An operation and maintenance monitoring system and method for an efficient machine room

Abstract

The application discloses a kind of operation and maintenance monitoring system and method for high-efficiency machine room, it is related to operation and maintenance monitoring technical field, the present application includes: S10: the target machine room is divided into regions;S20: the operation and maintenance sequence of each equipment placed in each division area is determined;S30: the temperature of cooling water in each target cooling water pipe is regulated by cooling tower;S40: whether other each equipment is handled by operation and maintenance is selected.The present application is based on the sensitive index of each equipment in the target machine room to the cooling water pipe, realizes the zoned management of each equipment in the target machine room, the sensitive index can consider the heat exchange between the equipment located in adjacent position in the cooling water pipe cooling process of equipment, it is favorable to realize the accurate control of the temperature of the target machine room by cooling water pipe, in this process, without considering the stability of voltage supplied into the target machine room, improve the analysis efficiency of the operation and maintenance monitoring result of target machine room.

Description

Technical Field

[0001] This invention relates to the field of operation and maintenance monitoring technology, specifically an operation and maintenance monitoring system and method for high-efficiency data centers. Background Technology

[0002] The full name of a high-efficiency computer room is a high-efficiency air conditioning and refrigeration computer room. Under the premise of meeting the terminal cooling load demand, it comprehensively considers factors such as load matching degree and system energy consumption rationality, and uses intelligent control technology and management system to systematically optimize equipment parameters and operating strategies to achieve high energy efficiency.

[0003] Currently, operation and maintenance monitoring is typically achieved by comprehensively analyzing the operating parameters of each device placed in a high-efficiency data center, such as current and power, as well as the environmental parameters of each device's location, such as temperature and humidity. This method requires full consideration of the stability of the voltage supplied to the high-efficiency data center and the heat exchange between the devices. The analysis process is relatively complex and prone to errors. Furthermore, existing technologies cannot specifically analyze the fault attributes of the equipment based on the comprehensive analysis results, i.e., local faults and overall faults. This requires operation and maintenance personnel to analyze the fault attributes of the equipment on-site. Summary of the Invention

[0004] The purpose of this invention is to provide an operation and maintenance monitoring system and method for high-efficiency data centers to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for operation and maintenance monitoring of high-efficiency data centers, the method comprising:

[0006] S10: Real-time monitoring of the cooling load difference values ​​at various locations within the target computer room, and calculation of the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipes. Based on the calculation results, the target computer room is divided into areas.

[0007] S20: The temperature of the cooling water in the cooling water pipe is the temperature required by the target computer room. When the target computer room is at the required temperature, the correlation between the sensitivity index of any two devices placed in each division area to the cooling water pipe is analyzed, and the operation and maintenance coefficient of each device placed in each division area is predicted. Then, the operation and maintenance sequence of each device placed in each division area is determined.

[0008] S30: Based on the maintenance sequence of each device in each divided area, analyze the fault attributes of the first maintenance object, and based on the analysis results, regulate the temperature of the cooling water in the cooling water pipes of each target through the cooling tower;

[0009] S40: When regulating the temperature of the cooling water in each target cooling water pipe through the cooling tower, the control terminal controls the first maintenance object to be in a shutdown state. When the temperature at the location of the first maintenance object changes to the temperature value required by the target computer room, the cooling tower control inputs the temperature of the cooling water in each target cooling water pipe to the temperature value required by the target computer room. Then, based on the re-marking of the monitored sensitivity index, it is selected whether to perform maintenance on other equipment.

[0010] Furthermore, S10 includes:

[0011] S101: A temperature sensor is installed on each device placed in the target computer room. The temperature sensor monitors the difference in cooling load at each location in the target computer room in real time. The difference in cooling load is the difference between the real-time temperature value at each location in the target computer room and the temperature value required by the target computer room.

[0012] S102: Based on the real-time cooling load differences at various locations within the target computer room, calculate the sensitivity index of each piece of equipment placed within the target computer room to the cooling water pipes. The calculation formula is: M i =exp(-F i / t i ), where i=1,2,…,n represents the serial number of each device placed in the target computer room, n represents the total number of devices placed in the target computer room, and t i F represents the time required for the real-time cooling load difference value of device i to change from its initial value to 0 during the cooling process of the cooling water pipes cooling the target computer room. i This represents the initial value of the cooling load difference for device i;

[0013] S103: Determine if the sensitivity indices of two adjacent devices are the same. If they are the same, classify the two adjacent devices into the same area; otherwise, classify them into two different areas. Repeat this process for all devices placed in the target computer room to obtain several divided areas. The sensitivity index takes into account the heat exchange between adjacent devices during the cooling process of the cooling water pipes, which is beneficial for achieving precise temperature control of the target computer room by the cooling water pipes.

