Intelligent constant-temperature single-phase immersion data center composite liquid cooling control method and system

Abstract

The invention relates to the technical field of single-phase immersion liquid cooling data processing, and particularly provides an intelligent constant-temperature single-phase immersion data center composite liquid cooling control method and system, and the method comprises the steps: monitoring the temperature change of a server mainboard in real time through a temperature sensing bulb, converting the temperature data into an electric signal through the temperature sensing bulb, and transmitting the electric signal to the server mainboard; the electric signal is transmitted to the data integration processor as an input signal; after the data integration processor receives the electric signal, an internal processor performs calculation according to a preset algorithm to generate a refrigerant flow control signal; the electromagnetic valve adjusts the refrigerant distribution amount of the liquid separator according to the refrigerant flow control signal, and regulation and control of the refrigerant flow are achieved. The adjusted refrigerant flow acts on the phase change heat absorption process of the refrigerant in the coil pipe, so that the temperature of the fluorinated liquid is stable. The system comprises a coil pipe, a server, a liquid separator, a data integration processor, a server mainboard, a temperature wrap, an evaporator and the like. The data integration processor generates a refrigerant flow control signal according to a feedback signal to form a closed-loop control loop.

Description

Technical Field

[0001] The present invention relates to the technical field of single-phase immersion liquid cooling data processing technology, and in particular to an intelligent constant-temperature single-phase immersion data center composite liquid cooling control method and system. Background Art

[0002] As society becomes increasingly information-based, the scale of data centers continues to grow, bringing with it significant energy consumption issues. Data center cooling systems consume approximately 40% of their total energy consumption. In recent years, the maximum heat dissipation of server-type loads in computer rooms has climbed to over 20 kW per cabinet. The existing air-cooled precision air conditioning system, with a maximum cooling capacity of 4 kW / m2 per unit area under ideal air supply conditions, is no longer sufficient. Liquid cooling is more efficient than traditional air cooling, but two-phase liquid cooling suffers from slow response, unstable temperature control, and high equipment requirements. Furthermore, the phase change process generates bubbles, and bubble erosion increases the failure rate of data center motherboard systems. Single-phase liquid cooling also suffers from uneven cooling capacity distribution among heat-generating units, resulting in large temperature differences between units.

[0003] Prior art 1, Chinese patent application number 202510529095.4, discloses a method for immersion-type liquid cooling and flame retardancy in data centers. The method involves placing the data center in a sealed enclosure and adding a coolant to the enclosure. The coolant comprises a mixture of 70 to 95 parts by weight of a base synthetic oil and 0.5 to 1 part by weight of an additive composition. After the coolant covers the upper surface of the data center, a fluoride with a boiling point of no more than 100 degrees Celsius is added to the enclosure until the fluoride concentration in the enclosure reaches a predetermined flame retardant concentration. While this method effectively improves the inertness and stability of the coolant at a low cost, it lacks a dynamic temperature control mechanism and relies solely on static immersion cooling, making it impossible to dynamically adjust the refrigerant flow rate based on server temperature changes, limiting cooling efficiency. The coolant composition relies on chemical flame retardancy rather than active temperature control, relying on fluoride concentration to enhance flame retardancy rather than precise temperature control to prevent overheating, posing a risk of local hot spots. The method also lacks a closed-loop feedback system and integrated closed-loop control of temperature monitoring and flow regulation, making it difficult to adapt to dynamic load changes.

[0004] Prior art two, Chinese patent application number 202211085078.9, discloses a phase-change immersion liquid cooling cabinet for a data center. The cabinet includes a cabinet with servers and condensers inside. The servers and condensers are immersed in a liquid area. The condensers are immersed in the liquid area. Coolant flows inside the cabinet and is located at the top of the liquid area. When the heat source surface in the server boils and bubbles rise to the top of the liquid area, the bubbles release heat to the condenser under a combined mechanism of "forced convection + condensation". The bubbles do not escape the liquid surface, and the top area of ​​the liquid where the condenser is located presents a state of gas-liquid coexistence. Although the structure is simple and easy to use, it relies on natural phase change and has low adjustment accuracy. It only passively dissipates heat through "forced convection + condensation", and cannot accurately control the refrigerant flow to adapt to different heat loads. Server-level temperature control is not achieved, and the bubble condensation mechanism cannot independently adjust the temperature of a single server motherboard, which may cause local overheating. There is no intelligent flow distribution, and the heat exchange between the condenser and the server lacks active flow distribution components such as liquid distributors and solenoid valves, resulting in insufficient flexibility.

[0005] Currently, existing technologies 1 and 2 have problems with low cooling efficiency, local overheating, and energy efficiency imbalance due to the lack of dynamic closed-loop temperature control. Therefore, the present invention provides an intelligent constant temperature single-phase immersion data center composite liquid cooling control method and system. Summary of the Invention

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: In one aspect of the present invention, a method for controlling composite liquid cooling of an intelligent constant temperature single-phase immersion data center is provided, comprising the following steps: After the data integration processor receives the electrical signal converted from the temperature data of the server motherboard, the internal processor calculates according to the preset algorithm to generate a refrigerant flow control signal; the refrigerant flow control signal is output to the solenoid valve as a regulation instruction to guide its action.

[0007] In an optional embodiment, the process of generating the refrigerant flow control signal includes the following steps: Obtain a collection of electrical signals output by multiple temperature sensors, each of which corresponds to a temperature-time gradient at a specific location on the server motherboard. The processor's built-in heterogeneous computing core maps electrical signal groups into a three-dimensional heat flow channel network based on physical location. Based on the thermal conductivity characteristics of the fluorinated liquid and the coil layout, the superposition heat flow is decoupled to obtain the transient heat flux density scalar of each channel and generate the heat flux density distribution. The required heat absorption equivalent of each coil unit is calculated based on the heat flux density distribution and the phase change hysteresis parameter of the two-phase coil. The heat absorption equivalent is converted into the refrigerant mass flow rate vector of the corresponding distributor branch and output as a discrete control signal executable by the solenoid valve.

[0008] In an optional embodiment, the process of outputting a discrete control signal executable by the solenoid valve includes the following steps: Input the transient heat flux density scalar of each channel. Each transient heat flux density scalar corresponds to the spatiotemporal distribution of thermal energy in the service area of ​​a single coil. Based on the two-phase coil phase change hysteresis parameter and the response delay of the refrigerant from liquid to gas, a thermal inertia decay function is established; the transient heat flux density scalar is convolved with the phase change leader along the time axis to output a net heat absorption flux value that matches the coil phase change rate in real time; Based on the latent heat of phase change per unit mass of the refrigerant in the coil, the net heat absorption flux is converted into the instantaneous value of the refrigerant mass flow rate required to maintain thermal balance; according to the mapping relationship between the distributor branch and the coil unit, the flow values ​​are aggregated to generate the branch flow rate vector.

