Intelligent extensible integrated circuit water cooling liquid circulation heat dissipation system and control method

Abstract

The invention discloses an intelligent extensible integrated circuit water cooling liquid circulation heat dissipation system and a control method, and relates to the technical field of integrated circuit temperature control, and the control method comprises the steps: S1, numbering all integrated water cooling blocks according to 1-N, deploying a conductivity sensor at a backwater collecting pipe, and monitoring the temperature insulation performance attenuation coefficient of cooling liquid flowing through all the integrated water cooling blocks in real time; s2, judging the service life of the cooling liquid through AI prediction and decision based on an LSTM neural network; s3, the alarm unit carries out processing by judging the alarm level; s4, heat dissipation is assisted through the phase change material; and a metal-organic framework material layer is filled between the hot end of the refrigeration sheet and the heat dissipation fins to absorb transient thermal shock. The temperature of each semiconductor is monitored in real time by utilizing an intelligent predictive control technology while the semiconductor is cooled by utilizing water cooling circulation, so that accurate temperature control is realized through AI, and the temperature of each water cooling sheet is kept within a certain range.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of integrated circuit temperature control, in particular to an intelligent and expandable integrated circuit water cooling liquid circulation heat dissipation system and control method. BACKGROUND

[0002] With the improvement of the scale of component integration, integrated circuits are increasingly developing towards large-scale, super-large and extra-large integrated circuit groups, and the heat power generated per unit volume is gradually increasing, while the board heat dissipation area remains unchanged, resulting in that the heat dissipation per unit area cannot meet the requirements. At present, the mode of metal heat dissipation fins and forced air cooling is still used for large-scale integrated circuits, which cannot meet the needs of the development of integrated circuits.

[0003] In the inventor's previous application, the application for an invention patent for an integrated circuit water cooling liquid circulation heat dissipation architecture (publication number: CN110211936A, publication date: September 6, 2019, and name: an integrated circuit water cooling liquid circulation heat dissipation architecture), which includes a precise heat exchange unit and a refrigeration heat dissipation unit, the inside of the precise heat exchange unit includes a plurality of integrated circuit water cooling blocks, a return water manifold and an outlet water manifold, a plurality of the integrated circuit water cooling blocks are located on the left side of the outlet water manifold, the return water manifold is located below the outlet water manifold, and the inside of the precise heat exchange unit includes an air-cooled water discharge, a water pump and a refrigeration fin refrigeration unit. The present application uses cooling water with an insulation resistance higher than air as the cooling liquid in the water cooling circulation of the architecture, compared with the existing refrigeration fin water cooling technology, which is not currently applied to the heat dissipation of integrated circuits, the architecture uses cooling water with insulation capacity to replace ordinary cooling water, realizes the application of the technology in the integrated circuit architecture, and avoids the possibility of short circuit of the circuit board caused by leakage of the cooling water.

[0004] In the prior art including the above-mentioned patent, the water temperature detection device acquires the temperature of the return water manifold in real time, the temperature controller controls the on-off of the power supply of the refrigeration fin according to the return water temperature, and the control of the circulating water temperature is realized, but it is difficult to monitor the temperature of each water cooling block, so that the heat dissipation of each water cooling block is uneven, and the temperature of the integrated circuit is difficult to accurately adjust. SUMMARY

[0005] The purpose of the present application is to provide an intelligent and expandable integrated circuit water cooling liquid circulation heat dissipation system and control method to solve the above-mentioned deficiencies in the prior art.

[0006] In order to achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0007] The control method of the intelligent and expandable integrated circuit water cooling liquid circulation heat dissipation system, the method comprising the following steps:

[0008] S1: Number each integrated water cooling block from 1 to N, and deploy an electrical conductivity sensor at the return water manifold to monitor the temperature insulation performance decay coefficient of the cooling liquid flowing through each integrated water cooling block in real time;

[0009] S2: AI prediction and decision-making based on LSTM neural network to determine the service life of the cooling liquid;

[0010] S3: The alarm unit processes by judging the alarm level;

[0011] S4: Phase change material assisted heat dissipation; fill the metal-organic framework material layer between the hot end of the refrigeration sheet and the heat dissipation fin to absorb transient thermal shock;

[0012] S5: Waste heat recovery, convert the waste heat at the hot end of the refrigeration sheet into electrical energy to power the control unit;

[0013] S6: Dynamic topology adjustment, automatically switch between series or parallel mode according to the temperature data of each integrated water cooling block;

[0014] S7: Collaborative workflow through the signal interaction system of each module.

[0015] As a further description of the above technical solution, S1 includes the following steps:

[0016] S11: The cooling liquid flows through the detection cavity to each electrical conductivity sensor;

[0017] S12: Each electrical conductivity sensor outputs a 4-20mA electrical signal to the signal conditioning circuit with a sampling frequency of 1-2Hz;

[0018] S13: The conditioning circuit converts the electrical signal into a digital quantity and transmits it to the ADS module;

[0019] S14: The ADS module uploads data packets to the edge computing unit every 500ms;

[0020] S15: The edge computing unit calculates the temperature insulation performance decay coefficient of each integrated water cooling block.

[0021] The signal transmission protocol of the electrical signal is Modbus RTU over RS-485;

[0022] The electrical conductivity-temperature compensation formula of the temperature insulation performance decay coefficient of each integrated water cooling block is σcal=σmeas×[1+0.02(T−25)].

[0023] As a further description of the above technical solution, the AI prediction and decision-making in S2 includes the following steps:

[0024] S21: Real-time monitoring of data from each electrical conductivity sensor through AI;

[0025] S22: Preprocessing the data of each conductivity sensor by AI;

[0026] S23: Continuously reading real-time data within 30 minutes and transmitting to the LSTM model for inference and judgment.