[0014] Furthermore, S20 includes:

[0015] S201: When the target computer room is at the required temperature, monitor the sensitivity index of each device in the divided area j to the cooling water pipe in real time, where j=1,2,…,m, representing the number of each divided area, and m represents the total number of divided areas;

[0016] Within the defined region j, a device is randomly selected, denoted as device p, where p = 1, 2, ..., n. The sensitivity index M of device p monitored at time t is... pt With sensitivity index M p The difference W between them pt→p Perform calculations, M pt =M p -exp[-(V pt -C) / (t-t0)], if the difference is not equal to 0, then the sensitivity index M pt Mark the value; if the difference is 0, then do not apply the sensitivity index M. pt Mark the values, where C represents the required temperature value of the target computer room, and V... pt t represents the temperature value monitored by the temperature sensor installed on device p at time t, and t0 represents the time when the temperature sensor installed on device p last monitored the temperature value C at the location of device p. Based on the sensitivity index, the abnormal operation of the device is initially identified. Compared with judging whether the device is operating abnormally by the current, voltage, etc., the amount of data processing is reduced and the stability of the input voltage does not need to be considered, which further improves the analysis efficiency.

[0017] S202: Determine the equipment to which the marked sensitivity index belongs, and use the marked sensitivity index sets corresponding to any two determined equipment as training data groups to train the linear regression model. Use the weights of the trained linear regression model as the correlation coefficient between any two equipment. Equipment failure is a common cause of abnormally increased heat generated by equipment. When the heat conduction of equipment increases due to failure, it will reduce the equipment's sensitivity index to the cooling water pipe. Based on the principle of heat exchange, it is known that the failure equipment will indirectly reduce the sensitivity index of adjacent equipment to the cooling water pipe. Therefore, the equipment's sensitivity index and correlation coefficient can better reflect the equipment's failure status.

[0018] S203: According to U ht =W ht→h ×G h→f The operation and maintenance coefficient of device h at time t is calculated, where h represents the ID of a randomly selected device within region j that is associated with sensitive data, h=1,2,…,n, and f represents the ID of a device within region j that has a correlation coefficient with device h, f=1,2,…,n. ht→h The sensitivity index M of device h monitored at time t is represented by... ht With sensitivity index M h The difference between them, G h→f This represents the correlation coefficient between device h and device f;

[0019] Iterate through all devices within region j and determine the maintenance order of each device within region j according to the maintenance coefficient from largest to smallest.

[0020] Furthermore, S30 includes:

[0021] S301: The device corresponding to the maximum value of the operation and maintenance coefficient within the divided area j is taken as the first operation and maintenance object. Based on the distribution of the placement of each device within the divided area j of the target data room, the devices adjacent to the first operation and maintenance object are determined. If the discrete coefficient of the data group composed of the operation and maintenance coefficients corresponding to the determined adjacent devices is less than a set threshold, it indicates that the fault attribute of the first operation and maintenance object is an overall fault. If the discrete coefficient of the data group composed of the operation and maintenance coefficients corresponding to the determined adjacent devices is greater than or equal to the set threshold, it indicates that the fault attribute of the first operation and maintenance object is a local fault. By analyzing the fault attributes of the first operation and maintenance object, it is convenient to regulate the temperature of the related devices when regulating the temperature at the location of the first operation and maintenance object. This can ensure the operating environment of the related devices and quickly and accurately regulate the temperature of the target data room.

[0022] S302: When the fault attribute of the first maintenance object is a global fault, the cooling water pipe used to regulate the temperature of the location of the first maintenance object and the cooling water pipe used to regulate the temperature of the location of the equipment adjacent to the first maintenance object are used as target cooling water pipes, and the temperature of the cooling water in the target cooling water pipes is regulated by the cooling tower.

[0023] The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the location of the first maintenance object is: 2×C-S1, where S1 represents the real-time temperature value of the location of the first maintenance object.

[0024] The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the equipment located adjacent to the first maintenance object is: 2×C-S2, where S2 represents the average temperature value of the locations of all equipment adjacent to the first maintenance object.

[0025] S303: When the fault attribute of the first maintenance object is a local fault, the maximum value of the correlation coefficient between the first maintenance object and each adjacent device is found, and the cooling water pipe used to regulate the temperature of the location of the first maintenance object and the cooling water pipe used to regulate the temperature of the location of the adjacent device corresponding to the maximum value of the correlation coefficient found above are taken as the target cooling water pipes, and the temperature of the cooling water in the target cooling water pipes is regulated by the cooling tower.

[0026] The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the location of the first maintenance object is: 2×C-S1, where S1 represents the real-time temperature value of the location of the first maintenance object.

[0027] The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the adjacent equipment corresponding to the maximum value of the correlation coefficient found above is: 2×C-S3, where S3 represents the real-time temperature value of the adjacent equipment corresponding to the maximum value of the correlation coefficient found above.

[0028] Furthermore, S40 includes: when the temperature of the cooling water in each target cooling water pipe is the temperature value required by the target computer room, the sensitivity index of each device in the divided area j to the cooling water pipe is monitored in real time.

