Magnetic suspension phase change cooling energy-saving control method and control system based on AI algorithm

Abstract

The invention discloses a magnetic suspension phase change cooling energy-saving control method and system based on an AI algorithm, and relates to the technical field of equipment control and regulation, and the control system comprises an environment layout data collection module, a control critical condition setting module, a current environment monitoring module and a control planning module. Equipment data and equipment adjustment historical data in a data machine room are acquired through an environment layout data acquisition module, a critical condition setting module is controlled to set a critical trigger condition for adjusting the temperature in the data machine room by using a magnetic suspension phase change cooling technology, and equipment operation is monitored in real time through a current environment monitoring module. When it is monitored that the current environment data meet the critical triggering condition, a temperature adjustment control instruction is sent, a parameter control planning model is established through a control planning module, and after the instruction is received, the temperature of an air conditioner in the data machine room is adjusted and controlled; and on the basis that the magnetic suspension phase change cooling technology has an energy-saving effect, the energy consumption is further reduced, and energy conservation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment control and regulation, and in particular to a magnetic suspension phase change cooling energy-saving control method and control system based on an AI algorithm. Background Art

[0002] The use of magnetic suspension phase change cooling technology can realize the cooling function of different spaces, such as data rooms, and at the same time play an energy-saving role. The magnetic suspension phase change cooling technology uses evaporative condensers, high-efficiency refrigerant pumps, magnetic suspension compressors and different indoor terminals as basic hardware, and uses an intelligent control system to achieve intelligent regulation of the temperature in the space. The process of using magnetic suspension phase change cooling technology to achieve energy saving while cooling includes: the gaseous refrigerant at the outlet of the air-conditioning terminal evaporator is sucked in and pressurized by the compressor and then enters the heat exchanger for condensation. The top cooling water nozzle acts evenly on the large-area stainless steel heat exchange coil to absorb the condensation heat of the refrigerant. The heated cooling water enters the packing layer and is fully contacted with the dry air under the action of the fan to force evaporation. The evaporated cooling water will absorb the heat of the remaining cooling water around the packing to reduce the cooling water temperature. The low-temperature cooling water falls into the water tank at the bottom of the tower. Under the action of the spray pump, it is pumped to the spray pipe on the top of the tower and then sprayed out from the nozzle to circulate and cool the heat exchange coil. The float valve in the water tank senses the water level. When the water level drops to the threshold value, the valve can be automatically opened for water replenishment.

[0003] However, when applying magnetic levitation phase change cooling technology to data rooms, there are a large number of precision electronic equipment in the data room. These devices will generate a lot of heat during operation. The heat generated by the equipment directly affects the temperature of the room. Effective heat dissipation measures need to be taken to reduce the temperature of the room. Using magnetic levitation phase change cooling technology to control the temperature of air-conditioning equipment in the room can solve this problem. However, due to the different equipment layouts in different data rooms, the required space temperature will also be different. Some equipment in the data room may have continuous failures and cannot operate. When a large number of concentrated equipment fails, the heat generated by the equipment will drop significantly. At this time, there is no need to maintain the original low temperature of the air-conditioning in the room. It can be adjusted to a certain extent, which can reduce the energy consumption of the refrigeration system. The existing technology fails to detect this abnormal situation in time to use magnetic levitation phase change cooling technology to make adaptive automatic adjustment to the temperature of the air-conditioning in the room when necessary, and cannot further reduce energy consumption. Summary of the invention

[0004] The purpose of the present invention is to provide a magnetic suspension phase change cooling energy-saving control method and control system based on AI algorithm to solve the problems raised in the prior art.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a magnetic levitation phase change cooling energy-saving control system based on an AI algorithm, comprising an environment layout data acquisition module, a control critical condition setting module, a current environment monitoring module and a control planning module, wherein the environment layout data acquisition module is used to collect equipment layout information, equipment operation history data and parameter adjustment history data in the data room; the control critical condition setting module is used to analyze the equipment operation history data, and set the critical trigger condition for adjusting the temperature in the data room using the magnetic levitation phase change cooling technology; the current environment monitoring module is used to monitor the operation of the equipment in the current environment in real time, and a temperature adjustment control instruction is sent when it is monitored that the current environment data meets the critical trigger condition; the control planning module is used to establish a parameter control planning model, and after receiving the temperature adjustment control instruction, the temperature in the data room is adjusted and controlled according to the model.