[0027] As a further description of the above technical solution, S3 includes the following steps:

[0028] S31: When the conductivity is 2-5 μS / cm, the alarm level is level one, and a short message is sent to notify the maintenance personnel to handle within 72 hours;

[0029] S32: When the conductivity is >5 μS / cm, the alarm level is level two, the alarm unit records the fault code F005 and uploads it to the cloud, and each refrigeration fin automatically runs at a lower frequency. The cloud pushes a work order through a short message to notify the maintenance personnel to handle urgently within 24 hours.

[0030] As a further description of the above technical solution, the heat dissipation assisted by the phase change material in S4 includes S41 heat absorption stage and S42 heat release stage;

[0031] S41 heat absorption stage includes the following steps:

[0032] S411: The hot end of the refrigeration fin transfers heat to the cold end of the refrigeration fin;

[0033] S412: The cold end of the refrigeration fin transfers heat to the thermal interface material;

[0034] S413: The thermal interface material transfers heat to the phase change material for latent heat absorption, keeping the temperature at 43-47℃;

[0035] S414: The thermal interface material transfers the heat after latent heat absorption to the metal-organic framework material layer for absorption.

[0036] As a further description of the above technical solution, when the ambient temperature is <40℃, the metal-organic framework material layer starts to crystallize and release heat, and the heat is transferred to the heat dissipation fin through the TEG module. The heat dissipation fin performs forced convection heat dissipation.

[0037] As a further description of the above technical solution, the S5 waste heat recycling, when the temperature difference ΔT between the hot end and the cold end of each refrigeration fin is ≥60℃, the heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers the heat to the energy storage super capacitor for storage to supply power to the control unit.

[0038] As a further description of the above technical solutions, the S6 dynamic topology adjustment comprises S61 mode switching judgment and S62 dynamic threshold calculation steps.

[0039] S611: Real-time monitoring of the cooling liquid temperature flowing through each integrated water cooling block by the conductivity sensor;

[0040] S612: Determining the temperature difference ΔT = max(T n )-min(T n ) of the hot end and the cold end of each integrated water cooling block by the LSTM; n n

[0041] S613: If ΔT = max(T n )-min(T n ) ≤ 5~7℃, maintain the series mode, if the single-point temperature of the hot end of each integrated water cooling block > 85℃, start the parallel mode, PID control the opening of the electric proportional valve to make the water pump speed value 120%;

[0042] S614: If the temperature difference ΔT = max(T n )-min(T n ) of the hot end and the cold end of each integrated water cooling block ≥ 10℃, start the parallel mode, PID control the opening of the electric proportional valve to make the water pump speed value 120%;

[0043] The S62 dynamic threshold calculation Threshold = 10−0.5×(Qcurrent−1) (unit: ℃), wherein Q is the current flow (L / min).

[0044] As a further description of the above technical solutions, the S7 comprises the following steps:

[0045] S71: Real-time monitoring of the cooling liquid temperature flowing through each integrated water cooling block by the conductivity sensor and the temperature sensor;

[0046] S72: When the temperature sensor detects that the temperature of each integrated water cooling block is 75~78℃, the controller switches the parallel mode PID control of the electric proportional valve opening, the heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers the heat to the energy storage super capacitor for storage to supply power to the control unit;

[0047] S73: When the temperature sensor detects that the temperature of each integrated water cooling block is 78℃, the controller switches the parallel mode PID control of the electric proportional valve opening, switches the parallel mode, PID control of the electric proportional valve opening, triggers the sound-light alarm and runs at a reduced frequency;

[0048] S74: When the temperature sensor detects that the temperature of each integrated water cooling block is 43~47℃, maintain the series mode, and the ADS module samples.

[0049] A smart and scalable integrated circuit water-cooled liquid circulation heat dissipation system includes a precision heat exchange unit, a cooling heat dissipation unit, a control unit, an insulating coolant circulation module, and a dynamic topology control module.

[0050] As a further description of the above technical solution, the precision heat exchange unit includes multiple integrated water-cooled blocks and multiple electric proportional valves. A metal-organic framework layer is filled between the hot end of each integrated water-cooled block and the heat dissipation fins. An Al2O3 nano-coating is sprayed on the hot end of each integrated water-cooled block. The cold end of each integrated water-cooled block is reinforced by a heat spreader. A thermal interface material layer is attached to the cold end of each water-cooled block. A metal-organic framework material layer is attached to the thermal interface material layer.

[0051] As a further description of the above technical solution, the cooling and heat dissipation unit includes multiple cooling chips, model TEC-12705, and each cooling chip is coated with a paraffin phase change layer.

[0052] As a further description of the above technical solution, the control unit includes a temperature sensor, a flow sensor, a conductivity sensor, an edge computing module, and an MPPT controller, wherein the edge computing module is an STM32H743.

[0053] As a further description of the above technical solution, the insulating coolant circulation module includes an outlet manifold, a return manifold, a fixed pipeline, an air-cooled water drain, and a water pump. The insulating coolant is a fluorinated liquid with an insulation strength ≥ 45 kV / mm.

[0054] As a further description of the above technical solution, the dynamic topology control module includes an ADS module and a TEG module.

[0055] In the above technical solution, the present invention provides an intelligent and scalable integrated circuit water-cooled liquid circulation heat dissipation method. It uses water cooling circulation to dissipate heat from semiconductors while using intelligent predictive control technology to monitor the temperature of each semiconductor in real time. Then, it uses AI to achieve precise temperature control, so that the temperature of each water-cooled chip is kept within a certain range.

[0056] It should be understood that the foregoing general description and the following detailed description are exemplary and illustrative only, and are not intended to limit this disclosure.

[0057] This application provides an overview of various implementations or examples of the technology described in this disclosure, and is not a full disclosure of the entire scope or all features of the disclosed technology. Attached Figure Description

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed in the embodiments will be briefly introduced as follows. Obviously, the accompanying drawings described below only represent some embodiments of the present application, and all other drawings obtained by those of ordinary skill in the art based on the drawings without any creative effort belong to the protection scope of the present application.

[0059] Figure 1 The integrated circuit water cooling liquid circulation heat dissipation system structure schematic diagram provided for the embodiment of the present application is shown in the figure.