[0029] If the detected sensitivity index is no longer marked, then there is no need to perform maintenance on the other devices within the divided area j.

[0030] If the detected sensitivity index is still marked, repeat the operations from S20 to S40 until no maintenance is required for other devices within the partitioned area j.

[0031] An operation and maintenance monitoring system for high-efficiency data centers, the system comprising a target data center area division module, an operation and maintenance analysis module, a temperature control module, and an operation and maintenance control module;

[0032] The target computer room area division module is used to calculate the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipe based on the real-time monitoring of the cooling load difference value of each location in the target computer room, and to divide the target computer room into areas.

[0033] The operation and maintenance analysis module is used to analyze the correlation between the sensitivity index of any two devices placed in each divided area to the cooling water pipe when the target computer room is at the required temperature, and to predict the operation and maintenance coefficient of each device placed in each divided area. Then, the operation and maintenance sequence of each device placed in each divided area is determined.

[0034] The temperature control module is used to analyze the fault attributes of the first maintenance object and to control the temperature of the cooling water in each target cooling water pipe through the cooling tower.

[0035] The operation and maintenance control module is used to control the working status of the first operation and maintenance object through the control terminal, and to adaptively regulate the temperature of the cooling water in each target cooling water pipe.

[0036] Furthermore, the target computer room area division module includes a cold load difference value monitoring unit, a sensitivity index calculation unit, and an area division unit;

[0037] The cooling load difference monitoring unit monitors the cooling load difference values ​​at various locations within the target computer room in real time through temperature sensors installed on various devices placed within the target computer room.

[0038] The sensitivity index calculation unit calculates the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipe based on the real-time cooling load difference value at each location in the target computer room.

[0039] The area division unit divides the target computer room into areas based on the judgment result that the sensitivity indices of two devices located in adjacent positions are the same.

[0040] Furthermore, the operation and maintenance analysis module includes a sensitivity index marking unit, a correlation coefficient calculation unit, and an operation and maintenance coefficient calculation unit;

[0041] The sensitivity index marking unit selectively marks the sensitivity index of the device based on the real-time changes in the device's sensitivity index.

[0042] The correlation coefficient calculation unit uses the labeled sensitive datasets corresponding to any two devices as training data groups to train the linear regression model, and uses the weights of the trained linear regression model as the correlation coefficient between any two devices.

[0043] The operation and maintenance coefficient calculation unit calculates the real-time operation and maintenance coefficient of the equipment based on the equipment's sensitivity index and correlation coefficient.

[0044] Furthermore, the temperature control module includes a fault attribute analysis unit, a first temperature control unit, and a second temperature control unit.

[0045] The fault attribute analysis unit analyzes the fault attributes of the first maintenance object based on the discrete coefficient of the data group composed of the maintenance coefficients of the devices adjacent to the first maintenance object.

[0046] When the fault attribute of the first maintenance object is an overall fault, the first temperature control unit uses the cooling water pipe used to control the temperature of the location of the first maintenance object and the cooling water pipe used to control the temperature of the location of the equipment adjacent to the first maintenance object as target cooling water pipes, and controls the temperature of the cooling water in the target cooling water pipes through the cooling tower.

[0047] When the fault attribute of the first maintenance object is a local fault, the second temperature control unit searches for the maximum value of the correlation coefficient between the first maintenance object and each adjacent device, and uses the cooling water pipe used to control the temperature of the location of the first maintenance object, and the cooling water pipe used to control the temperature of the location of the adjacent device corresponding to the maximum value of the correlation coefficient found above as the target cooling water pipe, and controls the temperature of the cooling water in the target cooling water pipe through the cooling tower.

[0048] Furthermore, the operation and maintenance control module includes an operation and maintenance control unit and an operation and maintenance reanalysis unit;

[0049] When the operation and maintenance control unit regulates the temperature of the cooling water in each target cooling water pipe through the cooling tower, the control terminal controls the first operation and maintenance object to be in a shutdown state, and when the temperature at the location of the first operation and maintenance object changes to the temperature value required by the target computer room, the cooling tower control inputs the temperature of the cooling water in each target cooling water pipe to the temperature value required by the target computer room.

[0050] When the temperature of the cooling water in each target cooling water pipe is the temperature value required by the target computer room, the operation and maintenance reanalysis unit monitors the sensitivity index of each device in each divided area to the cooling water pipe in real time, and selects whether to perform operation and maintenance on other devices based on the monitoring results.

[0051] Compared with the prior art, the beneficial effects of the present invention are:

[0052] 1. This invention divides the target computer room into zones based on the sensitivity index of each device to the cooling water pipes. This facilitates zoned management of the devices within the target computer room, reducing the workload of equipment operation and maintenance monitoring to a certain extent. At the same time, the sensitivity index can take into account the heat exchange between adjacent devices during the cooling process of the cooling water pipes, which is conducive to achieving precise temperature control of the target computer room by the cooling water pipes. In this process, there is no need to consider the stability of the voltage supplied to the target computer room, further improving the efficiency of analyzing the operation and maintenance monitoring results of the target computer room.