[0006] Furthermore, the environmental layout data acquisition module includes a computer room equipment layout acquisition unit, an equipment operation data acquisition unit and an adjustment data acquisition unit. The computer room equipment layout acquisition unit is used to establish a three-dimensional spatial model of the data room, collect equipment space layout information in the data room cabinet, and divide the equipment placed in adjacent positions into the same right parallelepiped space. The right parallelepiped here refers to a cuboid or a cube, and the adjacent positions refer to adjacent layers in a cabinet. For example: if two devices are placed on the first and second layers in a cabinet, the relative positions between the two devices belong to adjacent positions, and the number of devices contained in the different divided right parallelepiped spaces and the center point position information of the different divided right parallelepiped spaces are collected; the equipment operation data acquisition unit is used to collect historical operation fault data of equipment in the cabinet of the same data room, including the previous simultaneous failures that caused the equipment to stop The number of running equipment and the duration of equipment failure, the duration of the failure refers to the time interval from the start of monitoring to the time when the equipment is repaired and can operate normally; the adjustment data acquisition unit is used to collect different data of air-conditioning temperature in the computer room controlled by magnetic levitation phase change cooling technology when there was no equipment failure in the past. The magnetic levitation phase change cooling technology uses evaporative condensers, high-efficiency refrigerant pumps, magnetic levitation compressors and different indoor terminals as basic hardware, and is equipped with an intelligent control system, which can control the temperature of indoor air-conditioning equipment to achieve cooling and energy saving at the same time. For example: the evaporative condenser adopts two-stage evaporative cooling, namely coil spray evaporation and cooling water evaporation in the packing area. After the first-stage evaporation, the downward flowing water is heated, passes through the packing area, and continues the second-stage forced air cooling evaporation. After cooling, it is recycled, which can maximize the use of natural cold sources, greatly reduce energy consumption and achieve energy saving.

[0007] Furthermore, the control critical condition setting module includes an abnormal data analysis unit and a critical coefficient setting unit. The abnormal data analysis unit is used to retrieve historical operation failure data and analyze the necessary coefficients for adjusting the air-conditioning temperature in the data room when equipment failures occurred at different times in the past. The critical coefficient setting unit is used to set the critical coefficient for triggering the air-conditioning temperature adjustment based on the necessary coefficients obtained by analysis.

[0008] Furthermore, the current environment monitoring module includes an environment data monitoring unit, a data comparison unit and a parameter control trigger selection unit. The environment data monitoring unit is used to perform real-time monitoring of the equipment operation status in the current data room, count the number of equipment failures and estimate the duration of the equipment failure when an equipment failure occurs, and analyze the necessary coefficient for adjusting the temperature in the room using the magnetic levitation phase change cooling technology; the data comparison unit is used to compare the necessary coefficient and the critical coefficient for adjusting the temperature in the room using the magnetic levitation phase change cooling technology; the parameter control trigger selection unit is used to trigger the control and adjustment of the temperature in the current room when the current necessary coefficient is greater than or equal to the critical coefficient.

[0009] Furthermore, the control planning module includes a control planning model building unit, an environmental data input unit and a parameter adjustment control unit. The control planning model building unit is used to generate training samples based on the collected equipment space layout information, and establish a parameter control planning model after fitting the training samples; the environmental data input unit is used to input the spatial layout data of non-faulty equipment in the current data room into the parameter control planning model after receiving a trigger instruction to control and adjust the temperature in the current computer room; the parameter adjustment control unit is used to use magnetic levitation phase change cooling technology to adjust and control the temperature in the data room based on the data output by the model.

[0010] The magnetic suspension phase change cooling energy-saving control method based on AI algorithm includes the following steps:

[0011] Z100: collects equipment layout information, equipment operation history data and parameter adjustment history data in the data room;

[0012] Z101: Analyze the historical data of equipment operation and set the critical trigger conditions for adjusting the temperature in the data center using magnetic suspension phase change cooling technology;

[0013] Z102: Monitor the operation of equipment in the current environment in real time, and send temperature adjustment control instructions when the current environmental data meets the critical trigger conditions;

[0014] Z103: Establish a parameter control planning model, and after receiving the temperature adjustment control instruction, adjust and control the temperature in the data center according to the model.

[0015] Furthermore, in step Z100: a three-dimensional spatial model of the data room is established, the spatial layout information of the equipment in the cabinet of the data room is collected, the equipment placed in adjacent positions is divided into the same right parallelepiped space, and a random data room is collected and a total of n right parallelepipeds are divided, and the number of equipment contained in the divided n right parallelepiped spaces is set to {E1, E2, ...Er, ...En}, the center point position coordinates of the n right parallelepipeds are collected, and the minimum circle radius R that can cover the center points of the n right parallelepipeds is obtained, and the minimum circle radius R can be obtained using the minimum circle coverage derivative algorithm, and the historical operation failure data of the equipment in the cabinet of the same data room is collected, including the number of equipment that stopped running due to simultaneous failures in the past and the duration of the equipment failures, and the air-conditioning temperature data in different data rooms controlled by magnetic levitation phase change cooling technology when no equipment failures have occurred in the past are collected.

