", "second" etc. in the embodiments of the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. In addition, the terms "comprise" and "have" and any deformation thereof are intended to cover non-exclusive inclusions, such as, the process, method, system, product or equipment comprising a series of steps or units need not be limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or that are intrinsic to these processes, methods, products or equipment. In addition, "and / or" is used in the present application to represent at least one of connected objects, such as A and / or B and / or C, and represents comprising independent A, independent B, independent C, and A and B all exist, B and C all exist, A and C all exist, and 7 situations that A, B and C all exist.

[0048] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of a method for controlling a computer room cooling system according to an embodiment of the present application. Figure 1 As shown, the control method of the computer room cooling system may include the following steps:

[0049] Step 101: Acquire N first user data matching the target computer room, where the first user data is Internet behavior data generated by the corresponding user based on the user account during the process of using broadband, and N is a positive integer.

[0050] In this embodiment, the target computer room is one containing an Optical Line Terminal (OTL) device. An OLT is a device used for fiber-optic access networks, primarily responsible for converting fiber signals into electrical signals and performing data processing and management. It should be noted that this application predicts the computer room load based on the Internet behavior of home broadband users and adjusts the air conditioning accordingly. Therefore, choosing a computer room containing an OLT and primarily carrying broadband services will yield better air conditioning energy savings.

[0051] The N first user data are Internet behavior data generated by users when using broadband through their user accounts. For example, the N first user data are extracted through Radius home broadband AAA call bill data, wherein AAA is a security management framework for network access control, embedded in the process of user terminals initiating PPPoE Internet access requests, which determines which users can access the network, and which resources or services the users can access. AAA includes three elements: Authentication, Authorization, and Accounting, which can be implemented through a variety of protocols, including RADIUS, TACACS, LDAP, and Diameter, among which RADIUS (Remote Authentication Dial-In User Service) is the current standard protocol. The RADIUS protocol can be divided into an authentication protocol and an accounting protocol, which are defined by IETF RFC 2865 and RFC 2866 respectively.

[0052] Radius home broadband AAA call records contain both customer and call record data. Each time a home broadband user accesses the internet, a call record is generated. By extracting the user account, internet start and end time, uplink traffic, downlink traffic, the IP address used to connect to the OLT, and the IP address used to connect to the BRAS, a model of home broadband user online behavior can be constructed. Based on this model, using the user account field as a unique identifier, the online behavior data of a single home broadband user over a period of time, i.e., the primary user data, can be obtained.

[0053] Step 102: Calculate the N first user data to obtain a load prediction value of the target computer room, where the load prediction value is the cooling load required in the target computer room.

[0054] In this embodiment, the N first user data items are calculated to calculate a load forecast value for the target computer room. The load forecast value is the required cooling load in the target computer room. The cooling load refers to the required cooling capacity to maintain a specific environment (e.g., an indoor space), expressed as the amount of heat that needs to be removed from the space. The cooling load is typically measured in kilowatts (kW) or tons of cooling capacity (TR) and is used to calculate and design the capacity of air conditioning and refrigeration systems.

[0055] Step 103: Input the load prediction value, the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room, and the switch parameters corresponding to the target computer room into the target deep learning model for prediction to obtain a refrigeration control strategy. The environmental variables include the temperature and humidity parameters corresponding to the target computer room. The refrigeration control strategy is used to adjust the operating parameters corresponding to the refrigeration system of the target computer room. The operating time parameter is the business operation time of the target computer room, and the switch parameter is the number of times the refrigeration system of the target computer room is switched on and off during the business operation time.

[0056] In this embodiment, after obtaining the load forecast value, it is also necessary to obtain the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room, and the switch parameters corresponding to the target computer room, thereby inputting the above parameters into the target deep learning model for prediction to obtain the cooling control strategy.

[0057] Specifically, the environmental variables include temperature and humidity parameters corresponding to the target computer room. Specifically, the environmental variables include at least one of the following: Internet power, total power of the computer room, ratio of energy consumed to energy consumed by the Internet load, refrigeration equipment supply air temperature, refrigeration equipment return air temperature, cold channel temperature and humidity, hot channel temperature and humidity, outdoor temperature and humidity, refrigeration equipment start and stop status, fan start and stop status, compressor start and stop status, fan speed, compressor speed and outdoor temperature.

[0058] The operating time parameter is the target computer room's service operating time, that is, the target computer room's normal operating time. The switching parameter is the number of times the target computer room's cooling system is switched on and off during the service operating time, such as the number of times each cooling device is switched on and off.