[0053] 2. This invention analyzes the fault attributes of the maintenance object, which facilitates the control of the ambient temperature at the location of the relevant equipment, thereby ensuring a good operating environment for the relevant equipment. By simultaneously cooling the maintenance equipment and related equipment, it is beneficial to quickly control the temperature in the target computer room.

[0054] 3. This invention analyzes the correlation between any two devices by identifying the correlation between two sets of labeled sensitive datasets, and determines the operation and maintenance sequence of each device by combining the real-time sensitivity index of each device. By dividing the area and the operation and maintenance sequence, the scope of the operation and maintenance object is gradually narrowed, which improves the system's operation and maintenance monitoring efficiency for the target data center. Attached Figure Description

[0055] Figure 1 This is a schematic diagram illustrating the workflow of an operation and maintenance monitoring method for high-efficiency data centers according to the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] like Figure 1 As shown, this invention provides a technical solution for an operation and maintenance monitoring system and method for high-efficiency data centers, and a method for operation and maintenance monitoring of high-efficiency data centers, the method comprising:

[0058] S10: Real-time monitoring of the cooling load difference values ​​at various locations within the target computer room, and calculation of the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipes. Based on the calculation results, the target computer room is divided into areas.

[0059] S10 includes:

[0060] S101: A temperature sensor is installed on each device placed in the target computer room. The temperature sensor monitors the difference in cooling load at each location in the target computer room in real time. The difference in cooling load is the difference between the real-time temperature value at each location in the target computer room and the temperature value required by the target computer room.

[0061] S102: Based on the real-time cooling load differences at various locations within the target computer room, calculate the sensitivity index of each piece of equipment placed within the target computer room to the cooling water pipes. The calculation formula is: M i =exp(-F i / t i The cooling water pipes are laid under the raised floor of the target computer room, where i = 1, 2, ..., n represents the serial number of each piece of equipment placed in the target computer room, n represents the total number of pieces of equipment placed in the target computer room, and t i This represents the time required for the real-time cooling load difference value of device i to change from its initial value to 0 during the cooling process of the cooling water pipes in the target computer room (where all equipment is operating normally). F i Exp() represents the initial value of the cooling load difference of device i. exp() represents an exponential function with base e and e=2.73. The initial value is the difference between the temperature value monitored by the temperature sensor installed on device i and the temperature value required by the target computer room just before the cooling water pipe cools the target computer room.

[0062] S103: Determine whether the sensitivity indices of two adjacent devices are the same. If they are the same, divide the two adjacent devices into the same area. If they are not the same, divide the two adjacent devices into two different areas. Traverse all the devices placed in the target computer room to obtain several division areas.

[0063] S20: The temperature of the cooling water in the cooling water pipe is the temperature required by the target computer room. When the target computer room is at the required temperature, the correlation between the sensitivity index of any two devices placed in each division area to the cooling water pipe is analyzed, and the operation and maintenance coefficient of each device placed in each division area is predicted. Then, the operation and maintenance sequence of each device placed in each division area is determined.

[0064] S20 includes:

[0065] S201: When the target computer room is at the required temperature, monitor the sensitivity index of each device in the divided area j to the cooling water pipe in real time, where j=1,2,…,m, representing the number of each divided area, and m represents the total number of divided areas;

[0066] Within the defined region j, a device is randomly selected, denoted as device p, where p = 1, 2, ..., n. The sensitivity index M of device p monitored at time t is... pt With sensitivity index M p The difference W between them pt→p Perform calculations, M pt =M p -exp[-(V pt -C) / (t-t0)], if the difference is not equal to 0, then the sensitivity index M pt Mark the value; if the difference is 0, then do not apply the sensitivity index M. pt Mark the values, where C represents the required temperature value of the target computer room, and V... pt t represents the temperature value monitored by the temperature sensor installed on device p at time t, and t0 represents the time when the temperature sensor installed on device p last monitored the temperature value C at the location of device p.

[0067] S202: Determine the device to which the sensitivity index of the label belongs, and use the label sensitivity index sets corresponding to any two determined devices as training data groups to train the linear regression model. The linear regression model is: Y=k×X+b. Use the weights of the trained linear regression model as the correlation coefficient between any two devices.