[0016] Further, in step Z101: for a random data room: the number of devices that have failed at the same time and stopped running is obtained as A = {A1, A2, ... Ai, ... Am}, m represents the number of times the equipment in the cabinet of a random data room fails, the total number of devices in the cabinet of a random data room is counted as M, and a total of k devices fail at the same time at a random time, k = Ai, the failure maintenance time of k faulty devices is T = {T1, T2, ... Tj, ... Tk}, j represents the jth faulty device, according to Wi = (Ai) / M + [∑ k j＝1 (Tj) / k analyzes the necessary coefficients Wi for adjusting the air conditioning temperature in the data room when a random equipment failure occurs in the past, and obtains the necessary coefficients set for adjusting the air conditioning temperature in the data room when the equipment failure occurs m times in the past as {W1, W2, ...Wi, ...Wm}. The necessary coefficients are sorted in order from large to small, and the sorted necessary coefficients are randomly divided into f groups. All the necessary coefficients in the first group are greater than those in the second group. The necessary coefficient mean set in each group in a random grouping result is obtained as {w1, w2, ...we, ...wf}, where we represents the necessary coefficient mean in the eth group. According to the formula L=[(∑ f e＝1 (we-∑ f e＝1 (we) / f)) 2 / f] 1 / 2Calculate the reference degree L of a randomly selected grouping result for setting the critical coefficient, compare the reference degrees of different grouping results for setting the critical coefficient, select the grouping result with the highest reference degree, obtain the necessary coefficients of the first group in the grouping result with the highest reference degree, calculate the average value of the necessary coefficients of the first group, and set the obtained average value as the critical coefficient for triggering air-conditioning temperature adjustment, resulting in the critical coefficient U.

[0017] Further, in step Z102: Monitor that the number of currently failed and stopped devices in the data computer room is b, obtain the failure type of a currently randomly selected failed device, retrieve the set of previous failure duration of devices with the corresponding type of failure in the past as {t1, t2,... tv,... tg}, where g represents the number of devices with the corresponding type of failure in the past, estimate the failure duration Fa of the currently randomly selected failed device, Fa = ∑ g v＝1 (tv) / g, where a represents the a-th currently failed device, and v represents the v-th device with the corresponding type of failure in the past. By the same method, estimate the set of failure durations of currently failed and stopped devices as {F1, F2,... Fa,... Fb}, and analyze the necessary coefficient W for currently using the magnetic levitation phase change cooling technology to adjust the temperature in the computer room ’ , W ’ = b / M + [∑ b a＝1 (Fa)] / b, compare W ’ and U. If W ’ ≥U, send a temperature adjustment control instruction; if W ’ <U, do not send a temperature adjustment control instruction;

[0018] The purpose of setting the critical coefficient is to consider the situation that there may be a small number of devices with persistent failures. The failure of a small number of devices and the inability to operate have a small impact on the overall heat generation of the devices in the computer room, that is, the heat generation will not decrease significantly, and there may be a situation where devices fail for a short time, that is, the failure duration is short. Because the devices will be repaired and operate normally in a short time, if the temperature is adjusted during the failure, the air-conditioning temperature needs to be adjusted again after the failure. Frequent temperature adjustment in a short time will also increase energy consumption, which is not conducive to energy conservation. Therefore, the critical coefficient is set, and the temperature adjustment is triggered only when the device data in the computer room reaches the critical condition, effectively avoiding the above two situations, further reducing energy consumption while effectively increasing the role of temperature adjustment, and further realizing the energy conservation of the refrigeration system.

[0019] Further, in step Z103: According to Dc = ∑ n r＝1 (Er) / n + 1 / (π*R2 ) is analyzed to obtain the equipment layout concentration index Dc of a random data room, r represents the rth straight parallelepiped divided in a random data room, c represents the cth data room, and the equipment layout concentration index set of different data rooms is analyzed in the same way as {D1, D2, ...Dc, ...Ds}, s represents the total number of data rooms, and the air-conditioning temperature set controlled by the magnetic levitation phase change cooling technology when there is no equipment failure in the s data rooms is retrieved as Q = {Q1, Q2, ...Qc, ...Qs}. When there is no equipment failure, the room is in a constant temperature state, that is, in a constant temperature state, and the training samples {(D1, Q1), (D2, Q2), ...(Dc, Qc), ...(Ds, Qs)} are generated. After linear fitting of the training samples, a parameter control planning model is established: y = θ1*x + θ2, where θ1 and θ2 represent fitting coefficients, x represents the variable in the model that refers to the equipment layout concentration index, y represents the variable in the model that refers to the temperature, and * represents the multiplication sign. After receiving the temperature adjustment control instruction, the equipment layout concentration index of the non-faulty equipment in the current data room is analyzed to be q. The calculation method of q is the same as that of Dc. When calculating the equipment layout concentration index of the non-faulty equipment, the data of the faulty equipment is eliminated, and only the data of the non-faulty equipment is analyzed, that is, the location and quantity information of the faulty equipment are not within the analysis range. Substitute q into the model: Let x = q, and use the magnetic levitation phase change cooling technology to adjust the temperature in the current data room to θ1*q + θ2;