[0059] The cooling control strategy is used to adjust the operating parameters corresponding to the cooling system of the target computer room, so that the target computer room meets the current cooling load demand while saving energy.

[0060] Step 104: Control the cooling system of the target computer room based on the cooling control strategy.

[0061] In this embodiment, the cooling system of the target computer room is controlled according to the cooling control strategy. Specifically, Figure 2 As shown, Figure 2Figure 1 is a schematic diagram of the cooling control strategy used in this embodiment. This strategy includes at least one of the following: cooling equipment supply / return air temperature control strategy, cooling equipment fan speed adjustment strategy, cooling equipment compressor speed adjustment strategy, cooling equipment on / off quantity control strategy, humidification / dehumidification control strategy, fluorine gas switching strategy, fresh air switching strategy, dual cooling source switching strategy, and compressor load setting strategy. These control variables directly affect the computer room environment, thereby achieving the control objectives of maintaining stable room temperature and optimizing energy consumption.

[0062] The cooling system of the target computer room is controlled through a cooling control strategy, which combines the existing data computer room environmental variables, time and switch variables, and load forecast values. Based on artificial intelligence deep reinforcement learning, the control strategy is automatically learned and optimized to reduce the power consumption of the computer room cooling system while ensuring the stability of the computer room temperature.

[0063] This application obtains N first user data that matches the target computer room, and then calculates the load prediction value of the target computer room based on the N first user data, inputs the load prediction value, environmental variables, operating time parameters and switch parameters into the target deep learning model for prediction, and obtains the refrigeration control strategy, thereby improving the control effect of the refrigeration system control strategy on the refrigeration system.

[0064] In some feasible implementations, optionally, obtaining N first user data matching the target computer room includes:

[0065] Obtaining M pieces of second user data, where the M pieces of second user data correspond one-to-one to M user accounts, the second user data being online behavior data generated by the corresponding user based on the user account during broadband use, where M is greater than or equal to N and is a positive integer;

[0066] Parsing the M second user data to obtain M parsing results, where the M parsing results correspond one-to-one to the M second user data, and the parsing results are used to indicate whether the computer room information corresponding to the second user data matches the computer room information corresponding to the target computer room;

[0067] The N first user data are determined from the M second user data according to the M parsing results, and the computer room information indicated by the parsing results corresponding to the first user data matches the computer room information corresponding to the target computer room.

[0068] In this embodiment, M second user data are obtained within a certain cell or a certain communication range. Specifically, the M second user data are in a one-to-one correspondence with the M user accounts, wherein the second user data is the Internet behavior data generated by the corresponding user based on the user account during the use of broadband.

[0069] By parsing the M pieces of second user data, M parsing results are obtained. The parsing results are used to determine whether the computer room information of the second user data matches the computer room information of the target computer room. If so, it indicates that the second user data was processed by the target computer room. Thus, N pieces of first user data are determined from the M pieces of second user data. The computer room information indicated by the parsing results corresponding to the first user data matches the computer room information corresponding to the target computer room, indicating that the first user data was transmitted by a single OTL device, and the OTL is located in the target computer room. Thus, the N pieces of first user data associated with the target computer room are determined.

[0070] Optionally, parsing the M second user data to obtain M parsing results includes:

[0071] performing data cleaning and data parsing on the M second user data to obtain the M parsing results, wherein each of the M parsing results includes a first field and a second field, the first field being used to indicate user access line identification information of the corresponding second user data, and the second field being used to indicate user access location information of the corresponding second user data, the user access line identification information being used to indicate a device identification of a user accessing a network, and the user access location information being used to indicate a geographical location of a user accessing a network, and when the device identification and the geographical location are the same as the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result matches the computer room information corresponding to the target computer room; and when the device identification and the geographical location are different from the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result does not match the computer room information corresponding to the target computer room;

[0072] Determining the N first user data from the M second user data according to the M parsing results includes:

[0073] For each piece of second user data, generating computer room information corresponding to each piece of second user data according to the first field and the second field corresponding to the second user data, the computer room information including the device identifier and the geographic location;

[0074] The N first user data are determined from the M second user data according to the computer room information corresponding to each second user data.

[0075] In this embodiment, when M pieces of second user data are acquired, data cleaning and parsing are performed on the M pieces of second user data to obtain M parsing results. Specifically, the parsing results include a first field and a second field. The first field is used to indicate user access line identification information of the corresponding second user data, and the second field is used to indicate user access location information of the corresponding second user data. The first field is the logicalportno field, and the second field is the nas-identifier field. The user access line identification information indicates the device identifier of the user accessing the network, and the user access location information indicates the geographic location of the user accessing the network. If the device identifier and the geographic location are identical to the computer room information corresponding to the target computer room, then whether the parsing results match the computer room information of the target computer room is determined based on the first field and the second field.