[0068] S203: According to U ht =W ht→h ×G h→fThe operation and maintenance coefficient of device h at time t is calculated, where h represents the ID of a randomly selected device within region j that is associated with sensitive data, h=1,2,…,n, and f represents the ID of a device within region j that has a correlation coefficient with device h, f=1,2,…,n. ht→h The sensitivity index M of device h monitored at time t is represented by... ht With sensitivity index M h The difference between them, G h→f This represents the correlation coefficient between device h and device f;

[0069] Traverse all devices within the partitioned region j, and determine the maintenance order of each device within partitioned region j according to the maintenance coefficient from largest to smallest, prioritizing the maintenance processing of devices with large maintenance data;

[0070] S30: Based on the maintenance sequence of each device in each divided area, analyze the fault attributes of the first maintenance object, and based on the analysis results, regulate the temperature of the cooling water in the cooling water pipes of each target through the cooling tower;

[0071] S30 includes:

[0072] S301: The device corresponding to the maximum value of the operation and maintenance coefficient within the divided area j is taken as the first operation and maintenance object. Based on the distribution of the placement of each device within the divided area j of the target computer room, the devices adjacent to the first operation and maintenance object are determined. If the discrete coefficient of the data group composed of the operation and maintenance coefficients corresponding to the determined adjacent devices is less than the set threshold, it indicates that the fault attribute of the first operation and maintenance object is an overall fault. If the discrete coefficient of the data group composed of the operation and maintenance coefficients corresponding to the determined adjacent devices is greater than or equal to the set threshold, it indicates that the fault attribute of the first operation and maintenance object is a local fault.

[0073] S302: When the fault attribute of the first maintenance object is a global fault, the cooling water pipe used to regulate the temperature of the location of the first maintenance object and the cooling water pipe used to regulate the temperature of the location of the equipment adjacent to the first maintenance object are used as target cooling water pipes, and the temperature of the cooling water in the target cooling water pipes is regulated by the cooling tower.

[0074] The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the location of the first maintenance object is: 2×C-S1, where S1 represents the real-time temperature value of the location of the first maintenance object.

[0075] The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the equipment located adjacent to the first maintenance object is: 2×C-S2, where S2 represents the average temperature value of the locations of all equipment adjacent to the first maintenance object, and the average temperature value = the sum of the temperatures of the locations of all equipment adjacent to the first maintenance object / the total number of equipment adjacent to the first maintenance object.

[0076] S303: When the fault attribute of the first maintenance object is a local fault, the maximum value of the correlation coefficient between the first maintenance object and each adjacent device is found, and the cooling water pipe used to regulate the temperature of the location of the first maintenance object and the cooling water pipe used to regulate the temperature of the location of the adjacent device corresponding to the maximum value of the correlation coefficient found above are taken as the target cooling water pipes, and the temperature of the cooling water in the target cooling water pipes is regulated by the cooling tower.

[0077] The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the location of the first maintenance object is: 2×C-S1, where S1 represents the real-time temperature value of the location of the first maintenance object.

[0078] The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the adjacent equipment corresponding to the maximum value of the correlation coefficient found above is: 2×C-S3, where S3 represents the real-time temperature value of the adjacent equipment corresponding to the maximum value of the correlation coefficient found above.

[0079] S40: When regulating the temperature of the cooling water in each target cooling water pipe through the cooling tower, the control terminal controls the first maintenance object to be in a shutdown state. When the temperature at the location of the first maintenance object changes to the temperature value required by the target computer room, the cooling tower control inputs the temperature of the cooling water in each target cooling water pipe to the temperature value required by the target computer room. Then, based on the re-marking of the monitored sensitivity index, it selects whether to perform maintenance on other equipment. The control terminal is used to control the working status of each equipment in the target computer room. The cooling water pipes laid in the target computer room can individually cool each piece of equipment.

[0080] S40 includes: when the temperature of the cooling water in each target cooling water pipe is the temperature value required by the target computer room, the sensitivity index of each device in the divided area j to the cooling water pipe is monitored in real time.

[0081] If the detected sensitivity index is no longer marked, then there is no need to perform maintenance on the other devices in the partitioned area j. The other devices refer to all devices in the partitioned area j except for the first maintenance object.

[0082] If the detected sensitivity index is still marked, repeat the operations from S20 to S40 until no maintenance is required for other devices within the partitioned area j.

[0083] An operation and maintenance monitoring system for high-efficiency data centers, comprising a target data center area division module, an operation and maintenance analysis module, a temperature control module, and an operation and maintenance control module;

[0084] The target computer room area division module is used to calculate the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipes based on the real-time monitoring of the cooling load difference values ​​of each location in the target computer room, and to divide the target computer room into areas.

[0085] The target data center area division module includes a cooling load difference monitoring unit, a sensitivity index calculation unit, and an area division unit.

[0086] The cooling load difference monitoring unit monitors the cooling load difference values ​​at various locations within the target computer room in real time through temperature sensors installed on various devices placed within the target computer room.

[0087] The sensitivity index calculation unit calculates the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipes based on the real-time cooling load difference value at each location in the target computer room.

[0088] The area division unit divides the target computer room into areas based on the judgment result of whether the sensitivity indices of two devices located in adjacent positions are the same;

[0089] The operation and maintenance analysis module is used to analyze the correlation between the sensitivity index of any two devices placed in each division area to the cooling water pipe when the target computer room is under the required temperature, and to predict the operation and maintenance coefficient of each device placed in each division area. Then, the operation and maintenance sequence of each device placed in each division area is determined.