[0020] The equipment layout data in the cabinets of different computer rooms are collected through big data technology, and the equipment layout concentration index is analyzed. The equipment is divided into straight parallelepipeds and then the equipment layout concentration index is analyzed. The more equipment is covered in the hexahedron space and the closer the geometric centers of multiple hexahedrons are, that is, the smaller the area covering all center points, the more concentrated the equipment distribution in the computer room is, the more likely it is to have local overheating problems. At this time, the air-conditioning equipment needs to be adjusted to an appropriate temperature to effectively achieve heat dissipation. The linear regression algorithm in the AI ​​algorithm is used to establish a parameter control planning model. The temperature adjustment plan is made by establishing a parameter control planning model. When the current environment requires temperature adjustment, the data of equipment that has stopped running and does not generate heat is excluded, and the layout concentration index of the remaining equipment is analyzed. The index is then substituted into the model to obtain the current optimal temperature that needs to be adjusted. The temperature is adjusted to the optimal temperature using magnetic levitation phase change cooling technology. When necessary, the temperature of the air conditioner in the computer room is adaptively and automatically adjusted using magnetic levitation phase change cooling technology. On the basis of the energy-saving function of magnetic levitation phase change cooling technology itself, energy consumption is further reduced and energy saving is achieved.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] The present invention collects the equipment layout data in the cabinets of different computer rooms through big data technology, analyzes the equipment layout concentration index, divides the equipment into straight parallelepiped bodies and then analyzes the equipment layout concentration index, uses the linear regression algorithm in the AI ​​algorithm to establish a parameter control planning model, and plans the temperature adjustment by establishing the parameter control planning model. When the current environment requires temperature adjustment, the equipment data that stops running and does not generate heat is excluded, and the layout concentration index of the remaining equipment is analyzed, and then the index is substituted into the model to obtain the current optimal temperature that needs to be adjusted, and the temperature is adjusted to the optimal temperature by using the magnetic suspension phase change cooling technology. When necessary, the magnetic suspension phase change cooling technology is used to make adaptive automatic adjustment to the temperature of the air conditioner in the computer room, and the energy consumption is further reduced on the basis of the energy-saving effect of the magnetic suspension phase change cooling technology itself, and energy saving is achieved;

[0023] The purpose of setting the critical coefficient is to take into account the possibility of a small number of equipment experiencing continuous failures and the possibility of short-term equipment failures. The temperature adjustment is triggered only when the equipment data in the computer room reaches the critical conditions, which effectively avoids the above two situations, effectively increases the effect of temperature regulation, further reduces energy consumption, and further realizes energy saving of the refrigeration system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a structural schematic diagram of the magnetic suspension phase change cooling energy-saving control system based on AI algorithm of the present invention;

[0025] Figure 2 It is a flow chart of the magnetic suspension phase change cooling energy-saving control method based on AI algorithm of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Embodiment 1:

[0028] like Figure 1 As shown, this embodiment provides a magnetic suspension phase change cooling energy-saving control system based on an AI algorithm, and the system includes: an environment layout data acquisition module, a control critical condition setting module, a current environment monitoring module, and a control planning module;

[0029] The output end of the environment layout data acquisition module is connected to the input end of the control critical condition setting module and the control planning module, the output end of the control critical condition setting module is connected to the input end of the current environment monitoring module, and the output end of the current environment monitoring module is connected to the input end of the control planning module;

[0030] The environment layout data collection module collects equipment layout information, equipment operation history data, and parameter adjustment history data in the data room;

[0031] By controlling the critical condition setting module to analyze the historical data of equipment operation, set the critical trigger conditions for adjusting the temperature in the data center using magnetic suspension phase change cooling technology;

[0032] The current environment monitoring module monitors the operation of the equipment in the current environment in real time, and sends a temperature adjustment control instruction when the current environment data meets the critical trigger condition;

[0033] A parameter control planning model is established through the control planning module, and after receiving the temperature adjustment control instruction, the temperature in the data room is adjusted and controlled according to the model.