[0009] In an optional implementation, the process of aggregating flow values ​​to generate branch flow rate vectors includes the following steps: Continuously track the operating status of the compressor and the pressure changes inside the evaporator, while taking into account the time delay required for the refrigerant to change from liquid to gas. Based on the real-time data of pressure changes and time delays, the actual heat exchange value currently required is obtained, and the amount of heat required in the future is estimated, ultimately providing a net heat absorption flux value for the adjusted heat removal requirement. Based on the latent heat constant per unit mass of the refrigerant at a specific saturation pressure, the net heat absorption flux value is divided by the latent heat constant per unit mass to obtain the refrigerant mass flow rate per unit time required for the current thermal balance; the refrigerant mass flow rate per unit time required to maintain the current thermal balance is output, and its value is linearly proportional to the heat absorption flux; Based on the spatial mapping topology of the distributor branch and the coil unit, which is determined by the parallel coil structure in the evaporator, the flow values ​​of multiple coil units belonging to the same distributor branch are scalared and superimposed. The branch flow rate vectors of each distributor outlet are output, and the vector dimensions match the solenoid valve execution port.

[0010] In an optional embodiment, the process of scalaring and superimposing the flow values ​​of multiple coil units belonging to the same distributor branch includes the following steps: A collection of refrigerant mass flow values ​​per unit time output, each value corresponding to the real-time mass requirement of a single coil unit to maintain thermal balance; Based on the hard-connected mapping relationship between the distributor branch and the coil unit, the physical piping arrangement of the parallel coils in the evaporator is uniquely determined. The mass flow of all coil units under the same distributor branch is calculated by performing an unweighted scalar algebraic sum operation, and the total mass flow of each branch is output to satisfy the law of fluid continuity. The total mass flow values ​​of each branch are arranged according to the inherent spatial sequence of the dispenser outlet, and a discrete flow rate instruction sequence that is completely isomorphic to the number of dispenser outlets and the solenoid valve execution ports is generated.

[0011] In an optional embodiment, the process of outputting the total mass flow value of each branch includes the following steps: The outlet flow passages of all coil units under the same distributor branch complete physical convergence at the distributor branch main pipe section; The mass flow of each coil unit automatically forms a flux density distribution on the main pipe cross section according to the flow channel geometric projection ratio; the flux density of the main pipe cross section is surface integrated and the scalar integral value is output; The scalar integral value is directly equivalent to the steady-state output total mass flux of the distributor branch according to the fluid continuity law; the total mass flux forms a dynamic balance with the suction characteristics of the downstream compressor.

[0012] In an optional implementation, the process of outputting a scalar integral value includes the following steps: The mass flow of each coil unit on the main pipe cross section of the distributor branch is automatically decomposed into normal components perpendicular to the pipe wall based on the spatial orientation angle between its outlet flow channel and the main pipe axis. All normal components are continuously distributed on the main pipe cross section according to the position coordinates to form a flux density field. The flux conservation integral is performed on the flux density field over the entire cross section of the main pipe; the output of the integral is equivalent to the total mass flux scalar of the cross section.

[0013] In an optional embodiment, the temperature sensing package monitors the temperature changes of the server mainboard in real time, and the temperature sensing package converts the temperature data into an electrical signal, which is transmitted as an input signal to the data integration processor.

[0014] In an optional embodiment, the solenoid valve adjusts the refrigerant distribution amount of the distributor according to the refrigerant flow control signal to achieve refrigerant flow control; the adjusted refrigerant flow acts on the phase change heat absorption process of the refrigerant in the coil to stabilize the temperature of the fluorinated liquid.

[0015] Another aspect of the present invention provides an intelligent constant temperature single-phase immersion data center composite liquid cooling control system, which implements the intelligent constant temperature single-phase immersion data center composite liquid cooling control method, and includes: a coil, a server, a liquid distributor, a gas collecting pipe, a compressor, a condenser, a liquid storage tank, a ball valve, a pump, a one-way valve, a data integration processor, a server motherboard, a two-phase liquid cooling coil, a solenoid valve, a temperature sensor, and an evaporator.

[0016] Among them, multiple groups of coils and servers are arranged in the evaporator, the coils are connected to the gas collecting pipe, the servers are arranged between adjacent coils, and the liquid separator is arranged at the coolant inlet end of the lower left end of the evaporator; the outlet end of the gas collecting pipe is connected to the inlet end of the compressor through a pipeline, the outlet end of the compressor is connected to the inlet end of the condenser through a pipeline, the outlet end of the condenser is connected to the inlet end of the liquid storage tank through a pipeline, the outlet end of the liquid storage tank is connected to one end of the ball valve through a pipeline, the other end of the ball valve is connected to one end of the pump through a pipeline, the other end of the pump is connected to one end of the one-way valve through a pipeline, and the other end of the one-way valve is connected to the inlet end of the liquid separator through a pipeline; The server consists of a server motherboard, which is installed between adjacent coils. The coils are made up of multiple two-phase liquid cooling coils connected together. The data integration processor is connected to the electromagnetic valve through electrical signals, and the electromagnetic valve is connected to the dispenser; the data integration processor is connected to multiple temperature sensing packages through electrical signals, and the temperature sensing packages are installed on both sides of the server motherboard.

[0017] The present invention realizes the establishment of a closed-loop temperature control system. The temperature sensing package monitors the temperature change of the server mainboard to form a temperature feedback signal. The data integration processor generates a refrigerant flow control signal according to the feedback signal to form a closed-loop control circuit. The solenoid valve responds to the control signal to adjust the flow of the dispenser to achieve dynamic cooling capacity matching; the phase change cooling process is precisely regulated, the refrigerant phase change heat absorption rate in the coil is directly regulated by the refrigerant flow control signal, the dynamic adjustment of the dispenser distribution amount maintains the temperature stability of the fluorinated liquid, and the opening change of the solenoid valve forms a corresponding relationship with the mainboard temperature change; the system response characteristics are optimized, the temperature sensing package electrical signal transmission shortens the temperature detection delay, the preset algorithm of the data integration processor realizes the rapid generation of the control signal, and the solenoid valve-dispenser linkage mechanism improves the flow regulation response speed; the thermal management accuracy is improved, the full-link closed-loop control of the temperature signal, control signal, and flow regulation, the dynamic balance of the refrigerant phase change heat absorption process and the heat load of the heat source is achieved, and the thermal inertia influence of the traditional open-loop control is eliminated; the system reliability is enhanced, the electrical signal transmission avoids the failure risk of the mechanical transmission link, the preset algorithm ensures the consistency of the control logic, and the flow graded regulation prevents system oscillation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of the intelligent constant temperature single-phase immersion data center composite liquid cooling control method provided in Example 1 of the present invention; Figure 2 This is a schematic diagram of the intelligent constant temperature single-phase immersion data center composite liquid cooling control method provided in Example 1 of the present invention; Figure 3 This is a diagram showing the process of converting temperature data into an electrical signal by the temperature sensing package provided in Example 2 of the present invention; Figure 4 This is a process diagram of generating a refrigerant flow control signal provided in Example 4 of the present invention; Figure 5 This is a process diagram for achieving refrigerant flow control provided in Example 11 of the present invention; Figure 6 This is a schematic diagram of the structure of the intelligent constant temperature single-phase immersion data center composite liquid cooling control system provided in Example 12 of the present invention; Figure 7 This is a schematic diagram of the server box structure provided in Example 12 of the present invention; Figure 8 This is a schematic structural diagram of the data integration processor and spot shampooing provided in Example 12 of the present invention; Figure 9 This is a schematic structural diagram of a server motherboard provided in Example 12 of the present invention; Figure 10 A block diagram of the electronic device provided by the present invention; Figure 11 A block diagram of a computer-readable storage medium provided by the present invention; Figure numerals: 1. Coil; 2. Server; 3. Liquid distributor; 4. Gas collecting pipe; 5. Compressor; 6. Condenser; 7. Liquid storage tank; 8. Ball valve; 9. Pump; 10. One-way valve; 11. Data integration processor; 12. Server motherboard; 13. Two-phase liquid cooling coil; 14. Solenoid valve; 15. Temperature sensor; 16. Evaporator; 17. Central processing unit / microprocessor / main control chip; 18. Storage medium; 19. Data bus; 20. Input / output bus / external bus / device bus; 21. Display; 22. Input / output device; 23. Computer-readable instructions; 24. Non-temporary computer-readable storage medium. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0020] In the following, the terms "first," "second," etc., are used for descriptive convenience only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature identified with "first," "second," etc., may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0021] In the present invention, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense. For example, "connection" can be a fixed mechanical connection, a detachable mechanical connection, or an integrated one; or, "connection" can be a direct connection or an indirect connection through an intermediate medium. In addition, unless otherwise clearly specified and limited, the term "coupling" should be understood in a broad sense. For example, "coupling" can be a direct electrical connection, such as physical contact and electrical conduction between two components, or it can be understood as the electrical connection between different components in a circuit structure through a physical line that can transmit electrical signals, such as printed circuit board (PCB) copper foil or wire, so as to transmit electrical signals; or, "coupling" can be an indirect electrical connection between two components through an intermediate medium; or, "coupling" can be an electrical connection between two components in an airless / non-contact manner, such as electrical connection between two components using capacitive coupling to transmit electrical signals.