[0060] Figure 2 The sensor data acquisition stage structure schematic diagram provided for the embodiment of the present application is shown in the figure.

[0061] Figure 3 The AI prediction and decision stage flow chart provided for the embodiment of the present application is shown in the figure.

[0062] Figure 4 The alarm system response flow chart provided for the embodiment of the present application is shown in the figure.

[0063] Figure 5 The mode switching determination structure schematic diagram provided for the embodiment of the present application is shown in the figure.

[0064] Figure 6 The module signal interaction schematic diagram provided for the embodiment of the present application is shown in the figure.

[0065] Figure 7 The fluorinated liquid parameter schematic diagram provided for the embodiment of the present application is shown in the figure.

[0066] Figure 8 The key parameter coordination schematic diagram provided for the embodiment of the present application is shown in the figure.

[0067] Figure 9 The prior art and the present application data comparison schematic diagram provided for the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure clearer, the technical solutions of the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings of the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by those of ordinary skill in the art without any creative effort belong to the protection scope of the present disclosure.

[0069] Embodiment 1

[0070] A control method of an intelligent expandable integrated circuit water cooling liquid circulation heat dissipation system, the method comprising the following steps:

[0071] S1: Number each integrated water cooling block from 1 to N, and deploy an electrical conductivity sensor at the return water manifold to monitor the temperature insulation performance decay coefficient of the cooling liquid flowing through each integrated water cooling block in real time; the S1 includes the following steps:

[0072] S11: The cooling liquid flows through the detection cavity to each electrical conductivity sensor;

[0073] S12: Each electrical conductivity sensor outputs a 4mA electrical signal to the signal conditioning circuit at a sampling frequency of 1Hz;

[0074] S13: The conditioning circuit converts the electrical signal into a digital quantity and transmits it to the ADS module;

[0075] S14: The ADS module uploads data packets to the edge computing unit every 500ms;

[0076] S15: The edge computing unit calculates the temperature insulation performance decay coefficient of each integrated water cooling block;

[0077] The signal transmission protocol of the electrical signal is Modbus RTU over RS-485;

[0078] The electrical conductivity-temperature compensation formula of the temperature insulation performance decay coefficient of each integrated water cooling block is σcal=σmeas×[1+0.02(T−25)].

[0079] S2: AI prediction and decision-making based on LSTM neural network to determine the service life of the cooling liquid;

[0080] The AI prediction and decision-making in the S2 include the following steps:

[0081] S21: Real-time monitoring of data from each electrical conductivity sensor by AI;

[0082] S22: Preprocessing of data from each electrical conductivity sensor by AI;

[0083] S23: Continuously reading real-time data within 30 minutes and transmitting it to the LSTM model for inference and judgment.

[0084] S3: The alarm unit processes by judging the alarm level, and does not alarm when the electrical conductivity is ≤2μS / cm.

[0085] S4: Assisted heat dissipation by phase change material; filling a metal-organic framework material layer between the hot end of the refrigeration sheet and the heat dissipation fin to absorb transient thermal shock; the composition of the metal-organic framework material layer (MOF-5) is ZnO(BDC), (BDC = terephthalic acid), which is a three-dimensional porous crystal structure with a specific surface area of up to 3000m 2 / g

[0086] The S4 heat dissipation assisted by the phase change material includes a heat absorption stage S41 and a heat release stage S42;

[0087] The S41 heat absorption stage includes the following steps:

[0088] S411: The hot end of the refrigeration fin transfers heat to the cold end of the refrigeration fin;

[0089] S412: The cold end of the refrigeration fin transfers heat to the thermal interface material;

[0090] S413: The thermal interface material transfers heat to the phase change material for latent heat absorption, so that the temperature is maintained at 43°C;

[0091] S414: The thermal interface material transfers the heat after latent heat absorption to the metal-organic framework material layer for absorption, wherein the thermal interface material is thermal conductive silicone grease or graphene, and the phase change material is filled with 70% ZnO(BDC) and 30% graphene.

[0092] The S42 heat release stage, when the ambient temperature <40℃, the metal-organic framework material layer starts to crystallize and releases heat to transfer heat to the heat dissipation fins through the TEG module, and the heat dissipation fins perform forced convection heat dissipation.

[0093] S5: Waste heat recovery, converting the waste heat of the hot end of the refrigeration fin into electrical energy to power the control unit;

[0094] In the S5 waste heat recovery, when the temperature difference ΔT between each refrigeration fin hot end and its cold end ≥60℃, heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers heat to the energy storage super capacitor for storage to power the control unit.

[0095] S6: Dynamic topology adjustment, automatically switching between series or parallel modes according to temperature data of each integrated water cooling block;

[0096] The S6 dynamic topology adjustment includes the following steps of S61 mode switching judgment and S62 dynamic threshold calculation:

[0097] S611: Real-time monitoring of the temperature of the cooling liquid flowing through each integrated water cooling block by an electrical conductivity sensor;

[0098] S612: Determining the temperature difference ΔT between the hot end and the cold end of each integrated water cooling block by LSTM, wherein ΔT=max(T n )-min(T n )≥10℃;

[0099] If ΔT=max(T n )-min(T nThe temperature of each integrated water-cooled block is ≤5℃, and the single-point temperature of the hot end of each integrated water-cooled block is ≤75℃, and the series mode is maintained;

[0100] The S62 dynamic threshold value is calculated as Threshold=10−0.5×(Qcurrent−1)(unit: ℃), wherein Q is the current flow (L / min).

[0101] S7: The modules work cooperatively through the signal interaction system.

[0102] The S7 includes the following steps:

[0103] S71: The temperature of the cooling liquid flowing through each integrated water-cooled block is monitored in real time by the conductivity sensor and the temperature sensor.

[0104] S72: When the temperature sensor detects that the temperature of each integrated water-cooled block is 75℃, the controller switches to the parallel mode PID control electric proportional valve opening, the heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers the heat to the energy storage super capacitor for storage as a control unit power supply.