[0076] It's important to note that Home Broadband Radius generates detailed AAA user-level daily call bill data. Since each record's user data must be mapped to a single OLT to identify the target data center, existing AAA call bill data cannot directly identify the OLT corresponding to each user. This is primarily due to two reasons: first, the call bill lacks the OLT IP field; second, different OLTs in different regions may have the same IP address. Therefore, to uniquely identify a specific OLT, in addition to the device IP address, the city (city) to which the device belongs is also required. Using the IP and city fields as a joint primary key, a single OLT can be uniquely identified.

[0077] The first field of parsing, logicalportno, is explained, where the example format of logicalportno is: trunk3 / 0 / 7:2125.18 172.31.251.84 / 0 / 0 / 7 / 0 / 16 / CMDCB238F0FA GP. First, determine whether the length of logicalportno characters is greater than 30. If logicalportno≤30, the determination process ends, and this call record is considered an invalid record and will not be added to the OLT peak user record. Get the position of each element in turn. If the acquisition fails, it is determined to be an abnormal character, and the process ends. If each element can be successfully obtained, then this call record can intercept OLT_ip. Specifically, if Figure 3 As shown, Figure 3 This is the flowchart of the parsing algorithm of logicalportno. By obtaining the position of each element in turn, it is determined whether the intercepted OLT_ip is completely obtained.

[0078] The second field, nas-identifier, is parsed as follows: HBJZO-MC-CMNET-BAS03-ESN-7750. The string before the first "-" is taken as the city, nas_city. For example, the parsing result is: HBJZO. Because the same OLT_ip may exist in different cities, nas_city and OLT_ip are concatenated as a composite primary key to determine the uniqueness of the OLT in each call record. This is used to associate Radius home broadband AAA call records with the target computer room. For example, nas_cityolt_ip:

[0079] HAJIZ172.31.251.84,

[0080] HAJIZ172.31.247.60,

[0081] HAJIZ172.31.247.60.

[0082] Whether the second user data matches the target computer room can be accurately determined by the first field and the second field, thereby improving the matching accuracy.

[0083] Optionally, calculating the N first user data to obtain a load prediction value of the target computer room includes:

[0084] Traversing the N first user data, obtaining Internet access information corresponding to each first user data, the Internet access information including the user's Internet access time period and the user's Internet access traffic;

[0085] Obtaining a target time period and a bandwidth capacity of the target computer room, wherein the target time period is a service start time period of the target computer room, and the bandwidth capacity of the computer room is a maximum bandwidth that the target computer room can provide;

[0086] Calculating the N pieces of Internet access information, the target time period, and the bandwidth capacity of the computer room to obtain a bandwidth utilization rate of the target computer room;

[0087] The bandwidth utilization is determined as the load prediction value of the target computer room.

[0088] In this embodiment, when calculating the load forecast value for the target computer room, the N first user data are traversed to obtain the corresponding online information for each first user data. The online information includes the user's online time period and the user's online traffic volume. For example, AAA call bill data within a three-day period is traversed. This is because the call bill data for a given day may contain only partial records. For example, if the device's active end time on a given day falls at a certain moment the next day, such a call bill record will be recorded in the next day's call bill data, and such a record will not appear on the current day. Similarly, the call bill records from the previous day are traversed for this reason. Therefore, to address the issue of inaccurate call bill segmentation, data from the past three days is used as the data source. After cleaning and parsing, the AAA call bill data within a three-day period is traversed to obtain relevant information such as the user's active start time, active end time, user upstream traffic volume, and user downstream traffic volume on each OLT device. Key information is shown in Table 1 below.

[0089] Table 1

[0090]

[0091]

[0092] Where OLT represents the OLT device, begintime represents the timestamp of the user's activity start on the current device, endtime represents the timestamp of the user's activity end on the current device, dataTime represents the time of the current call record, sendkbytes represents the cumulative upstream traffic during the user's activity period, in KB, and receivekbytes represents the cumulative downstream traffic during the user's activity period, in KB. Specifically, the user's online time period is determined by endtime and dataTime, and the user's online traffic is determined by sendkbytes and receivekbytes.