[0022] In an embodiment of the present invention, directional terms such as "up", "down", "left" and "right" may be defined including but not limited to the orientation relative to the schematic placement of the components in the drawings. It should be understood that these directional terms may be relative concepts, which are used for relative description and clarification, and may change accordingly according to changes in the orientation of the components in the drawings.

[0023] Example 1: Figure 1 As shown, an embodiment of the present invention provides an intelligent constant temperature single-phase immersion data center composite liquid cooling control method, comprising the following steps: Step S100: The temperature sensor package monitors the temperature change of the server motherboard in real time, converts the temperature data into an electrical signal, and transmits the electrical signal as an input signal to the data integration processor; Step S200: After the data integration processor receives the electrical signal, the internal processor performs calculations according to a preset algorithm to generate a refrigerant flow control signal; the refrigerant flow control signal is output as a regulation instruction to the solenoid valve to guide its operation; Step S300: the solenoid valve adjusts the refrigerant distribution amount of the distributor according to the refrigerant flow control signal to achieve refrigerant flow control; the adjusted refrigerant flow acts on the phase change heat absorption process of the refrigerant in the coil to stabilize the temperature of the fluorinated liquid.

[0024] In the above embodiments, the specific principles are shown in the attached Figure 2This embodiment realizes the establishment of a closed-loop temperature control system. The temperature sensor monitors the temperature change of the server motherboard and generates a temperature feedback signal. The data integration processor generates a refrigerant flow control signal based on the feedback signal to form a closed-loop control circuit. The solenoid valve responds to the control signal to adjust the flow of the dispenser to achieve dynamic cooling capacity matching. The phase change cooling process is precisely controlled. The refrigerant phase change heat absorption rate in the coil is directly adjusted by the refrigerant flow control signal. The dynamic adjustment of the dispenser distribution volume maintains the temperature stability of the fluorinated liquid. The change in the opening of the solenoid valve forms a corresponding relationship with the change in the motherboard temperature. The system response characteristics are optimized. The temperature sensor electrical signal transmission shortens the temperature detection delay. The preset algorithm of the data integration processor realizes the rapid generation of the control signal. The solenoid valve-dispenser linkage mechanism improves the flow regulation response speed. The thermal management accuracy is improved. The full-link closed-loop control of temperature signals, control signals, and flow regulation is carried out. The refrigerant phase change heat absorption process is dynamically balanced with the heat load of the heat source, eliminating the thermal inertia effect of traditional open-loop control. The system reliability is enhanced. The electrical signal transmission avoids the failure risk of the mechanical transmission link. The preset algorithm ensures the consistency of the control logic. The flow rate is adjusted in stages to prevent system oscillation.

[0025] Example 2: Figure 3 As shown, based on Example 1, the process of converting temperature data into an electrical signal by the temperature sensing package in step S100 provided in the embodiment of the present invention includes the following steps: Step S101: The server motherboard generates heat, causing the thermosensitive polymer layer embedded in the temperature sensor to absorb a non-uniform thermal field. The thermosensitive polymer is composed of a customized copolymer compatible with fluorinated liquids, and its molecular chains undergo directionally folding at a specific temperature threshold. Step S102: When the molecular chains of the thermosensitive polymer fold, microscale contraction stress is generated. The microscale contraction stress directly acts on the multi-stage linkage micro-arm structure below the polymer layer, converting the nanoscale folding energy into micron-scale deformation displacement; Step S103: The multi-stage linkage micro-arms displace and squeeze the stacked piezoelectric crystal array physically coupled thereto. The dipoles inside the stacked piezoelectric crystal array undergo directional deflection under mechanical stress, resulting in an asymmetric distribution of surface charges, and ultimately outputting an electrical signal that is nonlinearly related to the deformation.

[0026] In the above-mentioned embodiments, this embodiment achieves thermal-mechanical conversion. The customized copolymer compatible with fluorinated liquid triggers the directional folding of the molecular chain at a specific temperature threshold, achieving direct temperature-deformation conversion. The absorption of the non-uniform thermal field avoids the thermal inertia problem of traditional metal temperature sensors. The nanoscale contraction stress generated by the folding of the molecular chain is amplified into micron-level displacement by a multi-stage linkage micro-arm structure, ensuring that tiny temperature changes can be effectively captured. Mechanical-electrical signal conversion, coupled response of the piezoelectric crystal array, the stacked piezoelectric crystal array converts micro-arm displacement into mechanical stress through physical coupling, and the directional deflection of the dipole realizes the conversion of strain energy into electrical energy. The asymmetric distribution of surface charge makes the output electrical signal nonlinear with the deformation, enhancing the resolution of temperature changes.

[0027] In summary, the molecular chain folding mechanism of the thermosensitive polymer in this embodiment is combined with multi-stage micro-arm displacement amplification to achieve a high response to tiny temperature changes (nanoscale thermal stress, micron-scale displacement). The fluorinated liquid-compatible polymer material avoids the corrosion issues of traditional metal temperature sensors in immersion environments, while the physically coupled piezoelectric conversion eliminates the risk of electronic component aging. The nonlinear response of the piezoelectric crystal array matches the dynamic requirements of temperature regulation, providing a higher signal gradient in the critical temperature range. The vertical stacking design of the thermosensitive polymer layer, micro-arm structure, and piezoelectric array meets the requirements of immersion liquid cooling systems for sensor miniaturization.

