Anti-clogging water circulation radiator

Abstract

The application relates to the field of heat dissipation, in particular to an anti-blocking water circulation radiator, which comprises a radiator body, a filter screen structure, a temperature sensor array and other components arranged in the body, the radiator body is provided with a hot water outlet and a cold water outlet, a specific control method is adopted to perform heat dissipation, data is collected through the temperature sensor array, a model is established to calculate heat dissipation efficiency, and a phase change material heat storage module, a fault tolerance mechanism and the like are further arranged. The application achieves the technical effects that heat dissipation conditions of the radiator can be monitored and regulated in real time, blocking can be effectively prevented, working parameters of a micro-channel pump group can be dynamically adjusted according to a heat deposition accumulation trend, heat can be temporarily stored and directionally forced to dissipate when local temperature rises too fast, the radiator simultaneously has a fault tolerance capability, and stable operation of the radiator is guaranteed.

Description

A clog-resistant water circulation radiator Technical Field

[0001] This application relates to the field of heat dissipation technology, and in particular to an anti-clogging water circulation radiator. Background Technology

[0002] In today's era of rapid technological advancement, data centers, central processing unit (CPU) cabinets, and large-scale equipment play crucial roles across various fields. These devices generate significant amounts of heat during operation, making effective heat dissipation essential for ensuring stable operation and extending equipment lifespan. Efficient and reliable heat dissipation technologies can improve equipment performance, reduce malfunctions and downtime caused by overheating, thereby lowering operating costs and driving the sustainable development of related industries. As equipment performance continues to improve, the demands on heat dissipation technology are also increasing, prompting researchers to continuously explore more advanced heat dissipation solutions.

[0003] In the past, the industry has conventionally employed various methods to address the heat dissipation issues of similar equipment. The most common is traditional air cooling, which uses fans to accelerate airflow and carry away the heat generated by the equipment. This method is simple, direct, and relatively low-cost, making it widely used. Another method is water cooling, which utilizes the high specific heat capacity of water, using the circulation of coolant within the radiator to absorb and transfer heat; this method offers better heat dissipation than air cooling. Additionally, some large equipment employs heat pipe cooling technology, which utilizes the phase change of the working fluid within the heat pipe to achieve efficient heat transfer. These traditional heat dissipation methods, to a certain extent, meet the equipment's heat dissipation needs.

[0004] However, existing heat dissipation technologies have significant drawbacks. Particularly in water-cooled systems, when the liquid circulates internally, impurities and sediments in the water can easily enter the pump with the water flow, clogging the microchannels. Once the microchannels are blocked, the circulation of the coolant is hindered, heat dissipation efficiency is greatly reduced, seriously affecting the normal operation of the equipment, and may even lead to equipment damage due to overheating. Summary of the Invention

[0005] The purpose of this application is to overcome the above-mentioned technical problems and provide an anti-clogging water circulation radiator.

[0006] A clog-resistant water circulation radiator includes a radiator body, which incorporates a filter structure, a temperature sensor array, an ambient temperature and humidity sensor, a microchannel pump assembly, an infrared thermal imager, and a vortex cooling device. The radiator body has hot water and cold water inlets. The filter structure is located inside the water outlet and adjacent to the inlet of the microchannel pump assembly. The filter structure can be made of stainless steel, aluminum, or plastic. The temperature sensor array is evenly distributed on the surface of the microchannels within the radiator body. The ambient temperature and humidity sensor is installed at the air inlet on the outside of the radiator body. The microchannel pump assembly is connected to the cold water inlet of the radiator body via a flange. The infrared thermal imager is fixed to the top monitoring area of ​​the radiator body. The vortex cooling device is integrated downstream of the hot water inlet of the radiator body. The temperature sensor array and the ambient temperature and humidity sensor are connected to an external central controller via a CAN bus. The microchannel pump assembly and the vortex cooling device each receive PWM control signals, forming a closed-loop control system.