[0093] Therefore, after determining the target computer room, the identifiers of all OLT devices in the target computer room can be determined through the target computer room, that is, all home broadband Internet users carried by the OLT in the target computer room can be identified according to the Radius home broadband AAA call bill data.

[0094] In this embodiment, the OLT bandwidth utilization is determined as the load prediction value of the target computer room, where: The cooling load of the target equipment room is measured by the OLT bandwidth utilization.

[0095] In this embodiment, based on Radius home broadband AAA call bill data, the load of the computer room is predicted by predicting the user's Internet behavior, and then the strategy of the cooling system is formulated by combining traditional data computer room environmental variable parameters, cooling equipment parameters, equipment load parameters and other parameters. This is an active control strategy for the computer room cooling system, which improves the control effect.

[0096] Optionally, before inputting the load prediction value, the environmental variables corresponding to the target computer room, and other variables corresponding to the target computer room into a target deep learning model for prediction to obtain a cooling control strategy, the method further includes:

[0097] Obtaining a training data set, the training data set including historical load prediction values, historical environmental variables, historical operating time parameters, historical switch parameters, and training labels corresponding to the target computer room, the training labels being historical cooling control strategies corresponding to the target computer room, the historical load prediction values, the historical environmental variables, the historical operating time parameters, and the historical switch parameters being data within the same historical time period;

[0098] The deep learning model is trained based on the training data set to obtain the target deep learning model.

[0099] In this embodiment, the deep learning model is a Data Center Network (DCN), a structure used to design and build data center networks. The primary purpose of the DCN model is to provide efficient, high-speed, and reliable data transmission to meet the network performance and scalability requirements of modern data centers.

[0100] Based on a single Radius home broadband AAA call bill, we can accurately locate the OLT identifier, start time ts, end time te, and total user traffic (uplink traffic tup + downlink traffic tup) used by a AAA call bill to carry home broadband users. Based on all AAA call bill data, we can obtain the load on the computer room caused by all user behaviors within a certain time window, that is, the OLT bandwidth utilization of the target computer room. By correlating AAA user data with call bill data based on user ID, we can obtain the final DCN model input. Assuming user u i The call list is shown in Table 2 below:

[0101] Table 2

[0102]

[0103]

[0104] The user's uplink and downlink traffic per minute during this period are: and Assume that there are n users {u1, u2, ..., u n}, then the upstream and downstream traffic of the OLT device per minute are: and

[0105] The structural diagram of the DCN model in this embodiment is as follows Figure 4 As shown in the figure, it mainly consists of two parts: a cross network and a deep learning network, which increases the interaction between features. Among them, each cross layer of the cross network is used to capture the cross relationship between features. Each cross layer crosses the output of the previous layer with the input, and then uses the result as the input of the next layer. The specific superposition method is as follows:

[0106]

[0107] Among them, x l+1 is the output of the l+1th layer; x l is the output of layer l; w l is the weight matrix connecting the lth layer and the l+1th layer; b l is the bias; x0 is the initial value.

[0108] A deep network is a fully connected feedforward neural network, where each layer consists of multiple neurons. Each layer is processed as follows: a l+1 =∫(w l a l +b l ). Among them, a l+1 is the output of the l+1th layer of the neural network; a l is the output of the lth layer of the neural network; f(·) is the activation function; w l is the weight matrix connecting the lth layer and the l+1th layer; b l is the bias of the lth layer, which is used to adjust the activation value of each neuron to better fit the data.

[0109] In this embodiment, the training data set includes the historical load prediction value, historical environmental variables, historical operating time parameters, historical switch parameters and training labels corresponding to the target computer room. The training labels are the historical cooling control strategies corresponding to the target computer room. The historical load prediction value, historical environmental variables, historical operating time parameters and historical switch parameters are data within the same historical time period. The same historical time period can be a time period of three consecutive days, etc.

[0110] It should be noted that in this embodiment, the "actor-critic" deep reinforcement learning algorithm is used in the training process of the DCN model. Among them, the "actor-critic" model includes: Actor: The actor part is responsible for generating the control strategy, that is, deciding how to adjust the various environmental variables according to the current environmental state. For example, the air supply temperature of the air conditioner or the speed of the fan is determined according to the current temperature and power state. Critic: The critic part is responsible for evaluating the effectiveness of the current strategy. It measures whether the current strategy is effective based on the reward function, that is, by comparing the actual computer room environment state with the expected goal, giving feedback (reward or punishment). It should be noted that there are only two types of feedback, positive (reward) or negative (punishment).