[0028] Example 3: Based on Example 2, the process of outputting an electrical signal having a nonlinear relationship with the deformation amount in step S103 provided in this embodiment of the present invention includes the following steps: Step S1031: applying vertical pressure to the contact interface of the stacked piezoelectric crystal array by displacing the multi-stage linkage micro-arm; Step S1032: When the stacked piezoelectric crystal array is under pressure, periodic lattice distortion occurs, and the distortion phase difference of adjacent crystal layers generates an interference field; the interference field forces the charge carrier pairs inside the crystal to be cooperatively deflected along the interference polar axis direction; Step S1033: After deflection, the dipoles of the charge carrier pairs form a quantum potential well channel that penetrates the crystal layer boundary. The quantum potential well channel induces the electron cloud to undergo energy level transition, triggering directional tunneling of charge carriers between adjacent crystal layers; the tunneling charges form an asymmetric accumulation at the edge electrodes of the array, and the output is an electrical signal whose intensity is exponentially correlated with the displacement.

[0029] In the above-mentioned embodiment, the vertical displacement of the multi-stage linkage micro-arms of this embodiment is directly converted into periodic pressure at the crystal interface, triggering dynamic distortion of the lattice structure; the distortion phase difference between adjacent crystal layers forms an interference field, achieving coupled conversion of mechanical energy to electric field energy; the coordinated deflection of the interference polar axis forces the dipoles to align in a directional manner, providing a controllable path for charge transport. The quantum potential well induced by the dipole arrangement reduces the interlayer potential barrier, promotes directional tunneling of carriers, and achieves efficient charge transmission; the electron cloud energy level transition enhances the tunneling probability, making the output current sensitive to small displacements; the asymmetric accumulation of tunneling charge at the edge electrodes forms an exponential correlation characteristic, and the current increases nonlinearly with increasing displacement.

[0030] In summary, the cascade effect of lattice distortion, interference field, and quantum tunneling in this embodiment converts micron-scale displacement into a wide-range electrical signal, covering detection requirements from slight temperature changes to severe thermal shock. The exponential input-output relationship provides a higher signal slope in critical temperature ranges (such as near the phase transition point), improving the system's response speed to thermal shock. Quantum tunneling is based on physical coupling rather than electronic circuits, avoiding signal drift caused by electromagnetic interference. The solid-state response mechanism of the stacked crystal array eliminates the wear of moving parts and is suitable for long-term operation in immersion liquid cooling environments. This exponential correlation between displacement and electrical signal strength provides a temperature control system with an input reference that combines high sensitivity with a wide dynamic range.

[0031] Example 4: Figure 4 As shown, based on Example 1, the process of generating the refrigerant flow control signal in step S200 provided in this embodiment of the present invention includes the following steps: Step S201: Acquire a set of electrical signals output by multiple temperature sensing packages, where each electrical signal corresponds to a temperature-time gradient at a specific location on a server motherboard; Step S202: The processor's built-in heterogeneous computing core maps the electrical signal group into a three-dimensional heat flow channel network according to physical location; based on the thermal conductivity characteristics of the fluorinated liquid and the coil layout, the superposition state heat flow is decoupled to obtain the transient heat flux density scalar of each channel, and the heat flux density distribution is generated; Step S203: Calculate the heat absorption equivalent required by each coil unit based on the heat flux density distribution and the two-phase coil phase change hysteresis parameter; convert the heat absorption equivalent into a refrigerant mass flow rate vector corresponding to the liquid distributor branch, and output it as a discrete control signal executable by the solenoid valve.

[0032] Among the above-mentioned embodiments, this embodiment achieves the collaborative processing of multi-source thermal data, improving the efficiency of thermal flow field modeling by parallel processing of multi-dimensional temperature signals through heterogeneous computing cores. Precise decoupling of spatial thermal loads, combining a three-dimensional heat flow channel network with a superposition decoupling algorithm, eliminates thermal crosstalk and locates critical cooling areas. Adaptive flow control, based on dynamic decision-making based on real-time heat flux density and phase change hysteresis parameters, ensures that refrigerant distribution is strictly matched to local heat loads. System response is optimized, and discrete control signals adapt to the execution characteristics of the solenoid valve, reducing adjustment delays and improving closed-loop control stability.

[0033] In summary, this embodiment converts discrete temperature signals into high-precision refrigerant distribution instructions through cascade processing of distributed temperature acquisition, heat flow network modeling, heat absorption demand decision-making, and flow control conversion, thereby realizing dynamic thermal balance control of the immersion liquid cooling system.

[0034] Example 5: Based on Example 4, the process of generating the heat flux density distribution in step S202 provided in this embodiment of the present invention includes the following steps: Step S2021: A group of electrical signals of temperature-time gradient, each electrical signal carrying the spatiotemporal thermal characteristics of a corresponding position of the server motherboard; Step S2022: Using each temperature-sensing package as a thermal vortex core, a free decay vortex program is established based on the low viscosity of the fluorinated liquid. The interference noise of the heat flow superposition state is eliminated by interlocking the rotational phases of adjacent vortex cores. The three-dimensional thermal potential energy gradient field continuously distributed within the server immersion domain is output. Step S2023: cutting the potential flow field along the boundary of the physical channel where the coil and the server intersect, performing energy flux conservation integration on each cutting unit, and converting the continuous potential field into a transient heat flux density scalar of an independent channel.

[0035] In the aforementioned embodiments, this embodiment achieves spatiotemporal correlation encoding of discrete temperature measurement data through a group of temperature-time gradient electrical signals, providing a spatiotemporal reference for subsequent thermal flow field reconstruction. The discrete signals collected by the temperature sensor array are processed by a phase interlocking algorithm and converted into thermal characteristic inputs with spatiotemporal continuity. A free-decaying vortex model, constructed based on the low viscosity of fluorinated liquid, eliminates the superposition noise of multiple heat sources through the phase interlocking mechanism of vortex core rotation, expanding the discrete temperature measurement points into a continuous three-dimensional thermal potential energy gradient field, achieving a meshless description of heat flow within the immersion domain. The potential flow field cutting units strictly correspond to the physical channel boundaries, and the continuous potential field is converted into a discretized heat flux density scalar through energy flux conservation integration. Maintaining strict constraints of the first law of thermodynamics ensures the physical authenticity of the energy transfer process at the channel level. The resulting transient heat flux density scalar output constitutes a complete description of the server system-level heat dissipation characteristics, including dual thermodynamic characteristic parameters: spatial distribution dimension (three-dimensional channel network) and temporal dimension (phase-locked timing).

[0036] Example 6: Based on Example 5, the process of outputting a discrete control signal executable by the solenoid valve in step S203 provided in this embodiment of the present invention includes the following steps: Step S2031: inputting the transient heat flux density scalar of each channel, where each transient heat flux density scalar corresponds to the spatiotemporal distribution of thermal energy in the service area of ​​a single coil; Step S2032: Based on the two-phase coil phase change hysteresis parameter and the response delay of the refrigerant from liquid to gas, a thermal inertia attenuation function is established; the transient heat flux density scalar is convolved with the phase change leader along the time axis to output a net heat absorption flux value that matches the coil phase change rate in real time; Step S2033: Based on the latent heat of phase change per unit mass of the refrigerant in the coil, the net heat absorption flux is converted into the instantaneous value of the refrigerant mass flow required to maintain thermal balance; according to the mapping relationship between the distributor branch and the coil unit, the flow values ​​are aggregated to generate a branch flow rate vector.