[0007] The anti-clogging water circulation radiator uses the following control methods to perform heat dissipation:

[0008] S1. Real-time acquisition of multi-dimensional thermal field data on the surface of the radiator through a distributed temperature sensor array, combined with environmental temperature and humidity sensors to construct a three-dimensional thermal conduction feature matrix;

[0009] S2. Establish a dynamic thermal resistance model based on the unsteady heat conduction equation, calculate the heat dissipation efficiency attenuation gradient according to the characteristic matrix, and predict the local heat sink accumulation trend;

[0010] S3. Computational fluid dynamics simulation is used to generate an optimized flow velocity distribution map of the microchannel array on the surface of the heat sink, and the priority control area is determined in combination with the heat sink accumulation trend;

[0011] S4. Generate multi-level frequency conversion control signals based on fuzzy neural network algorithm, and dynamically adjust the operating parameters of the microchannel pump group in the priority control region;

[0012] S5. Monitor the changes in the surface temperature field of the radiator in real time using an infrared thermal imager, and use the feedback to correct the boundary condition parameters of the dynamic thermal resistance model;

[0013] S6. When the local temperature rise rate is detected to exceed the preset threshold, the phase change material heat storage module is activated to temporarily store heat, and the eddy current heat dissipation device is started simultaneously to implement directional forced heat dissipation.

[0014] By adopting the above technical solutions, the filter structure can prevent impurities and sediments from entering the water pump and clogging the microchannels; the temperature sensor array can collect the surface temperature data of the microchannels in real time, and the ambient temperature and humidity sensor can obtain the external ambient temperature and humidity. The two are combined to construct a three-dimensional heat conduction feature matrix; the microchannel pump group provides power for the circulation of coolant; the infrared thermal imager monitors the surface temperature field of the radiator in real time; the vortex cooling device enhances the heat dissipation effect; based on a series of control methods, the working parameters of the microchannel pump group can be precisely adjusted according to the thermal state of the radiator to improve the heat dissipation efficiency. It can also temporarily store heat through the phase change material heat storage module and promptly activate the vortex cooling device to deal with local excessive temperature rise.

[0015] Preferably, in step S1, the distributed temperature sensor array adopts a honeycomb topology layout, and the sensor spacing d satisfies d=(Q / (kΔT))^(1 / 3), where Q is the rated power of the heat sink (unit: W), k is the thermal conductivity of the substrate material (unit: W / m·K), and ΔT is the system's allowable temperature rise threshold (unit: °C).

[0016] By adopting the above technical solution, the distributed temperature sensor array with a honeycomb topology layout can more accurately collect multi-dimensional thermal field data of the radiator surface. The sensor spacing determined by a specific formula can make the sensor arrangement more reasonable. In addition, combined with the ambient temperature and humidity sensor, a three-dimensional heat conduction feature matrix can be constructed more accurately, providing a reliable data foundation for subsequent calculation and prediction, and improving the heat dissipation control accuracy and efficiency of the anti-clogging water circulation radiator.

[0017] Preferably, in step S2, the dynamic thermal resistance model is solved using the finite element method, and the heat flux density correction coefficient α is obtained through training a convolutional neural network. The training objective function is L=||T sim -T real ||²+λ|∇α|, where λ is the regularization coefficient (ranging from 0.01 to 0.1), T sim To simulate the temperature value, T real These are the measured temperature values.

[0018] By adopting the above technical solution, the dynamic thermal resistance model is solved by discretization using the finite element method, and the heat flux density correction coefficient is obtained by training a convolutional neural network. This allows for accurate calculation of the heat dissipation efficiency attenuation gradient, more precise prediction of local heat sink accumulation trends, and improved accuracy and reliability of the dynamic thermal resistance model.

[0019] Preferably, the generation of the optimized velocity distribution map in step S3 includes:

[0020] a) Establish an equivalent flow resistance network model for the microchannel array;

[0021] b) Using genetic algorithms to solve multi-objective optimization problems:

[0022] min[Σ(Ti-Topt)² + βΣvi³];

[0023] Where vi is the flow velocity of each microchannel (unit: m / s), Topt is the optimal temperature distribution (unit: ℃), and β is the energy consumption weighting coefficient (value range: 0.05-0.2).

[0024] By adopting the above technical solutions, the equivalent flow resistance network model of the microchannel array can be established, which facilitates the analysis of the flow characteristics in the microchannel. By using a genetic algorithm to solve the multi-objective optimization problem, the optimized flow velocity distribution map of the microchannel array on the radiator surface can be generated under the premise of considering heat dissipation effect and energy consumption. Combined with the heat sink accumulation trend, the priority control area can be determined to achieve efficient and energy-saving heat dissipation control.