[0111] Specifically, the process involves the following interactive processes: Strategy Generation: The actor generates an overall cooling system execution strategy (such as air conditioning temperature, fan speed, and chiller parameters) based on the current environmental variables. Strategy Evaluation: The critic evaluates the effectiveness of this strategy. If the strategy stabilizes the room temperature and optimizes energy consumption, the system receives positive rewards; otherwise, it receives negative feedback. Strategy Adjustment: Based on the critic's feedback, the actor adjusts the strategy to optimize future decisions, thereby improving environmental control through continuous feedback.

[0112] This application obtains N first user data that matches the target computer room, and then calculates the load prediction value of the target computer room based on the N first user data, inputs the load prediction value, environmental variables, operating time parameters and switch parameters into the target deep learning model for prediction, and obtains the refrigeration control strategy, thereby improving the control effect of the refrigeration system control strategy on the refrigeration system.

[0113] See also Figure 5 , Figure 5 This is a structural diagram of the control device of the computer room cooling system provided in the embodiment of the present application. Figure 5 As shown, the control device 500 of the computer room cooling system includes:

[0114] An acquisition module 510 is configured to acquire N first user data matching the target computer room, wherein the first user data is Internet surfing behavior data generated by the corresponding user based on the user account during the process of using broadband, and N is a positive integer;

[0115] A calculation module 520 is configured to calculate the N first user data to obtain a load prediction value of the target computer room, where the load prediction value is the cooling load required in the target computer room;

[0116] A prediction module 530 is configured to input the load prediction value, the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room, and the switching parameters corresponding to the target computer room into a target deep learning model for prediction to obtain a cooling control strategy, wherein the environmental variables include the temperature and humidity parameters corresponding to the target computer room; the cooling control strategy is configured to adjust the operating parameters corresponding to the cooling system of the target computer room; the operating time parameter is the service operating time of the target computer room; and the switching parameter is the number of times the cooling system of the target computer room is turned on and off during the service operating time;

[0117] The control module 540 is configured to control the cooling system of the target computer room based on the cooling control strategy.

[0118] Optionally, the calculation module 520 includes:

[0119] a first acquisition submodule, configured to acquire M pieces of second user data, wherein the M pieces of second user data correspond one-to-one to M user accounts, the second user data being online behavior data generated by the corresponding user based on the user account during broadband use, wherein M is greater than or equal to N and is a positive integer;

[0120] a parsing submodule, configured to parse the M pieces of second user data to obtain M parsing results, wherein the M parsing results correspond one-to-one to the M pieces of second user data, and the parsing results are used to indicate whether the computer room information corresponding to the second user data matches the computer room information corresponding to the target computer room;

[0121] The first determining submodule is configured to determine the N first user data from the M second user data according to the M parsing results, wherein the computer room information indicated by the parsing result corresponding to the first user data matches the computer room information corresponding to the target computer room.

[0122] Optional parsing submodules include:

[0123] a parsing unit, configured to perform data cleaning and data parsing on the M second user data to obtain the M parsing results, wherein each of the M parsing results includes a first field and a second field, the first field being used to indicate user access line identification information of the corresponding second user data, and the second field being used to indicate user access location information of the corresponding second user data, the user access line identification information being used to indicate a device identification of a user accessing a network, and the user access location information being used to indicate a geographical location of a user accessing a network, and when the device identification and the geographical location are the same as the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result matches the computer room information corresponding to the target computer room; and when the device identification and the geographical location are different from the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result does not match the computer room information corresponding to the target computer room;

[0124] The first determination submodule includes:

[0125] a generating unit, configured to generate, for each piece of second user data, computer room information corresponding to each piece of second user data according to the first field and the second field corresponding to the second user data, the computer room information including the device identifier and the geographic location;

[0126] A determining unit is configured to determine the N first user data from the M second user data according to the computer room information corresponding to each second user data.

[0127] Optionally, the acquisition module 510 includes:

[0128] A second acquisition submodule is configured to traverse the N first user data and acquire Internet access information corresponding to each first user data, wherein the Internet access information includes a user's Internet access time period and a user's Internet access traffic;

[0129] A third acquisition submodule is configured to acquire a target time period and a bandwidth capacity of the target computer room corresponding to the target computer room, wherein the target time period is a service start time period of the target computer room, and the bandwidth capacity of the computer room is a maximum bandwidth that the target computer room can provide;

[0130] a calculation submodule, configured to calculate the N pieces of Internet access information, the target time period, and the bandwidth capacity of the computer room to obtain a bandwidth utilization rate of the target computer room;

[0131] The second determining submodule is configured to determine the bandwidth utilization as the load prediction value of the target computer room.