[0037] In the above embodiment, this embodiment accurately maps the spatiotemporal distribution of thermal energy in the coil service area by inputting a channel-level transient heat flux density scalar. Combined with the phase change hysteresis characteristics of the two-phase coil, the thermal inertia attenuation function is used to perform phase change leading convolution, so that the net heat absorption flux value is synchronized with the refrigerant phase change process in real time, eliminating the timing mismatch between heat flux fluctuations and phase change delays. Based on the refrigerant unit mass phase change latent heat parameter, the net heat absorption flux is converted into the instantaneous value of mass flow required to maintain thermal balance. This conversion process strictly follows the energy equivalence relationship between the phase change latent heat and the heat flux density, ensuring the energy conservation of the refrigerant flow control signal and the heat load change. Based on the mapping relationship between the distributor branch and the coil unit, the discretized mass flow value is aggregated into a branch flow rate vector. This vector directly corresponds to the executable discrete control signal of the solenoid valve array, realizing the precise distribution of the refrigerant flow in the spatial dimension, and adapting to the dynamic adjustment requirements of the non-uniform thermal field in the server immersion area. The final output branch flow rate vector constitutes a complete solenoid valve control instruction set. Its temporal resolution (phase change leader convolution timing) and spatial resolution (distributor branch mapping) jointly ensure the refrigeration system's rapid response capability to transient heat loads.

[0038] Example 7: Based on Example 6, the process of aggregating flow values ​​to generate branch flow rate vectors in step S2033 provided in this embodiment of the present invention includes the following steps: Step S20331: Continuously track the operating status of the compressor and the pressure changes within the evaporator, while taking into account the time delay required for the refrigerant to change from liquid to gas. Based on the real-time data of pressure changes and time delays, the actual heat exchange value currently required is obtained, and the amount of heat required in the future is estimated, ultimately providing a net heat absorption flux value for the adjusted heat removal requirement. Step S20332: Based on the unit mass phase change latent heat constant of the refrigerant at a specific saturation pressure, the net heat absorption flux value is divided by the unit mass phase change latent heat constant to obtain the refrigerant mass flow rate per unit time required for the current thermal equilibrium; the refrigerant mass flow rate per unit time required to maintain the current thermal equilibrium is output, and its value is linearly proportional to the heat absorption flux; Step S20333: Based on the spatial mapping topology of the distributor branch and the coil unit, which is determined by the parallel coil structure in the evaporator, the flow values ​​of multiple coil units belonging to the same distributor branch are scalared and superimposed; the branch flow rate vectors of each distributor outlet are output, and the vector dimensions match the solenoid valve execution port.

[0039] In the above embodiment, the net heat absorption flux value of this embodiment is converted into an instantaneous value of mass flow rate through the phase change latent heat constant per unit mass of the refrigerant after being corrected for the phase change hysteresis, thereby ensuring that the heat removal demand is strictly matched with the refrigerant flow rate and maintaining the linear proportional relationship of the energy transfer process. Based on the spatial mapping topology of the liquid distributor branch and the coil unit, the flow of the parallel coil unit is scalared and superimposed to realize the coupled calculation of the multi-branch heat load in the evaporator, ensuring that the flow distribution corresponds accurately to the spatial heat distribution. The dimension of the generated liquid distributor outlet flow rate vector is fully matched with the solenoid valve execution port, so that the flow control signal has direct drive capability and adapts to the dynamic adjustment requirements of the multi-branch parallel structure of the evaporator. The final output flow rate vector constitutes a standardized interface executed by the solenoid valve array. Its numerical accuracy and topological logic together ensure the real-time response capability of the refrigeration system to non-uniform thermal fields.

[0040] Example 8: Based on Example 7, the process of scalaring and superimposing the flow values ​​of multiple coil units belonging to the same liquid distributor branch in step S20333 provided in this embodiment of the present invention includes the following steps: Step S203331: Output a set of refrigerant mass flow values ​​per unit time, each value corresponding to the real-time mass requirement of a single coil unit to maintain thermal balance; Step S203332: Based on the hard-connected mapping relationship between the distributor branches and the coil units, the physical piping arrangement of the parallel coils in the evaporator is uniquely determined; the mass flow rates of all coil units under the same distributor branch are subjected to an unweighted scalar algebraic sum operation, and the total mass flow rate value of each branch is output, satisfying the law of fluid continuity; Step S203333: Arrange the total mass flow values ​​of each branch according to the inherent spatial sequence of the dispenser outlet, and generate a discrete flow rate instruction sequence that is completely isomorphic to the number of dispenser outlets and the solenoid valve execution ports.

[0041] In the above embodiment, this embodiment realizes the flow coupling calculation of the parallel coil system by performing scalar algebraic sum operations on the real-time mass demand flow of a single coil unit; it follows the law of fluid continuity to ensure that the total flow of the distributor branch is accurately matched with the downstream pipeline delivery capacity. Based on the hard-connected mapping relationship between the distributor branch and the coil unit, the logical consistency of the physical pipeline arrangement inside the evaporator is maintained; the calculation process does not introduce weight correction, and the fluid distribution characteristics of the original pipeline structure are completely retained. The flow rate instruction sequence generated according to the inherent spatial sequence of the distributor outlet forms a strict isomorphic relationship with the solenoid valve control port of the execution terminal; this mapping ensures that the control signal can directly drive the actuator without the need for additional protocol conversion or signal processing. The final output discrete flow rate instruction sequence not only meets the local thermal balance requirements, but also ensures the conservation of the total mass flow of the entire refrigeration circuit through scalar superposition operations, providing a basic guarantee for system-level energy balance.

[0042] Example 9: Based on Example 8, the process of outputting the total mass flow value of each branch in step S203332 provided in this embodiment of the present invention includes the following steps: Step S2033321: The outlet flow channels of all coil units under the same liquid distributor branch complete physical confluence at the main pipe section of the liquid distributor branch; Step S2033322: The mass flow of each coil unit is automatically distributed on the main pipe cross section according to the flow channel geometric projection ratio; the surface integral is performed on the flux density of the main pipe cross section, and a scalar integral value is output; Step S2033323: The scalar integral value is directly equivalent to the steady-state output total mass flux of the liquid distributor branch according to the fluid continuity law; the total mass flux forms a dynamic balance with the suction characteristics of the downstream compressor.

[0043] In the above embodiment, this embodiment establishes the geometric boundary conditions for multi-channel fluid mixing through the physical confluence process of the coil unit outlet flow channel in the main pipe of the distributor branch, ensuring that the flux density distribution of the confluence section is consistent with the real flow field characteristics. Based on the flux density distribution on the main pipe cross section, the surface integral operation is performed to convert the output flow of the discrete coil unit into a scalar integral value of continuous medium mechanics, thereby realizing the mathematical conversion of microscopic flow field characteristics to macroscopic flow parameters. The cross-sectional integral value is directly mapped to the total mass flux of the distributor branch through the law of fluid continuity, ensuring the mass conservation characteristics of the confluence process, and at the same time establishing a dynamic equilibrium relationship with the suction characteristics of the downstream compressor. The total mass flux of the output serves as a key parameter connecting the evaporator branch and the compressor subsystem, providing accurate boundary conditions for the global mass flow balance of the refrigeration cycle.