[0105] System-level cooperative working process timing example:

[0106] 00:00:00 [Monitoring starts]

[0107] ├- Conductivity sensor: sampling every 500ms (CS1500, ±0.1 μS / cm accuracy)

[0108] ├- Temperature sensor: PT100, 1Hz refresh rate

[0109] ├- Data synchronization: summarized to STM32H743 controller through RS-485 bus

[0110] ├- Value: 2.0 μS / cm

[0111] └- System status: normal (green indicator light)

[0112] 00:00:05 [Emergency response]

[0113] ├- Trigger condition: water-cooled sheet temperature ≤75℃.

[0114] ├- Action: maintain series

[0115] └- Water pump: pressure is maintained at ≤0.1 MPa

[0116] 00:00:10 [Energy management]

[0117] ├- TEG output voltage detection: MPPT algorithm locks 4.2V maximum power point

[0118] ├─ Power Switch: Turn off external power supply, sensor group powered by TEG

[0119] └─ Super Capacitor: Store excess energy (3.7V / 10F)

[0120] 00:00:15 [Phase Change Regulation]

[0121] ├─ MOF-5 Temperature Reach Phase Change Point: Absorb transient heat shock of water cooling fin (180J / g)

[0122] ├─ Temperature Stabilization: Control at 43℃ (Through PID regulation of fin power)

[0123] └─ Heat Flow Diversion: Waste heat converted to electrical energy by TEG (Efficiency ≥ 3%)

[0124] 00:00:30 [Mode Recovery]

[0125] ├─ ΔT≤5℃ Maintain Series State

[0126] ├─ Valve Reset: Maintain Series Mode (Flow reduced to 1.0L / min)

[0127] └─ System Self-Check: Confirm all sensor data back to normal range

[0128] An intelligent and scalable integrated circuit water cooling liquid circulation cooling system, comprising a precise heat exchange unit, a refrigeration cooling unit, a control unit, an insulation cooling liquid circulation module, and a dynamic topology control module.

[0129] The precise heat exchange unit comprises a plurality of integrated water cooling blocks and a plurality of electric proportional valves, the hot end of each integrated water cooling block is filled with a metal organic framework layer between the heat dissipation fins, the hot end of each integrated water cooling block is sprayed with an Al2O3 nano coating, the cold end of each integrated water cooling block is reinforced with a uniform temperature plate, the cold end of each water cooling block is attached with a thermal interface material layer, and the thermal interface material layer is attached with a metal-organic framework material layer.

[0130] The refrigeration cooling unit comprises a plurality of refrigeration fins, the model of each refrigeration fin is TEC-12705, each refrigeration fin is attached with a paraffin phase change layer, and the melting point of the paraffin phase change layer is 45℃.

[0131] The control unit comprises a temperature sensor, a flow sensor, an electrical conductivity sensor, an edge computing module, and an MPPT controller, the model of the edge computing module is STM32H743.

[0132] The insulation cooling liquid circulating module comprises a water outlet manifold, a water return manifold, a fixed pipeline, an air-cooled water discharge and a water pump, the insulation cooling liquid is a fluorinated liquid, the model is FS270, and the insulation strength is greater than or equal to 45 kV / mm.

[0133] The dynamic topology control module comprises an ADS module and a TEG module.

[0134] Embodiment 2

[0135] A control method of an intelligent and scalable integrated circuit water cooling liquid circulating heat dissipation system, the method comprising the following steps:

[0136] S1: number each integrated water cooling block as 1-N, and deploy an electrical conductivity sensor in the water return manifold to monitor the temperature insulation performance attenuation coefficient of the cooling liquid flowing through each integrated water cooling block in real time; the S1 comprises the following steps:

[0137] S11: the cooling liquid flows to each electrical conductivity sensor through the detection cavity;

[0138] S12: each electrical conductivity sensor outputs an electrical signal of 16 mA to the signal conditioning circuit at a sampling frequency of 1.5 Hz;

[0139] S13: the conditioning circuit converts the electrical signal into a digital quantity and transmits it to the ADS module;

[0140] S14: the ADS module uploads a data packet to the edge computing unit every 500 ms;

[0141] S15: the edge computing unit calculates the temperature insulation performance attenuation coefficient of each integrated water cooling block;

[0142] The signal transmission protocol of the electrical signal is Modbus RTU over RS-485;

[0143] The electrical conductivity-temperature compensation formula of the temperature insulation performance attenuation coefficient of each integrated water cooling block is σcal=σmeas×[1+0.02(T−25)].

[0144] S2: AI prediction and decision-making are performed on the cooling liquid life based on an LSTM neural network;

[0145] The AI prediction and decision-making in the S2 comprise the following steps:

[0146] S21: AI is used to monitor the data of each electrical conductivity sensor in real time;

[0147] S22: AI is used to pre-process the data of each electrical conductivity sensor;

[0148] S23: Real-time data is continuously read and transmitted to the LSTM model for inference within 30 minutes.

[0149] S3: The alarm unit processes by judging the alarm level;

[0150] Among them, when the conductivity is 2.1 μS / cm, the alarm level is one, and a short message is sent to inform the maintenance personnel to handle within 72 hours;

[0151] S4: Heat dissipation assisted by phase change material; a metal-organic framework material layer is filled between the hot end of the refrigeration fin and the heat dissipation fin to absorb transient thermal shock; the composition of the metal-organic framework material layer is ZnO(BDC), (BDC=terephthalic acid), which is a three-dimensional porous crystal structure with a specific surface area of up to 3000 m 2 / g

[0152] The S4 heat dissipation assisted by phase change material includes S41 heat absorption stage and S42 heat release stage;

[0153] S41 heat absorption stage includes the following steps;

[0154] S411: The hot end of the refrigeration fin transfers heat to the cold end of the refrigeration fin;

[0155] S412: The cold end of the refrigeration fin transfers heat to the thermal interface material;

[0156] S413: The thermal interface material transfers heat to the phase change material for latent heat absorption, keeping the temperature at 45℃;

[0157] S414: The thermal interface material transfers the heat absorbed after latent heat absorption to the metal-organic framework material layer for absorption, wherein the thermal interface material is thermal conductive silicone grease or graphene, and the phase change material is composed of 70% ZnO(BDC) and 30% graphene.