[0132] Optionally, also include:

[0133] a data acquisition module, configured to acquire a training data set, the training data set including historical load prediction values, historical environmental variables, historical operating time parameters, historical switch parameters, and training labels corresponding to the target computer room, the training labels being historical cooling control strategies corresponding to the target computer room, the historical load prediction values, historical environmental variables, historical operating time parameters, and historical switch parameters being data within the same historical time period;

[0134] A training module is used to train the deep learning model based on the training data set to obtain the target deep learning model.

[0135] Optionally, the refrigeration control strategy includes at least one of the following: a refrigeration equipment supply / return air temperature control strategy, a refrigeration equipment fan speed adjustment strategy, a refrigeration equipment compressor speed adjustment strategy, a refrigeration equipment switch quantity control strategy, a humidification / dehumidification control strategy, a fluorine gas switching strategy, a fresh air switching strategy, a dual cold source switching strategy and a compressor load setting strategy.

[0136] Optionally, the environmental variables include at least one of the following: Internet power, total power of the computer room, ratio of energy consumed to energy consumed by Internet load, refrigeration equipment supply air temperature, refrigeration equipment return air temperature, cold channel temperature and humidity, hot channel temperature and humidity, outdoor temperature and humidity, refrigeration equipment start and stop status, fan start and stop status, compressor start and stop status, fan speed, compressor speed and outdoor temperature.

[0137] This application obtains N first user data that matches the target computer room, and then calculates the load prediction value of the target computer room based on the N first user data, inputs the load prediction value, environmental variables, operating time parameters and switch parameters into the target deep learning model for prediction, and obtains the refrigeration control strategy, thereby improving the control effect of the refrigeration system control strategy on the refrigeration system.

[0138] The present application also provides an electronic device. Figure 6 , the electronic device may include a processor 601, a memory 602, and a program 6021 stored in the memory 602 and executable on the processor 601.

[0139] When the program 6021 is executed by the processor 601, it can achieve Figure 1 Any step in the corresponding method embodiment:

[0140] Obtaining N first user data matching the target computer room, where the first user data is Internet behavior data generated by the corresponding user based on the user account during the process of using broadband, where N is a positive integer;

[0141] Calculating the N first user data to obtain a load prediction value of the target computer room, where the load prediction value is the cooling load required in the target computer room;

[0142] Input the load prediction value, the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room, and the switching parameters corresponding to the target computer room into the target deep learning model for prediction to obtain a cooling control strategy, wherein the environmental variables include the temperature and humidity parameters corresponding to the target computer room, and the cooling control strategy is used to adjust the operating parameters corresponding to the cooling system of the target computer room, the operating time parameter is the service operation time of the target computer room, and the switching parameter is the number of times the cooling system of the target computer room is turned on and off during the service operation time;

[0143] The cooling system of the target computer room is controlled based on the cooling control strategy.

[0144] Optionally, obtaining N first user data matching the target computer room includes:

[0145] Obtaining M pieces of second user data, where the M pieces of second user data correspond one-to-one to M user accounts, the second user data being online behavior data generated by the corresponding user based on the user account during broadband use, where M is greater than or equal to N and is a positive integer;

[0146] Parsing the M second user data to obtain M parsing results, where the M parsing results correspond one-to-one to the M second user data, and the parsing results are used to indicate whether the computer room information corresponding to the second user data matches the computer room information corresponding to the target computer room;

[0147] The N first user data are determined from the M second user data according to the M parsing results, and the computer room information indicated by the parsing results corresponding to the first user data matches the computer room information corresponding to the target computer room.

[0148] Optionally, parsing the M second user data to obtain M parsing results includes:

[0149] performing data cleaning and data parsing on the M second user data to obtain the M parsing results, wherein each of the M parsing results includes a first field and a second field, the first field being used to indicate user access line identification information of the corresponding second user data, and the second field being used to indicate user access location information of the corresponding second user data, the user access line identification information being used to indicate a device identification of a user accessing a network, and the user access location information being used to indicate a geographical location of a user accessing a network, and when the device identification and the geographical location are the same as the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result matches the computer room information corresponding to the target computer room; and when the device identification and the geographical location are different from the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result does not match the computer room information corresponding to the target computer room;

[0150] Determining the N first user data from the M second user data according to the M parsing results includes:

[0151] For each piece of second user data, generating computer room information corresponding to each piece of second user data according to the first field and the second field corresponding to the second user data, the computer room information including the device identifier and the geographic location;

[0152] The N first user data are determined from the M second user data according to the computer room information corresponding to each second user data.