[0044] Example 10: Based on Example 9, the process of outputting the scalar integral value in step S2033322 provided in this embodiment of the present invention includes the following steps: Step S20333221: The mass flow rate of each coil unit on the manifold cross section of the distributor branch is automatically decomposed into normal components perpendicular to the pipe wall based on the spatial orientation angle between its outlet flow channel and the axis of the manifold. All normal components are continuously distributed on the manifold cross section according to the position coordinates to form a flux density field. Step S20333222: perform flux conservation integration on the flux density field along the entire main pipe cross section; the integral output value is equivalent to the total mass flux scalar of the cross section.

[0045] In the above embodiment, this embodiment decomposes the mass flow of each coil unit into normal components perpendicular to the pipe wall based on the spatial orientation angle between the outlet flow channel and the axis of the main pipe, and establishes a vector field expression that strictly matches the geometric structure of the main pipe. Through the continuous spatial distribution of the normal components, a flux density field on the main pipe cross section is formed, and the contribution characteristics of each coil unit to the total mass flow are fully retained, realizing the mathematical mapping of discrete flow channels to continuous flow fields. A full domain integration is performed on the flux density field along the main pipe cross section to ensure that the local flux distribution strictly follows the principle of mass conservation, and the output result directly represents the steady-state scalar value of the total mass flux of the cross section. The total mass flux scalar output by the integration constitutes the key coupling parameter of the evaporator branch and the downstream system, providing an accurate physical benchmark for the dynamic balance of the refrigerant flow.

[0046] Example 11: Figure 5 As shown, based on Example 10, the process of implementing the control of the refrigerant flow in step S300 provided in this embodiment of the present invention includes the following steps: Step S301: The solenoid valve coil current is controlled by a discrete control signal, which excites an axial confined magnetic field in the ferrite core. The magnetic field gradient forces the permanent magnet valve core to undergo millimeter-level precision displacement along the axial direction, and the displacement is rigidly linearly mapped to the current intensity. Step S302: The displacement of the valve core is synchronously transmitted to the inlets of each branch of the liquid distributor through the conical push rod mechanism. The push rod conical surface and the branch valve seat form an annular contraction flow channel. For every unit increase in the displacement, the cross-sectional area of ​​the flow channel decreases according to the inverse square law. Step S303: The contraction flow channel generates a viscous throttling effect on the refrigerant fluid, the throttling intensity is positively correlated with the cross-sectional area attenuation, and a physical correction value of the branch mass flux is output; Step S304: The branch mass flux is distributed to the parallel coils via the liquid distributor. Changes in the branch mass flux directly change the refrigerant phase change latent heat exchange rate in the coils. The exchange rate is dynamically matched with the server heat generation rate, so that the fluorinated liquid temperature in the immersion area converges to the set threshold.

[0047] In the above embodiment, this embodiment realizes millimeter-level linear conversion of electrical signals to valve core displacement through the axial confinement magnetic field excited by the solenoid valve coil current in the ferrite core, and establishes a deterministic mapping relationship between the control signal and the mechanical action. The axial displacement of the valve core is transmitted through the conical push rod mechanism, forming the inverse square law area attenuation characteristic of the annular contraction flow channel, realizing the amplification and control effect of the micro-displacement on the flow cross-sectional area. The viscous throttling effect generated by the contraction flow channel and the cross-sectional area attenuation form a positive feedback relationship, converting the mechanical displacement into a quantifiable mass flux correction value. The branch mass flux change is distributed through the liquid distributor, directly modulating the phase change latent heat exchange rate in the parallel coil, forming a closed-loop temperature control capability that adaptively matches the heat generation rate of the server. The solenoid valve current-displacement-flow channel cross-sectional area-mass flux-heat exchange rate constitutes a five-level series control chain, realizing the complete transmission of electrical signals to thermodynamic parameters, and ultimately achieving stable control of the immersion domain temperature.

[0048] Example 12: Figure 6-Figure 9 As shown, based on Examples 1 to 11, the intelligent constant temperature single-phase immersion data center composite liquid cooling control system provided by the embodiments of the present invention includes: a coil 1, a server 2, a liquid distributor 3, a gas collecting pipe 4, a compressor 5, a condenser 6, a liquid storage tank 7, a ball valve 8, a pump 9, a one-way valve 10, a data integration processor 11, a server motherboard 12, a two-phase liquid cooling coil 13, a solenoid valve 14, a temperature sensor 15, and an evaporator 16.

[0049] Among them, multiple groups of coils 1 and servers 2 are arranged in the evaporator 16, the coils 1 are connected to the gas collecting pipe 4, the servers 2 are arranged between adjacent coils 1, and the liquid separator 3 is arranged at the coolant inlet end of the lower left end of the evaporator 16; the outlet end of the gas collecting pipe 4 is connected to the inlet end of the compressor 5 through a pipeline, the outlet end of the compressor 5 is connected to the inlet end of the condenser 6 through a pipeline, the outlet end of the condenser 6 is connected to the inlet end of the liquid storage tank 7 through a pipeline, the outlet end of the liquid storage tank 7 is connected to one end of the ball valve 8 through a pipeline, the other end of the ball valve 8 is connected to one end of the pump 9 through a pipeline, the other end of the pump 9 is connected to one end of the one-way valve 10 through a pipeline, and the other end of the one-way valve 10 is connected to the inlet end of the liquid separator 3 through a pipeline; The server 2 is composed of a server motherboard 12, which is installed between adjacent coils 1. The coils 1 are composed of multiple two-phase liquid cooling coils 13 connected together; The data integration processor 11 is connected to the electromagnetic valve 14 through electrical signals, and the electromagnetic valve 14 is connected to the liquid dispenser 3 ; the data integration processor 11 is connected to multiple temperature sensing packages 15 through electrical signals, and the temperature sensing packages 15 are installed on both sides of the server motherboard 12 .

[0050] In the above embodiment, the cooling cycle process of this embodiment is as follows: the liquid coolant is distributed through the liquid separator 3 and enters the coil 1 in the evaporator 16; the coolant absorbs the heat generated by the server 2 and then vaporizes, and the vaporized medium is collected through the gas collecting pipe 4; the high-temperature gaseous medium is pressurized by the compressor 5 and then enters the condenser 6; the heat is released in the condenser 6 and condensed into liquid; the liquid medium is stored in the liquid storage tank 7, and is transported by the pump 9 after the flow is adjusted by the ball valve 8; the backflow is prevented by the one-way valve 10, and finally returns to the liquid separator 3 to complete the cycle.

[0051] Temperature Control System: A temperature sensor 15 monitors the temperature of the server motherboard 12 in real time. The data integration processor 11 receives the temperature signal and controls the solenoid valve 14, which adjusts the coolant distribution from the liquid distributor 3. Server Cooling Structure: The server motherboard 12 is installed between two parallel liquid cooling coils 13; these two-phase liquid cooling coils 13 directly contact the heat-generating components for heat exchange.

[0052] This embodiment achieves precise temperature control of server components, improves heat dissipation efficiency through immersion cooling, and significantly improves the heat transfer coefficient through two-phase heat exchange; the one-way valve 10 prevents backflow of the medium, the ball valve 8 provides flow regulation guarantee, and the multi-loop design enhances system redundancy; multiple groups of coils 1 in the evaporator 16 are connected in parallel to increase the heat exchange area, the compressor 5-condenser 6 system realizes heat energy recovery, and the precise control of the solenoid valve 14 reduces energy consumption; the symmetrical arrangement of the liquid distributor 3 and the gas collecting pipe 4, the compact arrangement of the server 2 and the coil 1, and the centralized control of the data integration processor 11; through the above structure and control method, efficient cooling and precise temperature control of data center equipment are achieved.