[0158] The S42 heat release stage, when the ambient temperature is <40℃, the metal-organic framework material layer begins to crystallize and release heat, allowing heat to be transferred to the heat dissipation fin through the TEG module, and the heat dissipation fin performs forced convection heat dissipation.

[0159] S5: Waste heat recovery, converting the waste heat of the hot end of the refrigeration fin into electrical energy to power the control unit;

[0160] Among them, in the S5 waste heat recovery, when the temperature difference ΔT between each hot end of the refrigeration fin and its cold end is ≥60℃, heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers heat to the energy storage super capacitor for storage to power the control unit.

[0161] S6: Dynamic topology adjustment, according to the temperature data of each integrated water-cooled block, automatically switch between series or parallel mode;

[0162] The S6 dynamic topology adjustment includes S61 mode switching judgment and S62 dynamic threshold calculation sub-steps;

[0163] S611: Real-time monitoring of cooling liquid temperature through conductivity sensor;

[0164] S612: Through LSTM, judge the temperature difference ΔT = max(T n )-min(T n )≥10℃ of the hot end and cold end of each integrated water-cooled block;

[0165] If the single-point temperature of the hot end of each integrated water-cooled block is >85℃, start parallel mode, PID control electric proportional valve opening to make water pump speed value 120%;

[0166] The S62 dynamic threshold calculation Threshold = 10−0.5×(Qcurrent−1) (unit: ℃), where Q is the current flow (L / min).

[0167] S7: Through the signal interaction system of each module, the collaborative work flow is realized.

[0168] The S7 includes the following sub-steps:

[0169] S71: Real-time monitoring of cooling liquid temperature through conductivity sensor and temperature sensor;

[0170] S72: When the temperature sensor detects that the temperature of each integrated water-cooled block is 75℃, the controller switches to parallel mode PID control electric proportional valve opening, the heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers the heat to the energy storage super capacitor for storage as a control unit power supply;

[0171] S73: When the temperature sensor detects that the temperature of each integrated water-cooled block is 78℃, the controller switches to parallel mode PID control electric proportional valve opening, switches to parallel mode, PID control electric proportional valve opening, triggers sound-light alarm and reduces frequency operation;

[0172] S74: When the temperature sensor detects that the temperature of each integrated water-cooled block is 45℃, maintain series mode, and the ADS module samples.

[0173] System-level collaborative work flow timing example:

[0174] 00:00:00 [monitoring start]

[0175] ├─ Conductivity sensor: sampled every 500ms (CS1500, ±0.1 μS / cm accuracy)

[0176] ├─ Temperature sensor: PT100, 1.5 Hz refresh rate

[0177] ├─ Data synchronization: aggregated to STM32H743 controller via RS-485 bus

[0178] ├─ Value: 2.1 μS / cm

[0179] └─ System status: OK (green indicator light)

[0180] 00:00:05 [Emergency response]

[0181] ├─ Trigger condition: Water-cooling plate temperature > 75 °C threshold → Trigger highest-priority control thread interrupt

[0182] ├─ Action performed: Electric three-way valve: switched to parallel mode within 30 ms (SMF-3P-12V model)

[0183] └─ Water pump: pressure increased from 0.1 MPa to 0.15 MPa (compensate parallel flow resistance)

[0184] 00:00:10 [Energy management]

[0185] ├─ TEG output voltage detection: MPPT algorithm locks 4.2 V maximum power point

[0186] ├─ Power switching: external power supply is disconnected, and the sensor group is directly powered by TEG

[0187] └─ Super capacitor: store excess energy (3.7 V / 10 F)

[0188] 00:00:15 [Phase change regulation]

[0189] ├─ MOF-5 temperature reaches phase transition point: absorbs transient heat shock of water-cooling plate (180 J / g)

[0190] ├─ Temperature stabilization: controlled at 45 °C (through PID regulation of cooling plate power)

[0191] └─ Heat flow diversion: waste heat is converted into electrical energy by TEG (efficiency ≥ 3%)

[0192] 00:00:30 [Mode recovery]

[0193] ├─ ΔT < 7 °C hysteresis threshold → avoid frequent switching

[0194] ├- Valve reset: switch back to series mode (flow rate drops to 1.0 L / min)

[0195] └- System self-test: confirm that all sensor data is back in normal range

[0196] An intelligent and scalable integrated circuit water cooling liquid circulation cooling system, comprising a precise heat exchange unit, a refrigeration cooling unit, a control unit, an insulation cooling liquid circulation module, and a dynamic topology control module.

[0197] The precise heat exchange unit comprises a plurality of integrated water cooling blocks and a plurality of electric proportional valves, the hot end of each integrated water cooling block is filled with a metal organic framework layer between the heat dissipation fins, the hot end of each integrated water cooling block is sprayed with an Al2O3 nano coating, the cold end of each integrated water cooling block is reinforced with a uniform temperature plate, the cold end of each water cooling block is attached with a thermal interface material layer, and the thermal interface material layer is attached with a metal-organic framework material layer.

[0198] The refrigeration cooling unit comprises a plurality of refrigeration pieces, the model of which is TEC-12705, and each refrigeration piece is attached with a paraffin phase change layer, and the melting point of the paraffin phase change layer is 45℃.

[0199] The control unit comprises a temperature sensor, a flow sensor, an electric conductivity sensor, an edge computing module, and an MPPT controller, and the model of the edge computing module is STM32H743.

[0200] The insulation cooling liquid circulation module comprises a water outlet manifold, a water return manifold, a fixed pipeline, an air-cooled water exhaust, and a water pump, the insulation cooling liquid is fluorinated liquid, the model of which is FS270, and the insulation strength is ≥ 45 kV / mm.