[0090] The operation and maintenance analysis module includes a sensitivity index marking unit, a correlation coefficient calculation unit, and an operation and maintenance coefficient calculation unit;

[0091] The sensitivity index marking unit selectively marks the sensitivity index of the equipment based on the real-time changes in the equipment's sensitivity index;

[0092] The correlation coefficient calculation unit uses the labeled sensitive datasets corresponding to any two devices as training data groups to train the linear regression model, and uses the weights of the trained linear regression model as the correlation coefficient between any two devices.

[0093] The operation and maintenance coefficient calculation unit calculates the real-time operation and maintenance coefficient of the equipment based on the equipment's sensitivity index and correlation coefficient;

[0094] The temperature control module is used to analyze the fault attributes of the first maintenance object and to control the temperature of the cooling water in the cooling water pipes of each target through the cooling tower.

[0095] The temperature control module includes a fault attribute analysis unit, a first temperature control unit, and a second temperature control unit.

[0096] The fault attribute analysis unit analyzes the fault attributes of the first maintenance object based on the discrete coefficient of the data group composed of the maintenance coefficients of the devices adjacent to the first maintenance object.

[0097] When the fault attribute of the first maintenance object is an overall fault, the first temperature control unit uses the cooling water pipe used to control the temperature of the location of the first maintenance object and the cooling water pipe used to control the temperature of the location of the equipment adjacent to the first maintenance object as target cooling water pipes, and controls the temperature of the cooling water in the target cooling water pipes through the cooling tower.

[0098] When the fault attribute of the first maintenance object is a local fault, the second temperature control unit searches for the maximum value of the correlation coefficient between the first maintenance object and each adjacent device, and takes the cooling water pipe used to control the temperature of the location of the first maintenance object and the cooling water pipe used to control the temperature of the location of the adjacent device corresponding to the maximum value of the correlation coefficient as the target cooling water pipe, and adjusts the temperature of the cooling water in the target cooling water pipe through the cooling tower.

[0099] The operation and maintenance control module is used to control the working status of the first operation and maintenance object through the control terminal, and to adaptively regulate the temperature of the cooling water in each target cooling water pipe;

[0100] The operation and maintenance control module includes an operation and maintenance control unit and an operation and maintenance reanalysis unit;

[0101] When the operation and maintenance control unit regulates the temperature of the cooling water in each target cooling water pipe through the cooling tower, the control terminal controls the first operation and maintenance object to be in a shutdown state, and when the temperature at the location of the first operation and maintenance object changes to the temperature value required by the target computer room, the cooling tower control inputs the temperature of the cooling water in each target cooling water pipe to the temperature value required by the target computer room.

[0102] When the temperature of the cooling water in each target cooling water pipe is within the temperature value required by the target computer room, the operation and maintenance reanalysis unit monitors the sensitivity index of each device in each divided area to the cooling water pipe in real time, and selects whether to perform operation and maintenance on other devices based on the monitoring results.

[0103] Example 1: Assume the required temperature value of the target computer room is C=24℃, and the temperature sensor installed on device 1 monitors the temperature value V at time t. 1t =26℃, and the temperature sensor installed on equipment 1 last detected a temperature of C at the location of equipment 1 at a time t0 = t-2 min. Then, the sensitivity index of equipment 1 to the cooling water pipe is:

[0104] M 1t =exp[-(V 1t -C)) / (t-t0)]=exp[-(26℃-24℃)) / (t-t+2min)]=0.37;

[0105] The sensitivity index of device 1 to the cooling water pipe is 0.37.