[0034] The environment layout data acquisition module includes a computer room equipment layout acquisition unit, an equipment operation data acquisition unit, and an adjustment data acquisition unit;

[0035] The equipment layout acquisition unit of the computer room is used to establish a three-dimensional spatial model of the data room, collect the spatial layout information of the equipment in the cabinet of the data room, divide the equipment placed in adjacent positions into the same right parallelepiped space, where the right parallelepiped refers to a cuboid or a cube, and the adjacent positions refer to the adjacent layers in a cabinet. For example, if two devices are placed on the first and second layers in a cabinet, the relative positions between the two devices belong to adjacent positions, and collect the number of devices contained in the divided different right parallelepiped spaces and the center point position information of the divided different right parallelepiped spaces;

[0036] The equipment operation data collection unit is used to collect the historical operation fault data of the equipment in the cabinet of the same data room, including the number of equipment that stopped running due to faults in the past and the duration of equipment fault maintenance. The fault maintenance duration refers to the time interval from the start of monitoring to the time when the equipment is repaired and can operate normally.

[0037] The adjustment data acquisition unit is used to collect the air conditioning temperature data in different data rooms controlled by magnetic suspension phase change cooling technology when no equipment failure has occurred in the past.

[0038] The control critical condition setting module includes an abnormal data analysis unit and a critical coefficient setting unit;

[0039] The input end of the abnormal data analysis unit is connected to the output end of the equipment operation data acquisition unit, and the output end of the abnormal data analysis unit is connected to the input end of the critical coefficient setting unit;

[0040] The abnormal data analysis unit is used to retrieve historical operation fault data and analyze the necessary coefficients for adjusting the air conditioning temperature in the data room when equipment failures occurred at different times in the past;

[0041] The critical coefficient setting unit is used to set a critical coefficient for triggering air conditioning temperature adjustment according to the necessary coefficient obtained through analysis.

[0042] The current environment monitoring module includes an environment data monitoring unit, a data comparison unit, and a parameter control trigger selection unit;

[0043] The output ends of the environmental data monitoring unit and the critical coefficient setting unit are connected to the input end of the data comparison unit, and the output end of the data comparison unit is connected to the input end of the parameter control trigger selection unit;

[0044] The environmental data monitoring unit is used to monitor the equipment operation status in the current data center in real time, count the number of equipment failures and estimate the duration of equipment failures when equipment failures occur, and analyze the necessary coefficients for adjusting the temperature in the room using magnetic suspension phase change cooling technology;

[0045] The data comparison unit is used to compare the necessary coefficient and the critical coefficient for adjusting the temperature in the computer room by using the magnetic suspension phase change cooling technology;

[0046] The parameter control trigger selection unit is used to trigger the control and adjustment of the temperature in the current equipment room when the current necessary coefficient is greater than or equal to the critical coefficient.

[0047] The control planning module includes a control planning model building unit, an environmental data input unit and a parameter adjustment control unit;

[0048] The input end of the control planning model establishment unit is connected to the output ends of the parameter control trigger selection unit, the computer room equipment layout acquisition unit and the adjustment data acquisition unit, the output end of the control planning model establishment unit is connected to the input end of the environment data input unit, and the output end of the environment data input unit is connected to the input end of the parameter adjustment control unit;

[0049] The control planning model building unit is used to generate training samples based on the collected equipment space layout information, and establish a parameter control planning model after fitting the training samples;

[0050] The environmental data input unit is used to input the spatial layout data of the non-faulty equipment in the current data room into the parameter control planning model after receiving the trigger instruction to control and adjust the temperature in the current room;

[0051] The parameter adjustment control unit is used to adjust and control the temperature in the data room using magnetic levitation phase change cooling technology based on the data output by the model.

[0052] Embodiment 2:

[0053] like Figure 2 As shown, this embodiment provides a magnetic suspension phase change cooling energy-saving control method based on an AI algorithm, which is implemented based on the control system in the embodiment and specifically includes the following steps:

[0054] Z100: Collect equipment layout information, equipment operation history data and parameter adjustment history data in the data room: Establish a three-dimensional spatial model of the data room, collect equipment space layout information in the cabinet of the data room, divide the equipment placed in adjacent positions into the same right parallelepiped space, collect a random data room divided into n right parallelepipeds, and the number of equipment contained in the divided n right parallelepiped spaces is {E1, E2, ...Er, ...En}. Collect the center point position coordinates of the n right parallelepipeds, and obtain the minimum circle radius R that can cover the center points of the n right parallelepipeds. The minimum circle radius R can be obtained using the minimum circle coverage derivative algorithm. Collect the historical operation fault data of the equipment in the cabinet of the same data room, including the number of equipment that stopped running due to simultaneous failures in the past and the duration of equipment failures. Collect the air conditioning temperature data in different data rooms controlled by magnetic suspension phase change cooling technology when there was no equipment failure in the past;