[0201] The dynamic topology control module comprises an ADS module and a TEG module.

[0202] Example 3

[0203] A control method of an intelligent and scalable integrated circuit water cooling liquid circulation cooling system, the method comprising the following steps:

[0204] S1: number each integrated water cooling block as 1~N, and deploy an electric conductivity sensor at the water return manifold to monitor the cooling liquid temperature insulation performance attenuation coefficient flowing through each integrated water cooling block in real time; the S1 comprises the following steps:

[0205] S11: the cooling liquid flows to each electric conductivity sensor through the detection cavity;

[0206] S12: each electric conductivity sensor outputs a 20mA electric signal to the signal conditioning circuit with a sampling frequency of 2Hz;

[0207] S13: The conditioning circuit converts the electrical signal into a digital quantity and transmits it to the ADS module;

[0208] S14: The ADS module uploads the data packet to the edge computing unit every 500 ms;

[0209] S15: The edge computing unit calculates the temperature insulation performance decay coefficient of each integrated water cooling block;

[0210] The signal transmission protocol of the electrical signal is Modbus RTU over RS-485;

[0211] The conductivity-temperature compensation formula of the temperature insulation performance decay coefficient of each integrated water cooling block is σcal=σmeas×[1+0.02(T−25)].

[0212] S2: AI prediction and decision-making based on LSTM neural network to predict the service life of the cooling liquid;

[0213] The AI prediction and decision-making in S2 include the following steps:

[0214] S21: Real-time monitoring of data from each conductivity sensor through AI;

[0215] S22: Preprocessing of data from each conductivity sensor through AI;

[0216] S23: Continuously reading real-time data within 30 minutes and transmitting it to the LSTM model for inference and judgment.

[0217] S3: The alarm unit processes by judging the alarm level;

[0218] When the conductivity is > 5 μS / cm, the alarm level is two, the alarm unit records the fault code F005 and uploads it to the cloud, and automatically reduces the frequency of each cooling fin. The cloud pushes a work order to notify the maintenance personnel through a short message within 24 hours for emergency treatment.

[0219] S4: Heat dissipation assisted by phase change material; a metal-organic framework material layer is filled between the hot end of the cooling fin and the heat dissipation fin to absorb transient thermal shock; the composition of the metal-organic framework material layer is ZnO(BDC), (BDC = terephthalic acid), which is a three-dimensional porous crystal structure with a specific surface area of up to 3000 m 2 / g

[0220] The heat dissipation assisted by phase change material in S4 includes a heat absorption stage S41 and a heat release stage S42;

[0221] S41 heat absorption stage includes the following steps;

[0222] S411: The hot end of the refrigeration fin transfers heat to the cold end of the refrigeration fin;

[0223] S412: The cold end of the refrigeration fin transfers heat to the thermal interface material;

[0224] S413: The thermal interface material transfers heat to the phase change material for latent heat absorption, keeping the temperature at 47°C;

[0225] S414: The thermal interface material transfers the heat absorbed after latent heat absorption to the metal-organic framework material layer for absorption, wherein the thermal interface material is thermal conductive silicone grease or graphene, and the phase change material is filled with 70% ZnO(BDC) and 30% graphene.

[0226] The S42 heat release stage, when the ambient temperature <40℃, the metal-organic framework material layer starts to crystallize and release heat to transfer heat to the heat dissipation fins through the TEG module, and the heat dissipation fins perform forced convection heat dissipation.

[0227] S5: Waste heat recovery, converting the waste heat of the hot end of the refrigeration fin into electrical energy to power the control unit;

[0228] In the S5 waste heat recovery, when the temperature difference ΔT between each refrigeration fin hot end and its cold end ≥60℃, heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers heat to the energy storage supercapacitor for storage to power the control unit.

[0229] S6: Dynamic topology adjustment, automatically switching between series or parallel mode according to the temperature data of each integrated water cooling block;

[0230] The S6 dynamic topology adjustment includes S61 mode switching judgment and S62 dynamic threshold calculation steps;

[0231] S611: Real-time monitoring of the cooling liquid temperature through the conductivity sensor;

[0232] S612: Determine the temperature difference ΔT between the hot end and the cold end of each integrated water cooling block through LSTM, max(T n )-min(T n )≥10℃,

[0233] If the temperature difference ΔT between the hot end and the cold end of each integrated water cooling block is max(T n )-min(T n )≥10℃, start parallel mode, PID control electric proportional valve opening to make water pump speed value 120%;

[0234] The S62 dynamic threshold calculation Threshold=10−0.5×(Qcurrent−1)(unit: ℃), where Q is the current flow (L / min).

[0235] S7: Collaborative workflow through module signal interaction system.

[0236] The S7 includes the following steps:

[0237] S71 Real-time monitoring of the temperature of the cooling liquid flowing through each integrated water cooling block by conductivity sensors and temperature sensors.

[0238] S72: When the temperature sensor detects that the temperature of each integrated water cooling block is at 78°C, the controller switches to parallel mode PID control of the electric proportional valve opening, and the heat is transferred to the MPPT controller through the TEG module. The MPPT controller transfers the heat to the energy storage supercapacitor for storage as a control unit power supply.

[0239] S73: When the temperature sensor detects that the temperature of each integrated water cooling block is at 78°C, the controller switches to parallel mode PID control of the electric proportional valve opening, switches to parallel mode, PID control of the electric proportional valve opening, triggers an audible and visual alarm, and runs at a reduced frequency.

[0240] S74: When the temperature sensor detects that the temperature of each integrated water cooling block is at 47°C, maintain series mode, and the ADS module performs sampling.