[0053] This embodiment is an intelligent temperature control system composite cooling system that combines single-phase immersion liquid cooling and two-phase cooling coils; it mainly consists of a box composed of multiple groups of servers 2, a data integration processor 11, a condenser 6, a compressor 5, a ball valve 8, a one-way valve 10, a liquid storage tank 7 and a pump 9. Figure 6 The box body of the server 2 is composed of a server motherboard 12, a two-phase liquid cooling coil 13, a liquid distributor 3, and an air collecting pipe 4. Figure 7 As shown, server 2 generates a significant amount of heat, raising the coolant temperature. This heat is then transferred through two-phase liquid cooling coil 13 to the refrigerant liquid within coil 1. The refrigerant absorbs heat and undergoes a phase change, removing a significant amount of heat. The refrigerant gas is compressed, condensed, and returns to storage tank 7, where it reenters liquid separator 3, completing the cycle.

[0054] The server 2 is completely immersed in the single-phase coolant fluorinated liquid, and the heat of the server 2 is quickly transferred to the fluorinated liquid, and the temperature of the server 2 is reduced; the temperature sensing package 15 is distributed around the server motherboard 12. Figure 8As shown, when the data center's heat generation increases, temperature sensor 15 generates an electrical signal that is fed back to data integration processor 11. Upon receiving the signal, the internal processor in data integration processor 11 calculates the required refrigerant flow rate according to the settings and provides feedback. The signal from data integration processor 11 is received by solenoid valve 14, precisely regulating the flow of refrigerant into liquid distributor 3.

[0055] Structurally, this embodiment adopts single-phase immersion liquid cooling to avoid uneven temperature distribution caused by two-phase phase change; the cooling coil 1 is directly in contact with the single-phase immersion liquid fluorinated liquid to avoid the decrease in heat dissipation rate caused by multi-structure heat conduction, and the composite cooling response speed is faster.

[0056] This embodiment utilizes a design that combines single-phase immersion and two-phase cooling coils. The heating element is completely immersed in the single-phase coolant, the fluorinated liquid. Heat is transferred to the coolant through heat conduction and convection, raising the temperature of the fluorinated liquid (heat absorption solely through temperature change, i.e., sensible heat). The heat from the coolant, after absorbing the heat, is then absorbed by the low-temperature refrigerant in coil 1 through heat conduction and convection. The refrigerant in coil 1 then absorbs the heat through phase change, thus completing the heat transfer process from the heating element to the fluorinated liquid and then to the refrigerant.

[0057] Data centers generate a lot of heat. Temperature sensor 15 detects temperature changes and generates an electrical signal to the data center processor. The processor then generates an electrical signal, which controls the flow of refrigerant in coil 1 via solenoid valve 14. The refrigerant in coil 1 quickly absorbs heat, changing phase from liquid to gas, absorbing and dissipating a large amount of heat. This rapid response keeps the temperature of the single-phase fluorinated immersion liquid within a stable range as heat generation from server 2 increases. The gaseous refrigerant in coil 1 is collected by manifold 4, converted to a room-temperature liquid by compressor 5 and condenser 6, and then redistributed into storage tank 7, completing the cycle.

[0058] The server body 12 is directly immersed in the single-phase coolant. The interior of the enclosure is filled with the single-phase coolant, a fluorinated liquid that rapidly transfers heat to the coolant coil 1. This single-phase immersion combined with the intelligently controlled two-phase liquid cooling coil 13 keeps the entire server 2 system within the optimal operating temperature range, preventing uneven temperatures among heat-generating modules and temperature rises caused by excessive heat generation.

[0059] Compared to traditional air-cooling processes, this embodiment has a lower PUE value. Single-phase immersion cooling has lower requirements for the server cabinet and the amount of fluorinated liquid used. It has a simple structure and is easy to operate. The server in the computer room is safer, easier to operate, and has low maintenance costs. In addition, the entire device is quieter and smaller than air-cooling equipment, saving on computer room construction and maintenance costs. The entire design chooses direct contact between the coil and the single-phase immersion liquid, which can quickly conduct heat. The temperature sensor, data integration processor, and solenoid valve work together to effectively deal with rapid heat generation. Only the refrigerant flow can be adjusted to ensure the temperature balance of the heat generation unit and keep the system within the optimal operating range.

[0060] Figure 10 A block diagram is shown of an exemplary electronic device suitable for implementing embodiments of the present invention.

[0061] The electronic device may include a central processing unit / microprocessor / main control chip, etc. 17; a storage medium 18, coupled to the central processing unit / microprocessor / main control chip, etc. 17, and storing computer executable instructions therein for performing the steps of each method of an embodiment of the present invention when executed by the processor.

[0062] The central processing unit / microprocessor / main control chip 17 may include but is not limited to one or more processors or microprocessors.

[0063] The storage medium 18 may include, but is not limited to, for example, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, computer storage media (such as hard disk, floppy disk, solid-state drive, removable disk, CD-ROM, DVD-ROM, Blu-ray disc, etc.).

[0064] In addition, the electronic device may also include (but not limited to) a data bus 19, an input / output bus / external bus / device bus 20, a display 21, and input / output devices 22 (for example, a keyboard, a mouse, a speaker, etc.).

[0065] The central processing unit / microprocessor / main control chip etc. 17 can communicate with external devices ( 21 , 22 etc.) via an I / O bus 20 via a wired or wireless network (not shown).

[0066] The storage medium 18 may also store at least one computer executable instruction for executing the various functions and / or method steps in the embodiments described in this technology when executed by the central processing unit / microprocessor / main control chip 17 .

[0067] In one embodiment, the at least one computer executable instruction may also be compiled into or constitute a software product, wherein one or more computer executable instructions are executed by a processor to perform the various functions and / or method steps in the embodiments described in the present technology.

[0068] Figure 11 A schematic diagram of a computer-readable storage medium according to an embodiment of the present invention is shown.

[0069] like Figure 11 As shown, instructions are stored on the non-transitory computer-readable storage medium 24, and the instructions are, for example, computer-readable instructions 23. When the computer-readable instructions 23 are executed by the processor, the various methods described above can be executed. Non-transitory computer-readable storage media include, but are not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. Non-transitory non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, the non-transitory computer-readable storage medium 24 can be connected to a computing device such as a computer, and then, when the computing device executes the computer-readable instructions 23 stored on the computer-readable storage medium 24, the various methods described above can be performed.

[0070] In the several embodiments provided herein, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of units is merely a logical functional division. In actual implementation, there may be other division methods, such as multiple units or components being combined or integrated into another system, or some features being ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interface, or the indirect coupling or communication connection of the system or unit may be electrical, mechanical, or other forms.