[0055] Z101: Analyze the historical data of equipment operation and set the critical triggering conditions for adjusting the temperature in the data room by using magnetic suspension phase change cooling technology: For a random data room: the number of equipment that stopped running due to failures in the past is obtained as A = {A1, A2, ...Ai, ...Am}, m represents the number of equipment failures in the cabinet of a random data room, the total number of equipment in the cabinet of a random data room is counted as M, and a total of k equipment failures at a random time are collected, k = Ai, the failure maintenance duration of k faulty equipment is T = {T1, T2, ...Tj, ...Tk}, j represents the jth faulty equipment, according to Wi = (Ai) / M+[∑ k j＝1(Tj)] / k Analyze the necessary coefficient Wi for adjusting the air-conditioning temperature in the data computer room when a device failure occurs randomly once in the past, and obtain the set of necessary coefficients for adjusting the air-conditioning temperature in the data computer room when a device failure occurs m times in the past as {W1, W2,... Wi,... Wm}. Sort the necessary coefficients in descending order, and randomly divide the sorted necessary coefficients into f groups. All the necessary coefficients in the previous group are greater than those in the subsequent group. Obtain the set of average values of the necessary coefficients in each group in a randomly obtained grouping result as {w1, w2,... we,... wf}, where we represents the average value of the necessary coefficients in the e-th group. According to the formula L = [(∑ f e＝1 (we - ∑ f e＝1 (we) / f)) 2 / f] 1 / 2 Calculate the reference degree L of a randomly obtained grouping result for setting the critical coefficient, compare the reference degrees of different grouping results for setting the critical coefficient, select the grouping result with the highest reference degree, obtain the necessary coefficients in the first group in the grouping result with the highest reference degree, calculate the average value of the necessary coefficients in the first group, and set the obtained average value as the critical coefficient for triggering air-conditioning temperature adjustment, and obtain the critical coefficient U;

[0056] Z102: Real-time monitor the operation of devices in the current environment, and send a temperature adjustment control instruction when it is detected that the current environmental data meets the critical trigger condition: Monitor that the number of devices that have failed and stopped running simultaneously in the data computer room is b, obtain the failure type of a randomly selected failed device currently, and retrieve the set of previous failure duration of devices that have had corresponding type failures in the past as {t1, t2,... tv,... tg}, where g represents the number of devices that have had corresponding type failures in the past. Estimate the failure duration Fa of a randomly selected failed device currently, Fa = ∑ g v＝1 (tv) / g, where a represents the a-th failed device currently, and v represents the v-th device that has had corresponding type failures in the past. Estimate the set of failure durations of devices that have failed and stopped running simultaneously currently in the same way as {F1, F2,... Fa,... Fb}, and analyze the necessary coefficient W ’ , W ’ = b / M + [∑ b a＝1 (Fa)] / b. Compare W ’ and U. If W ’ ≥ U, send a temperature adjustment control instruction; if W ’ < U, do not send a temperature adjustment control instruction;

[0057] For example, if the critical coefficient is set to U = 3.67, the necessary coefficient W for adjusting the temperature in the computer room using the current magnetic suspension phase change cooling technology is obtained. ’ =4.22,W ’ >U, sends temperature adjustment control instructions, and uses magnetic suspension phase change cooling technology to adjust the temperature in the current data center;

[0058] Z103: Establish a parameter control planning model, and after receiving the temperature adjustment control instruction, adjust the temperature in the data center according to the model: according to Dc = ∑ n r＝1 (Er) / n+1 / (π*R 2 ) is analyzed to obtain the equipment layout concentration index Dc of a random data room, r represents the rth straight parallelepiped divided in a random data room, c represents the cth data room, and the equipment layout concentration index set of different data rooms is analyzed in the same way as {D1, D2, ...Dc, ...Ds}, s represents the total number of data rooms, and the air-conditioning temperature set controlled by the magnetic levitation phase change cooling technology when there is no equipment failure in the s data rooms is retrieved as Q = {Q1, Q2, ...Qc, ...Qs}. When there is no equipment failure, the room is in a constant temperature state, that is, in a constant temperature state, and the training samples {(D1, Q1), (D2, Q2), ...(Dc, Qc), ...(Ds, Qs)} are generated. After linear fitting of the training samples, a parameter control planning model is established: y=θ1*x+θ2, where θ1 and θ2 represent fitting coefficients, x represents the variable representing the equipment layout concentration index in the model, y represents the variable representing the temperature in the model, and * represents the multiplication sign. After receiving the temperature adjustment control instruction, the equipment layout concentration index of the non-faulty equipment in the current data room is analyzed to be q. The calculation method of q is the same as that of Dc. When calculating the equipment layout concentration index of the non-faulty equipment, the faulty equipment data is eliminated, and only the data of the non-faulty equipment is analyzed, that is, the location and quantity information of the faulty equipment are not within the analysis range. Substitute q into the model: Let x=q, and use the magnetic levitation phase change cooling technology to adjust the temperature in the current data room to θ1*q+θ2.