[0241] System-level collaborative workflow timing example:

[0242] 00:00:00 [Monitoring starts]

[0243] ├- Conductivity sensor: sample every 500ms (CS1500, ±0.1 μS / cm accuracy)

[0244] ├- Temperature sensor: PT100, 2Hz refresh rate (±0.5°C)

[0245] ├- Data synchronization: aggregated through RS-485 bus to STM32H743 controller

[0246] ├- Value: > 5 μS / cm

[0247] └- System status: normal (green indicator light)

[0248] 00:00:05 [Emergency response]

[0249] ├- Trigger condition: water cooling sheet temperature > 78°C → trigger the highest priority control thread interruption

[0250] ├- Execute action: electric three-way valve: switch to parallel mode within 30ms (SMF-3P-12V model)

[0251] └─ Water pump: pressure from 0.1 MPa to 0.15 MPa (compensate parallel flow resistance)

[0252] 00:00:10 [Energy management]

[0253] ├─ TEG output voltage detection: MPPT algorithm locks 4.2V maximum power point

[0254] ├─ Power switching: turn off external power supply, and directly supply the sensor group by TEG

[0255] └─ Super capacitor: store excess energy (3.7V / 10F)

[0256] 00:00:15 [Phase change regulation]

[0257] ├─ MOF-5 temperature reaches phase change point: absorbs transient heat shock (180J / g) of water cooling fin

[0258] ├─ Temperature stabilization: control at 47℃ (regulate cooling fin power by PID)

[0259] └─ Heat flow diversion: waste heat is converted into electrical energy by TEG (efficiency ≥ 3%)

[0260] 00:00:30 [Mode recovery]

[0261] ├─ ΔT>10℃ hysteresis threshold

[0262] ├─ Valve reset: switch back to series mode (flow rate drops to 1.0L / min)

[0263] └─ System self-check: confirm that all sensor data returns to normal range

[0264] An intelligent and scalable integrated circuit water cooling liquid circulation heat dissipation system, comprising a precise heat exchange unit, a refrigeration heat dissipation unit, a control unit, an insulation cooling liquid circulation module and a dynamic topology control module.

[0265] The precise heat exchange unit comprises a plurality of integrated water cooling blocks and a plurality of electric proportional valves, a metal organic framework layer is filled between the hot end of each integrated water cooling block and the heat dissipation fin, an Al2O3 nano coating is sprayed on the hot end of each integrated water cooling block, the cold end of each integrated water cooling block is reinforced by using a uniform temperature plate, a thermal interface material layer is attached to the cold end of each water cooling block, and a metal-organic framework material layer is attached to the thermal interface material layer.

[0266] The refrigeration heat dissipation unit comprises a plurality of refrigeration fins, each of which is a TEC-12705, and each refrigeration fin is attached with a paraffin phase change layer, and the melting point of the paraffin phase change layer is 45℃.

[0267] The control unit comprises a temperature sensor, a flow sensor, a conductivity sensor, an edge calculation module and an MPPT controller, and the edge calculation module is an STM32H743.

[0268] The insulation cooling liquid circulation module comprises a water outlet manifold, a backwater manifold, a fixed pipeline, an air-cooled water discharge and a water pump, the insulation cooling liquid is a fluorinated liquid, the model is FS270, and the insulation strength is greater than or equal to 45 kV / mm.

[0269] The dynamic topology control module comprises an ADS module and a TEG module.

[0270] The above only describes certain exemplary embodiments of the present application in a descriptive manner, and it is needless to say that the described embodiments can be modified in various ways without departing from the spirit and scope of the present application for those skilled in the art. Therefore, the above drawings and descriptions are illustrative in nature and should not be understood as limiting the scope of protection of the claims of the present application.

Claims

1. A control method of an intelligent extensible integrated circuit water cooling liquid circulating heat dissipation system, characterized in that, The method comprises the following steps: S1: number each integrated water cooling block as 1-N, and deploy an electrical conductivity sensor at the water return manifold to monitor the temperature insulation performance attenuation coefficient of the cooling liquid flowing through each integrated water cooling block in real time; S2: predict and determine the cooling liquid life by AI based on the LSTM neural network; S3: the alarm unit processes by judging the alarm level; S4: auxiliary heat dissipation by phase change material; fill the metal-organic framework material layer between the hot end of the refrigeration fin and the heat dissipation fin to absorb transient thermal shock; S5: waste heat recovery, convert the waste heat at the hot end of the refrigeration fin into electrical energy to power the control unit; S6: dynamic topology adjustment, automatically switch between series or parallel modes according to the temperature data of each integrated water cooling block; S7: collaborative workflow through each module signal interaction system.

2. The control method of the intelligent scalable integrated circuit water cooling liquid circulation heat dissipation system according to claim 1, characterized in that, The S1 comprises the following steps: S11: the cooling liquid flows through the detection cavity to each electrical conductivity sensor; S12: each electrical conductivity sensor outputs a 4-20mA electrical signal to the signal conditioning circuit with a sampling frequency of 1-2Hz; S13: the conditioning circuit converts the electrical signal into a digital quantity and transmits it to the ADS module; S14: the ADS module uploads data packets to the edge computing unit every 500ms; S15: the edge computing unit calculates the temperature insulation performance attenuation coefficient of each integrated water cooling block.

3. The control method of the intelligent scalable integrated circuit water cooling liquid circulating heat dissipation system according to claim 2, characterized in that, The signal transmission protocol of the electrical signal is Modbus RTU over RS-485; The electrical conductivity-temperature compensation formula of the temperature insulation performance attenuation coefficient of each integrated water cooling block is σcal=σmeas×[1+0.02(T−25)].

4. The control method of the intelligent scalable integrated circuit water cooling liquid circulating heat dissipation system according to claim 1, characterized in that, The AI prediction and determination in S2 comprises the following steps: S21: monitor the data of each electrical conductivity sensor in real time through AI; S22: preprocess the data of each electrical conductivity sensor through AI; S23: continuously read real-time data within 30 minutes and transmit it to the LSTM model for inference and judgment.