[0071] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0072] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0073] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the various embodiments of the method of the present invention via a computer device (which can be a personal computer, server, or network device, etc.). The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0074] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent constant temperature single-phase immersion data center composite liquid cooling control method, characterized in that: The following steps are involved: After the data integration processor receives the electrical signal converted from the temperature data of the server motherboard, the internal processor calculates according to the preset algorithm and generates a refrigerant flow control signal. The refrigerant flow control signal is output as a regulation instruction to the solenoid valve to guide its operation. The process of generating a refrigerant flow control signal includes the following steps: Obtain a collection of electrical signals output by multiple temperature sensors, each of which corresponds to a temperature-time gradient at a specific location on the server motherboard. The processor's built-in heterogeneous computing core maps electrical signal groups into a three-dimensional heat flow channel network based on physical location. Based on the thermal conductivity characteristics of the fluorinated liquid and the coil layout, the superposition heat flow is decoupled to obtain the transient heat flux density scalar of each channel and generate the heat flux density distribution. The required heat absorption equivalent of each coil unit is calculated based on the heat flux density distribution and the phase change hysteresis parameter of the two-phase coil. The heat absorption equivalent is converted into the refrigerant mass flow rate vector of the corresponding distributor branch and output as a discrete control signal executable by the solenoid valve.

2. The intelligent constant temperature single-phase immersion data center composite liquid cooling control method according to claim 1, characterized in that: The process of outputting a discrete control signal that can be executed by the solenoid valve includes the following steps: Input the transient heat flux density scalar of each channel. Each transient heat flux density scalar corresponds to the spatiotemporal distribution of thermal energy in the service area of ​​a single coil. Based on the two-phase coil phase change hysteresis parameter and the response delay of the refrigerant from liquid to gas, a thermal inertia decay function is established; the transient heat flux density scalar is convolved with the phase change leader along the time axis to output a net heat absorption flux value that matches the coil phase change rate in real time; Based on the latent heat of phase change per unit mass of the refrigerant in the coil, the net heat absorption flux is converted into the instantaneous value of the refrigerant mass flow rate required to maintain thermal balance; according to the mapping relationship between the distributor branch and the coil unit, the flow values ​​are aggregated to generate the branch flow rate vector.

3. The intelligent constant temperature single-phase immersion data center composite liquid cooling control method according to claim 2, characterized in that: The process of aggregating flow values ​​to generate branch flow rate vectors includes the following steps: Continuously track the operating status of the compressor and the pressure changes inside the evaporator, while taking into account the time delay required for the refrigerant to change from liquid to gas. Based on the real-time data of pressure changes and time delays, the actual heat exchange value currently required is obtained, and the amount of heat required in the future is estimated, ultimately providing a net heat absorption flux value for the adjusted heat removal requirement. Based on the unit mass phase change latent heat constant of the refrigerant at a specific saturation pressure, the net heat absorption flux value is divided by the unit mass phase change latent heat constant to obtain the refrigerant mass flow rate per unit time required for the current heat balance; Output the refrigerant mass flow rate per unit time required to maintain the current thermal balance, and its value is linearly proportional to the heat absorption flux; Based on the spatial mapping topology of the distributor branch and the coil unit, which is determined by the parallel coil structure in the evaporator, the flow values ​​of multiple coil units belonging to the same distributor branch are scalared and superimposed. The branch flow rate vectors of each distributor outlet are output, and the vector dimensions match the solenoid valve execution port.

4. The intelligent constant temperature single-phase immersion data center composite liquid cooling control method according to claim 3, characterized in that: The process of scalaring and superimposing the flow values ​​of multiple coil units belonging to the same distributor branch includes the following steps: A collection of refrigerant mass flow values ​​per unit time output, each value corresponding to the real-time mass requirement of a single coil unit to maintain thermal balance; Based on the hard-connected mapping relationship between the distributor branch and the coil unit, the physical piping arrangement of the parallel coils in the evaporator is uniquely determined. The mass flow of all coil units under the same distributor branch is calculated by performing an unweighted scalar algebraic sum operation, and the total mass flow of each branch is output to satisfy the law of fluid continuity. The total mass flow values ​​of each branch are arranged according to the inherent spatial sequence of the dispenser outlet, and a discrete flow rate instruction sequence that is completely isomorphic to the number of dispenser outlets and the solenoid valve execution ports is generated.

5. The intelligent constant temperature single-phase immersion data center composite liquid cooling control method according to claim 4, characterized in that: The process of outputting the total mass flow value of each branch includes the following steps: The outlet flow passages of all coil units under the same distributor branch complete physical convergence at the distributor branch main pipe section; The mass flow of each coil unit automatically forms a flux density distribution on the main pipe cross section according to the flow channel geometric projection ratio; the flux density of the main pipe cross section is surface integrated and the scalar integral value is output; The scalar integral value is directly equivalent to the steady-state output total mass flux of the distributor branch according to the fluid continuity law; the total mass flux forms a dynamic balance with the suction characteristics of the downstream compressor.

6. The intelligent constant temperature single-phase immersion data center composite liquid cooling control method according to claim 5, characterized in that: The process of outputting the scalar integral value includes the following steps: The mass flow of each coil unit on the main pipe cross section of the distributor branch is automatically decomposed into normal components perpendicular to the pipe wall based on the spatial orientation angle between its outlet flow channel and the main pipe axis. All normal components are continuously distributed on the main pipe cross section according to the position coordinates to form a flux density field. The flux conservation integral is performed on the flux density field over the entire cross section of the main pipe; the output of the integral is equivalent to the total mass flux scalar of the cross section.

7. The intelligent constant temperature single-phase immersion data center composite liquid cooling control method according to claim 1, characterized in that: The temperature sensor package monitors the temperature changes of the server motherboard in real time, converts the temperature data into an electrical signal, and transmits the electrical signal as an input signal to the data integration processor.

8. The intelligent constant temperature single-phase immersion data center composite liquid cooling control method according to claim 1, characterized in that: The solenoid valve adjusts the refrigerant distribution amount of the distributor according to the refrigerant flow control signal to realize the regulation of the refrigerant flow; the adjusted refrigerant flow acts on the phase change heat absorption process of the refrigerant in the coil to stabilize the temperature of the fluorinated liquid.

9. An intelligent constant temperature single-phase immersion data center composite liquid cooling control system, implementing the intelligent constant temperature single-phase immersion data center composite liquid cooling control method according to any one of claims 1 to 8, characterized in that: Includes: coil, server, liquid distributor, gas collecting pipe, compressor, condenser, liquid storage tank, ball valve, pump, check valve, data integration processor, server motherboard, two-phase liquid cooling coil, solenoid valve, temperature sensor, evaporator; Among them, multiple groups of coils and servers are arranged in the evaporator, the coils are connected to the gas collecting pipe, the servers are arranged between adjacent coils, and the liquid separator is arranged at the coolant inlet end of the lower left end of the evaporator; the outlet end of the gas collecting pipe is connected to the inlet end of the compressor through a pipeline, the outlet end of the compressor is connected to the inlet end of the condenser through a pipeline, the outlet end of the condenser is connected to the inlet end of the liquid storage tank through a pipeline, the outlet end of the liquid storage tank is connected to one end of the ball valve through a pipeline, the other end of the ball valve is connected to one end of the pump through a pipeline, the other end of the pump is connected to one end of the one-way valve through a pipeline, and the other end of the one-way valve is connected to the inlet end of the liquid separator through a pipeline; The server consists of a server motherboard, which is installed between adjacent coils. The coils are made up of multiple two-phase liquid cooling coils connected together. The data integration processor is connected to the electromagnetic valve through electrical signals, and the electromagnetic valve is connected to the dispenser; the data integration processor is connected to multiple temperature sensing packages through electrical signals, and the temperature sensing packages are installed on both sides of the server motherboard.

Images