[0153] Optionally, calculating the N first user data to obtain a load prediction value of the target computer room includes:

[0154] Traversing the N first user data, obtaining Internet access information corresponding to each first user data, the Internet access information including the user's Internet access time period and the user's Internet access traffic;

[0155] Obtaining a target time period and a bandwidth capacity of the target computer room, wherein the target time period is a service start time period of the target computer room, and the bandwidth capacity of the computer room is a maximum bandwidth that the target computer room can provide;

[0156] Calculating the N pieces of Internet access information, the target time period, and the bandwidth capacity of the computer room to obtain a bandwidth utilization rate of the target computer room;

[0157] The bandwidth utilization is determined as the load prediction value of the target computer room.

[0158] Optionally, before inputting the load prediction value, the environmental variables corresponding to the target computer room, and other variables corresponding to the target computer room into a target deep learning model for prediction to obtain a cooling control strategy, the method further includes:

[0159] Obtaining a training data set, the training data set including historical load prediction values, historical environmental variables, historical operating time parameters, historical switch parameters, and training labels corresponding to the target computer room, the training labels being historical cooling control strategies corresponding to the target computer room, the historical load prediction values, the historical environmental variables, the historical operating time parameters, and the historical switch parameters being data within the same historical time period;

[0160] The deep learning model is trained based on the training data set to obtain the target deep learning model.

[0161] Optionally, the refrigeration control strategy includes at least one of the following: a refrigeration equipment supply / return air temperature control strategy, a refrigeration equipment fan speed adjustment strategy, a refrigeration equipment compressor speed adjustment strategy, a refrigeration equipment switch quantity control strategy, a humidification / dehumidification control strategy, a fluorine gas switching strategy, a fresh air switching strategy, a dual cold source switching strategy and a compressor load setting strategy.

[0162] Optionally, the environmental variables include at least one of the following: Internet power, total power of the computer room, ratio of energy consumed to energy consumed by Internet load, refrigeration equipment supply air temperature, refrigeration equipment return air temperature, cold channel temperature and humidity, hot channel temperature and humidity, outdoor temperature and humidity, refrigeration equipment start and stop status, fan start and stop status, compressor start and stop status, fan speed, compressor speed and outdoor temperature.

[0163] This application obtains N first user data that matches the target computer room, and then calculates the load prediction value of the target computer room based on the N first user data, inputs the load prediction value, environmental variables, operating time parameters and switch parameters into the target deep learning model for prediction, and obtains the refrigeration control strategy, thereby improving the control effect of the refrigeration system control strategy on the refrigeration system.

[0164] The present application also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the various processes of the aforementioned control method for a computer room cooling system, achieving the same technical effects. To avoid repetition, the details are omitted here. The computer-readable storage medium may be, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0165] An embodiment of the present application further provides a computer program product, which is stored in a storage medium. The computer program product is executed by at least one processor to implement the various processes of the above-mentioned control method embodiment of the computer room cooling system, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0166] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0167] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0168] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A method for controlling a computer room cooling system, characterized in that: The method comprises: Obtaining N first user data matching the target computer room, where the first user data is Internet behavior data generated by the corresponding user based on the user account during the process of using broadband, where N is a positive integer; Calculating the N first user data to obtain a load prediction value of the target computer room, where the load prediction value is the cooling load required in the target computer room; Input the load prediction value, the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room, and the switching parameters corresponding to the target computer room into the target deep learning model for prediction to obtain a cooling control strategy, wherein the environmental variables include the temperature and humidity parameters corresponding to the target computer room, and the cooling control strategy is used to adjust the operating parameters corresponding to the cooling system of the target computer room, the operating time parameter is the service operation time of the target computer room, and the switching parameter is the number of times the cooling system of the target computer room is turned on and off during the service operation time; The cooling system of the target computer room is controlled based on the cooling control strategy.

2. The method according to claim 1, characterized in that The obtaining of N first user data matching the target computer room includes: Obtaining M pieces of second user data, where the M pieces of second user data correspond one-to-one to M user accounts, the second user data being online behavior data generated by the corresponding user based on the user account during broadband use, where M is greater than or equal to N and is a positive integer; Parsing the M second user data to obtain M parsing results, where the M parsing results correspond one-to-one to the M second user data, and the parsing results are used to indicate whether the computer room information corresponding to the second user data matches the computer room information corresponding to the target computer room; The N first user data are determined from the M second user data according to the M parsing results, and the computer room information indicated by the parsing results corresponding to the first user data matches the computer room information corresponding to the target computer room.