[0106] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for operation and maintenance monitoring of high-efficiency data centers, characterized in that: The method includes: S10: Real-time monitoring of the cooling load difference values ​​at various locations within the target computer room, and calculation of the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipes. Based on the calculation results, the target computer room is divided into areas. S10 includes: S101: A temperature sensor is installed on each device placed in the target computer room. The temperature sensor monitors the difference in cooling load at each location in the target computer room in real time. The difference in cooling load is the difference between the real-time temperature value at each location in the target computer room and the temperature value required by the target computer room. S102: Based on the real-time cooling load differences at various locations within the target computer room, calculate the sensitivity index of each piece of equipment placed within the target computer room to the cooling water pipes. The calculation formula is: M i =exp(-F i / t i ), where i=1,2,…,n represents the serial number of each device placed in the target computer room, n represents the total number of devices placed in the target computer room, and t i F represents the time required for the real-time cooling load difference value of device i to change from its initial value to 0 during the cooling process of the cooling water pipes cooling the target computer room. i This represents the initial value of the cooling load difference for device i; S103: Determine whether the sensitivity indices of two adjacent devices are the same. If they are the same, divide the two adjacent devices into the same area. If they are not the same, divide the two adjacent devices into two different areas. Traverse all the devices placed in the target computer room to obtain several division areas. S20: The temperature of the cooling water in the cooling water pipe is the temperature required by the target computer room. When the target computer room is at the required temperature, the correlation between the sensitivity index of any two devices placed in each division area to the cooling water pipe is analyzed, and the operation and maintenance coefficient of each device placed in each division area is predicted. Then, the operation and maintenance sequence of each device placed in each division area is determined. S201: When the target computer room is at the required temperature, monitor the sensitivity index of each device in the divided area j to the cooling water pipe in real time, where j=1,2,…,m, representing the number of each divided area, and m represents the total number of divided areas; Within the defined region j, a device is randomly selected, denoted as device p, where p = 1, 2, ..., n. The sensitivity index M of device p monitored at time t is... pt With sensitivity index M p The difference W between them pt→p Perform calculations, M pt =M p -exp[-(V pt -C) / (t-t0)], if the difference is not equal to 0, then the sensitivity index M pt Mark the value; if the difference is 0, then do not apply the sensitivity index M. pt Mark the values, where C represents the required temperature value of the target computer room, and V... pt t represents the temperature value monitored by the temperature sensor installed on device p at time t, and t0 represents the time when the temperature sensor installed on device p last monitored the temperature value C at the location of device p. S202: Determine the device to which the sensitivity index of the label belongs, use the label sensitivity index set corresponding to any two determined devices as training data group to train the linear regression model, and use the weight of the trained linear regression model as the correlation coefficient between any two devices. S203: According to U ht =W ht→h ×G h→f The operation and maintenance coefficient of device h at time t is calculated, where h represents the ID of a randomly selected device within region j that is associated with sensitive data, h=1,2,…,n, f represents the ID of a device within region j that has a correlation coefficient with device h, f=1,2,…,n, W ht→h The sensitivity index M of device h monitored at time t is represented by... ht With sensitivity index M h The difference between them, G h→f This represents the correlation coefficient between device h and device f; S30: Based on the maintenance sequence of each device in each divided area, analyze the fault attributes of the first maintenance object. Based on the analysis results, regulate the temperature of the cooling water in the cooling water pipes of each target through the cooling tower. The fault attributes include overall faults and local faults. S40: When the temperature of the cooling water in each target cooling water pipe is regulated by the cooling tower, the control terminal controls the first maintenance object to be in a shutdown state. When the temperature at the location of the first maintenance object changes to the temperature value required by the target computer room, the cooling tower control inputs the temperature of the cooling water in each target cooling water pipe to the temperature value required by the target computer room. After that, the sensitivity index of each device in the divided area j to the cooling water pipe is monitored in real time. If the detected sensitivity index is no longer marked, then there is no need to perform maintenance on the other devices within the divided area j. If the detected sensitivity index is still marked, repeat the operations from S20 to S40 until no maintenance is required for other devices within the partitioned area j.

2. The operation and maintenance monitoring method for a high-efficiency data center according to claim 1, characterized in that: S20 further includes: Iterate through all devices within region j and determine the maintenance order of each device within region j according to the maintenance coefficient from largest to smallest.

3. The operation and maintenance monitoring method for a high-efficiency data center according to claim 2, characterized in that: S30 includes: S301: The device corresponding to the maximum value of the operation and maintenance coefficient within the divided area j is taken as the first operation and maintenance object. Based on the distribution of the placement of each device within the divided area j of the target computer room, the devices adjacent to the first operation and maintenance object are determined. If the discrete coefficient of the data group composed of the operation and maintenance coefficients corresponding to the determined adjacent devices is less than the set threshold, it indicates that the fault attribute of the first operation and maintenance object is an overall fault. If the discrete coefficient of the data group composed of the operation and maintenance coefficients corresponding to the determined adjacent devices is greater than or equal to the set threshold, it indicates that the fault attribute of the first operation and maintenance object is a local fault. S302: When the fault attribute of the first maintenance object is a global fault, the cooling water pipe used to regulate the temperature of the location of the first maintenance object and the cooling water pipe used to regulate the temperature of the location of the equipment adjacent to the first maintenance object are used as target cooling water pipes, and the temperature of the cooling water in the target cooling water pipes is regulated by the cooling tower. The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the location of the first maintenance object is: 2×C-S1, where S1 represents the real-time temperature value of the location of the first maintenance object. The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the equipment located adjacent to the first maintenance object is: 2×C-S2, where S2 represents the average temperature value of the locations of all equipment adjacent to the first maintenance object. S303: When the fault attribute of the first maintenance object is a local fault, the maximum value of the correlation coefficient between the first maintenance object and each adjacent device is found, and the cooling water pipe used to regulate the temperature of the location of the first maintenance object and the cooling water pipe used to regulate the temperature of the location of the adjacent device corresponding to the maximum value of the correlation coefficient found above are taken as the target cooling water pipes, and the temperature of the cooling water in the target cooling water pipes is regulated by the cooling tower. The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the location of the first maintenance object is: 2×C-S1, where S1 represents the real-time temperature value of the location of the first maintenance object. The temperature of the cooling water in the cooling water pipe used to regulate the temperature of the adjacent equipment corresponding to the maximum value of the correlation coefficient found above is: 2×C-S3, where S3 represents the real-time temperature value of the adjacent equipment corresponding to the maximum value of the correlation coefficient found above.