[0059] It will be apparent to those skilled in the art that the 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 the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. Magnetic suspension phase change cooling energy-saving control system based on AI algorithm, characterized by: It includes environment layout data acquisition module, control critical condition setting module, current environment monitoring module and control planning module; The environment layout data collection module collects equipment layout information, equipment operation history data and parameter adjustment history data in the data room; Analyze the equipment operation history data through the control critical condition setting module, and set the critical trigger condition for adjusting the temperature in the data room by using the magnetic suspension phase change cooling technology; The current environment monitoring module monitors the operation of the equipment in the current environment in real time, and sends a temperature adjustment control instruction when it is detected that the current environment data meets the critical trigger condition; A parameter control planning model is established through the control planning module, and after receiving the temperature adjustment control instruction, the temperature in the data room is adjusted and controlled according to the model.

2. The magnetic suspension phase change cooling energy-saving control system based on AI algorithm according to claim 1 is characterized by: The environment layout data acquisition module includes a computer room equipment layout acquisition unit, an equipment operation data acquisition unit and an adjustment data acquisition unit; The equipment layout acquisition unit of the computer room is used to establish a three-dimensional spatial model of the data room, collect the spatial layout information of the equipment in the cabinet of the data room, divide the equipment placed in adjacent positions into the same right parallelepiped space, and collect the number of equipment contained in different right parallelepiped spaces divided and the center point position information of different right parallelepiped spaces divided; The equipment operation data collection unit is used to collect the historical operation fault data of the equipment in the cabinet of the same data room, including the number of equipment that stopped running due to faults in the past and the maintenance time of the equipment faults, wherein the maintenance time of the faults refers to the time interval from the start of monitoring to the time when the equipment is repaired and can operate normally; The adjustment data acquisition unit is used to collect air conditioning temperature data in different data rooms that were controlled by magnetic suspension phase change cooling technology when no equipment failure occurred in the past.

3. The magnetic suspension phase change cooling energy-saving control system based on AI algorithm according to claim 2 is characterized by: The control critical condition setting module includes an abnormal data analysis unit and a critical coefficient setting unit; The abnormal data analysis unit is used to retrieve historical operation fault data and analyze the necessary coefficients for adjusting the air conditioning temperature in the data room when equipment failures occurred at different times in the past; The critical coefficient setting unit is used to set a critical coefficient for triggering air conditioning temperature adjustment according to the necessary coefficient obtained by analysis.

4. The magnetic suspension phase change cooling energy-saving control system based on AI algorithm according to claim 3 is characterized by: The current environment monitoring module includes an environment data monitoring unit, a data comparison unit and a parameter control trigger selection unit; The environmental data monitoring unit is used to monitor the operation status of the equipment in the current data room in real time, count the number of equipment failures and estimate the duration of equipment failures when equipment failures occur, and analyze the necessary coefficients for adjusting the temperature in the room using the magnetic suspension phase change cooling technology; The data comparison unit is used to compare the necessary coefficient and the critical coefficient for adjusting the temperature in the computer room using the magnetic suspension phase change cooling technology; The parameter control trigger selection unit is used to trigger the control and adjustment of the temperature in the current computer room when the current necessary coefficient is greater than or equal to the critical coefficient.

5. The magnetic suspension phase change cooling energy-saving control system based on AI algorithm according to claim 4 is characterized by: The control planning module includes a control planning model building unit, an environmental data input unit and a parameter adjustment control unit; The control planning model building unit is used to generate training samples based on the collected equipment space layout information, and to establish a parameter control planning model after fitting the training samples; The environmental data input unit is used to input the spatial layout data of the non-faulty equipment in the current data room into the parameter control planning model after receiving the trigger instruction to control and adjust the temperature in the current data room; The parameter adjustment control unit is used to adjust and control the temperature in the data room using magnetic suspension phase change cooling technology according to the data output by the model.

6. The magnetic suspension phase change cooling energy-saving control method based on AI algorithm is characterized by: The following steps are involved: Z100: collects equipment layout information, equipment operation history data and parameter adjustment history data in the data room; Z101: Analyze the historical data of equipment operation and set the critical trigger conditions for adjusting the temperature in the data center using magnetic suspension phase change cooling technology; Z102: Monitor the operation of equipment in the current environment in real time, and send temperature adjustment control instructions when the current environmental data meets the critical trigger conditions; Z103: Establish a parameter control planning model, and after receiving the temperature adjustment control instruction, adjust and control the temperature in the data center according to the model.