3. The method according to claim 2, characterized in that The parsing of the M second user data to obtain M parsing results includes: performing data cleaning and data parsing on the M second user data to obtain the M parsing results, wherein each of the M parsing results includes a first field and a second field, the first field being used to indicate user access line identification information of the corresponding second user data, and the second field being used to indicate user access location information of the corresponding second user data, the user access line identification information being used to indicate a device identification of a user accessing a network, and the user access location information being used to indicate a geographical location of a user accessing a network, and when the device identification and the geographical location are the same as the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result matches the computer room information corresponding to the target computer room; and when the device identification and the geographical location are different from the computer room information corresponding to the target computer room, the computer room information indicated by the parsing result does not match the computer room information corresponding to the target computer room; Determining the N first user data from the M second user data according to the M parsing results includes: For each piece of second user data, generating computer room information corresponding to each piece of second user data according to the first field and the second field corresponding to the second user data, the computer room information including the device identifier and the geographic location; The N first user data are determined from the M second user data according to the computer room information corresponding to each second user data.

4. The method according to claim 1, wherein The calculating the N first user data to obtain the load prediction value of the target computer room includes: Traversing the N first user data, obtaining Internet access information corresponding to each first user data, the Internet access information including the user's Internet access time period and the user's Internet access traffic; Obtaining a target time period and a bandwidth capacity of the target computer room, wherein the target time period is a service start time period of the target computer room, and the bandwidth capacity of the computer room is a maximum bandwidth that the target computer room can provide; Calculating the N pieces of Internet access information, the target time period, and the bandwidth capacity of the computer room to obtain a bandwidth utilization rate of the target computer room; The bandwidth utilization is determined as the load prediction value of the target computer room.

5. The method according to claim 1, wherein Before inputting the load prediction value, the environmental variables corresponding to the target computer room, and other variables corresponding to the target computer room into a target deep learning model for prediction to obtain a cooling control strategy, the method further includes: Obtaining a training data set, the training data set including historical load prediction values, historical environmental variables, historical operating time parameters, historical switch parameters, and training labels corresponding to the target computer room, the training labels being historical cooling control strategies corresponding to the target computer room, the historical load prediction values, the historical environmental variables, the historical operating time parameters, and the historical switch parameters being data within the same historical time period; The deep learning model is trained based on the training data set to obtain the target deep learning model.

6. The method according to any one of claims 1 to 5, characterized in that The refrigeration control strategy includes at least one of the following: refrigeration equipment supply / return air temperature control strategy, refrigeration equipment fan speed adjustment strategy, refrigeration equipment compressor speed adjustment strategy, refrigeration equipment switch quantity control strategy, humidification / dehumidification control strategy, fluorine gas switching strategy, fresh air switching strategy, dual cooling source switching strategy and compressor load setting strategy.

7. The method according to any one of claims 1 to 5, characterized in that The environmental variables include at least one of the following: Internet power, total power of the computer room, ratio of energy consumed to energy consumed by Internet load, refrigeration equipment supply air temperature, refrigeration equipment return air temperature, cold aisle temperature and humidity, hot aisle temperature and humidity, outdoor temperature and humidity, refrigeration equipment start and stop status, fan start and stop status, compressor start and stop status, fan speed, compressor speed and outdoor temperature.

8. A control device for a computer room cooling system, characterized in that: The device comprises: an acquisition module, configured to acquire N first user data matching the target computer room, wherein the first user data is online behavior data generated by the corresponding user based on the user account during the process of using broadband, and N is a positive integer; a calculation module, configured to calculate the N first user data to obtain a load prediction value of the target computer room, where the load prediction value is the cooling load required in the target computer room; a prediction module, configured to input the load prediction value, the environmental variables corresponding to the target computer room, the operating time parameters corresponding to the target computer room, and the switching parameters corresponding to the target computer room into a target deep learning model for prediction to obtain a refrigeration control strategy, wherein the environmental variables include the temperature and humidity parameters corresponding to the target computer room; the refrigeration control strategy is configured to adjust the operating parameters corresponding to the refrigeration system of the target computer room; the operating time parameter is the service operation time of the target computer room; and the switching parameter is the number of times the refrigeration system of the target computer room is switched on and off during the service operation time; A control module is used to control the refrigeration system of the target computer room based on the refrigeration control strategy.

9. An electronic device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which implements the steps of the method according to any one of claims 1 to 7 when executed by a processor.

11. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images