4. An operation and maintenance monitoring system for high-efficiency data centers, applied to the operation and maintenance monitoring method for high-efficiency data centers as described in any one of claims 1-3, characterized in that: The system includes a target data center area division module, an operation and maintenance analysis module, a temperature control module, and an operation and maintenance control module. The target computer room area division module is used to calculate the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipe based on the real-time monitoring of the cooling load difference value of each location in the target computer room, and to divide the target computer room into areas. The operation and maintenance analysis module is used to analyze the correlation between the sensitivity index of any two devices placed in each divided area to the cooling water pipe when the target computer room is at the required temperature, and to predict the operation and maintenance coefficient of each device placed in each divided area. Then, the operation and maintenance sequence of each device placed in each divided area is determined. The temperature control module is used to analyze the fault attributes of the first maintenance object and to control the temperature of the cooling water in each target cooling water pipe through the cooling tower. The operation and maintenance control module is used to control the working status of the first operation and maintenance object through the control terminal, and to adaptively regulate the temperature of the cooling water in each target cooling water pipe.

5. The operation and maintenance monitoring system for high-efficiency data centers according to claim 4, characterized in that: The target computer room area division module includes a cold load difference value monitoring unit, a sensitivity index calculation unit, and an area division unit; The cooling load difference monitoring unit monitors the cooling load difference values ​​at various locations within the target computer room in real time through temperature sensors installed on various devices placed within the target computer room. The sensitivity index calculation unit calculates the sensitivity index of each piece of equipment placed in the target computer room to the cooling water pipe based on the real-time cooling load difference value at each location in the target computer room. The area division unit divides the target computer room into areas based on the judgment result that the sensitivity indices of two devices located in adjacent positions are the same.

6. The operation and maintenance monitoring system for high-efficiency data centers according to claim 5, characterized in that: The operation and maintenance analysis module includes a sensitivity index marking unit, a correlation coefficient calculation unit, and an operation and maintenance coefficient calculation unit. The sensitivity index marking unit selectively marks the sensitivity index of the device based on the real-time changes in the device's sensitivity index. The correlation coefficient calculation unit uses the labeled sensitive datasets corresponding to any two devices as training data groups to train the linear regression model, and uses the weights of the trained linear regression model as the correlation coefficient between any two devices. The operation and maintenance coefficient calculation unit calculates the real-time operation and maintenance coefficient of the equipment based on the equipment's sensitivity index and correlation coefficient.

7. The operation and maintenance monitoring system for high-efficiency data centers according to claim 6, characterized in that: The temperature control module includes a fault attribute analysis unit, a first temperature control unit, and a second temperature control unit. The fault attribute analysis unit analyzes the fault attributes of the first maintenance object based on the discrete coefficient of the data group composed of the maintenance coefficients of the devices adjacent to the first maintenance object. When the fault attribute of the first maintenance object is an overall fault, the first temperature control unit uses the cooling water pipe used to control the temperature of the location of the first maintenance object and the cooling water pipe used to control the temperature of the location of the equipment adjacent to the first maintenance object as target cooling water pipes, and controls the temperature of the cooling water in the target cooling water pipes through the cooling tower. When the fault attribute of the first maintenance object is a local fault, the second temperature control unit searches for the maximum value of the correlation coefficient between the first maintenance object and each adjacent device, and uses the cooling water pipe used to control the temperature of the location of the first maintenance object and the cooling water pipe used to control the temperature of the location of the adjacent device corresponding to the maximum value of the correlation coefficient as the target cooling water pipe, and controls the temperature of the cooling water in the target cooling water pipe through the cooling tower.

8. The operation and maintenance monitoring system for high-efficiency data centers according to claim 7, characterized in that: The operation and maintenance control module includes an operation and maintenance control unit and an operation and maintenance reanalysis unit; When the operation and maintenance control unit regulates the temperature of the cooling water in each target cooling water pipe through the cooling tower, the control terminal controls the first operation and maintenance object to be in a shutdown state, and when the temperature at the location of the first operation and maintenance object changes to the temperature value required by the target computer room, the cooling tower control inputs the temperature of the cooling water in each target cooling water pipe to the temperature value required by the target computer room. When the temperature of the cooling water in each target cooling water pipe is the temperature value required by the target computer room, the operation and maintenance reanalysis unit monitors the sensitivity index of each device in each divided area to the cooling water pipe in real time, and selects whether to perform operation and maintenance on other devices based on the monitoring results.

Images