7. The magnetic suspension phase change cooling energy-saving control method based on AI algorithm according to claim 6 is characterized in that: In step Z100: establish a three-dimensional spatial model of the data room, collect the spatial layout information of the equipment in the cabinet of the data room, divide the equipment placed in adjacent positions into the same right parallelepiped space, collect a total of n right parallelepipeds divided in a random data room, and the number of equipment contained in the n right parallelepiped spaces is {E1, E2, ...Er, ...En}. Collect the center point position coordinates of the n right parallelepipeds, and obtain the minimum circle radius R that can cover the center points of the n right parallelepipeds. Collect the historical operation failure data of the equipment in the cabinet of the same data room, including the number of equipment that stopped running due to simultaneous failures in the past and the duration of the equipment failures. Collect the air-conditioning temperature data in different data rooms controlled by magnetic levitation phase change cooling technology when no equipment failures occurred in the past.

8. The magnetic suspension phase change cooling energy-saving control method based on AI algorithm according to claim 7 is characterized in that: In step Z101: for a random data room: the number of devices that have failed at the same time and stopped running in the past is obtained as A = {A1, A2, ... Ai, ... Am}, m represents the number of times the equipment in the cabinet of a random data room fails, the total number of devices in the cabinet of a random data room is counted as M, and a total of k devices fail at the same time at a random time, k = Ai, the failure maintenance time of k faulty devices is T = {T1, T2, ... Tj, ... Tk}, j represents the jth faulty device, according to Wi = (Ai) / M + [∑ k j＝1 (Tj) / k analyzes the necessary coefficients Wi for adjusting the air conditioning temperature in the data room when a random equipment failure occurs in the past, and obtains the necessary coefficients set for adjusting the air conditioning temperature in the data room when the equipment failure occurs m times in the past as {W1, W2, ...Wi, ...Wm}. The necessary coefficients are sorted in order from large to small, and the sorted necessary coefficients are randomly divided into f groups. All the necessary coefficients in the first group are greater than those in the second group. The necessary coefficient mean set in each group in a random grouping result is obtained as {w1, w2, ...we, ...wf}, where we represents the necessary coefficient mean in the eth group. According to the formula L=[(∑ f e＝1 (we-∑ f e＝1 (we) / f)) 2 / f] 1 / 2 Calculate the reference degree L of a random grouping result for the set critical coefficient, compare the reference degrees of different grouping results for the set critical coefficient, select the grouping result with the highest reference degree, obtain the necessary coefficients in the first group among the grouping results with the highest reference degree, calculate the average value of the necessary coefficients of the first group, set the average value as the critical coefficient for triggering air conditioning temperature adjustment, and obtain the critical coefficient U.

9. The magnetic suspension phase change cooling energy-saving control method based on AI algorithm according to claim 8 is characterized in that: In step Z102: The number of devices that currently have faults and have stopped running in the data room is monitored as b. The fault type of a currently randomly selected faulty device is obtained. The set of past fault duration times of devices that had corresponding type faults in the past is retrieved as {t1, t2,... tv,... tg}, where g represents the number of devices that had corresponding type faults in the past. The estimated fault duration time of a currently randomly selected faulty device is Fa, Fa = ∑ g v＝1 (tv) / g, where a represents the a-th faulty device currently, and v represents the v-th device that had a corresponding type fault in the past. The set of fault duration times of devices that currently have faults and have stopped running is estimated in the same way as {F1, F2,... Fa,... Fb}. Analyze the necessary coefficient W for currently using the magnetic levitation phase change cooling technology to adjust the temperature in the computer room ’ , W ’ = b / M + [∑ b a＝1 (Fa)] / b. Compare W ’ with U. If W ’ ≥ U, send a temperature adjustment control instruction; if W ’ < U, do not send a temperature adjustment control instruction.

10. The magnetic suspension phase change cooling energy-saving control method based on AI algorithm according to claim 9 is characterized in that: In step Z103: according to Dc=∑ n r＝1 (Er) / n+1 / (π*R 2 ) is analyzed to obtain the equipment layout concentration index Dc of a random data room, r represents the rth straight parallelepiped divided in a random data room, c represents the cth data room, and the equipment layout concentration index set of different data rooms is analyzed in the same way as {D1, D2, ...Dc, ...Ds}, s represents the total number of data rooms, and the air conditioning temperature set controlled by the magnetic levitation phase change cooling technology when there is no equipment failure in the s data rooms is Q = {Q1, Q2, ...Qc, ...Qs}, generating training samples {(D1, Q1), (D 2, Q2), ... (Dc, Qc), ... (Ds, Qs)}, a parameter control planning model is established after linear fitting of the training samples: y = θ1*x + θ2, wherein θ1 and θ2 represent fitting coefficients, x represents the variable in the model that refers to the equipment layout concentration index, and y represents the variable in the model that refers to the temperature. After receiving the temperature adjustment control instruction, the equipment layout concentration index of the non-faulty equipment in the current data room is analyzed to be q. Substitute q into the model: let x = q, and use the magnetic levitation phase change cooling technology to adjust the temperature in the current data room to θ1*q + θ2.

Images