5. The control method of the intelligent scalable integrated circuit water cooling liquid circulating heat dissipation system according to claim 1, characterized in that, S3 comprises the following steps: S31: when the electrical conductivity is 2-5 μS / cm, the alarm level is level one, and a short message is sent to notify the maintenance personnel to handle it within 72 hours; S32: when the electrical conductivity is >5 μS / cm, the alarm level is level two, the alarm unit records the fault code F005 and uploads it to the cloud, and each refrigeration fin automatically runs at a lower frequency, and the cloud pushes a work order by sending a short message to notify the maintenance personnel to handle it urgently within 24 hours.

6. The control method of the intelligent scalable integrated circuit water cooling liquid circulation heat dissipation system according to claim 1, characterized in that, The auxiliary heat dissipation by phase change material in S4 comprises S41 heat absorption stage and S42 heat release stage; S41 heat absorption stage comprises the following steps: S411: the hot end of the refrigeration fin transmits heat to the cold end of the refrigeration fin; S412: the cold end of the refrigeration fin transmits heat to the thermal interface material; S413: the thermal interface material transmits heat to the phase change material for latent heat absorption, keeping the temperature at 43-47℃; S414: the thermal interface material transmits the heat absorbed after latent heat absorption to the metal-organic framework material layer for absorption.

7. The control method of the intelligent scalable integrated circuit water cooling liquid circulating heat dissipation system according to claim 6, characterized in that, The S42 heat release stage, when the ambient temperature <40℃, the metal-organic framework layer starts to crystallize and release heat, which is transferred to the heat dissipation fins through the TEG module, and the heat dissipation fins perform forced convection heat dissipation.

8. The control method of the intelligent scalable integrated circuit water cooling liquid circulating heat dissipation system according to claim 1, characterized in that, The S5 waste heat recycling, when the temperature difference ΔT between the hot end and the cold end of each refrigeration sheet ≥60℃, the heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers the heat to the energy storage super capacitor for storage to power the control unit.

9. The control method of the intelligent scalable integrated circuit water cooling liquid circulating heat dissipation system according to claim 1, characterized in that, The S6 dynamic topology adjustment includes S61 mode switching judgment and S62 dynamic threshold calculation sub-steps; S611: Real-time monitoring of the temperature of the cooling liquid flowing through each integrated water cooling block by the conductivity sensor; S612: Determine the temperature difference ΔT between the hot end and the cold end of each integrated water-cooled block by LSTM, ΔT = max(T n ) - min(T n ) ≥ 10℃; S613: If ΔT = max(T n ) - min(T n ) ≤ 5~7℃, maintain series mode, if single point temperature of each integrated water cooling block hot end > 85℃, start parallel mode, PID control electric proportional valve opening to make water pump speed value 120%; S614: If the temperature difference ΔT = max(T n ) - min(T n ) of the hot end and the cold end of each integrated water cooling block is greater than or equal to 10℃, start the parallel mode, and PID control the opening of the electric proportional valve to make the water pump speed value 120%.

10. The control method of the intelligent scalable integrated circuit water cooling liquid circulating heat dissipation system according to claim 9, characterized in that, The S62 dynamic threshold calculation Threshold=10−0.5×(Qcurrent−1) (unit: ℃), where Q is the current flow (L / min).

11. The control method of the intelligent scalable integrated circuit water cooling liquid circulating heat dissipation system according to claim 1, characterized in that, The S7 includes the following sub-steps: S71 Real-time monitoring of the temperature of the cooling liquid flowing through each integrated water cooling block by the conductivity sensor and the temperature sensor; S72: When the temperature sensor detects that the temperature of each integrated water cooling block is in the range of 75~78℃, the controller switches the parallel mode PID control of the electric proportional valve opening, the heat is transferred to the MPPT controller through the TEG module, and the MPPT controller transfers the heat to the energy storage super capacitor for storage to power the control unit; S73: When the temperature sensor detects that the temperature of each integrated water cooling block is in the range of 78℃, the controller switches the parallel mode PID control of the electric proportional valve opening, switches the parallel mode, PID control of the electric proportional valve opening, triggers the sound and light alarm and reduces the frequency of operation; S74: When the temperature sensor detects that the temperature of each integrated water cooling block is in the range of 43~47℃, maintain the series mode, and the ADS module performs sampling.

12. An intelligent scalable integrated circuit water cooling liquid circulating heat sink system, characterized in that, It includes an accurate heat exchange unit, a refrigeration and heat dissipation unit, a control unit, an insulating cooling liquid circulation module, and a dynamic topology control module.

13. The intelligent scalable integrated circuit water cooling liquid circulating heat sink system of claim 12, wherein, The accurate heat exchange unit includes a plurality of integrated water cooling blocks and a plurality of electric proportional valves, the hot end of each integrated water cooling block is filled with a metal-organic framework layer between the heat dissipation fins, the hot end of each integrated water cooling block is sprayed with an Al2O3 nano coating, the cold end of each integrated water cooling block is reinforced with a uniform temperature plate, the cold end of each water cooling block is attached with a thermal interface material layer, and the thermal interface material layer is attached with a metal-organic framework material layer.

14. The intelligent scalable integrated circuit water cooling liquid circulating heat sink system of claim 12, wherein, The refrigeration and heat dissipation unit includes a plurality of refrigeration sheets, the model of which is TEC-12705, and each refrigeration sheet is attached with a paraffin phase change layer.

15. The intelligent scalable integrated circuit water cooling liquid circulating heat sink system of claim 12, wherein, The control unit includes a temperature sensor, a flow sensor, a conductivity sensor, an edge computing module, and an MPPT controller, the model of the edge computing module is STM32H743.

16. The intelligent scalable integrated circuit water cooling liquid circulating heat sink system of claim 12, wherein, The insulating cooling liquid circulation module includes a water outlet manifold, a water return manifold, a fixed pipeline, an air-cooled water exhaust, and a water pump, the insulating cooling liquid is fluorinated liquid, and the insulation strength is ≥45 kV / mm.

17. The intelligent scalable integrated circuit water cooling liquid circulating heat sink system of claim 12, wherein, The dynamic topology control module includes an ADS module and a TEG module.

Images