[0025] Preferably, the fuzzy neural network in step S4 comprises a 5-layer structure:

[0026] Input layer: heat sink accumulation rate, ambient temperature and humidity, historical control records;

[0027] Fuzzification layer: Three fuzziness levels are defined using Gaussian membership functions;

[0028] Rule base: Stores 27 control rules generated from expert experience;

[0029] Inference layer: Fuzzy inference is performed using the Mamdani method;

[0030] Defuzzification layer: The centroid method is used to output the frequency conversion control signal.

[0031] By adopting the above technical solution, based on the structure including the radiator body, filter structure, temperature sensor array, ambient temperature and humidity sensor, microchannel pump group, infrared thermal imager, and vortex heat dissipation device, and with hot water inlet and cold water inlet set, and the filter structure located inside the water drain and adjacent to the inlet of the microchannel pump group, the five-layer structure of the fuzzy neural network can dynamically adjust the working parameters of the microchannel pump group in the priority control area according to the heat sink accumulation rate, ambient temperature and humidity, and historical control records. After dividing the fuzzy level by Gaussian membership function, applying expert experience rules, Mamdani method inference, and centroid method output, the working parameters of the microchannel pump group in the priority control area can be dynamically adjusted to achieve precise and intelligent control of radiator heat dissipation.

[0032] Preferably, the boundary condition correction in step S5 includes:

[0033] a) Extract the isotherm curvature feature C=|∇²T| (unit: 1 / m²) from thermal imaging data;

[0034] b) When C > 0.8C max Time (C) max To achieve the maximum curvature, the contact thermal resistance parameters of the thermal resistance model are automatically adjusted.

[0035] c) Introduce a Kalman filter to calibrate the model parameters in real time (sampling period 50ms).

[0036] By adopting the above technical solution, the curvature characteristics of isotherms can be extracted using thermal imaging data, which can accurately identify abnormal heat distribution. When the curvature characteristics exceed the maximum curvature of the design by a certain proportion, the contact thermal resistance parameters are automatically adjusted, which can correct the thermal resistance model in a timely manner. The introduction of a Kalman filter to calibrate the model parameters in real time ensures the accuracy of the boundary condition parameters of the dynamic thermal resistance model, enabling the heat sink to better adapt to the actual heat dissipation situation and improve the accuracy and stability of heat dissipation control.

[0037] Preferably, the phase change material thermal storage module in step S6 includes a graded activation mechanism:

[0038] Primary thermal storage: Paraffin-based phase change materials undergo phase change in the 45-50℃ range (latent heat value 180-200kJ / kg).

[0039] Secondary thermal storage: Metal-organic framework materials adsorb and release heat in the 55-60℃ range (specific surface area 3000m² / g);

[0040] Three-stage thermal storage: Nanoencapsulated salt water solution undergoes phase separation at temperatures above 65°C (particle size distribution 50-200nm).

[0041] By adopting the above technical solution, when the local temperature rise rate is detected to exceed the preset threshold, the phase change material thermal storage module can perform staged thermal storage. The paraffin-based phase change material can store a certain amount of heat during phase change in the range of 45-50℃, the metal-organic framework material can further store heat by adsorbing and releasing heat in the range of 55-60℃, and the nano-encapsulated salt solution can continue to temporarily store heat by undergoing phase separation above 65℃. At the same time, the vortex heat dissipation device can be used to implement directional forced heat dissipation, which can effectively deal with the problem of excessively rapid local temperature rise and avoid local overheating of the radiator.

[0042] Preferably, after step S6, the following steps are also included:

[0043] A1: Thermal response time constant τ = RC (R is thermal resistance, C is heat capacity;)

[0044] A2: Heat dissipation efficiency per unit power consumption η=ΔP / (W·t) (ΔP is the heat dissipation, W is the power consumption, and t is the time)

[0045] A3: Temperature uniformity index U = 1 - [max(Ti) - min(Ti)] / Tavg;

[0046] When any indicator deteriorates by more than 15%, the system self-check procedure is triggered.

[0047] By adopting the above technical solution, the performance of the radiator can be comprehensively evaluated by calculating the thermal response time constant, heat dissipation efficiency per unit power consumption, and temperature uniformity index. Once any indicator deteriorates by more than 15%, the system self-test program is triggered, which can promptly detect and handle possible system faults and ensure the normal and stable operation of the anti-clogging water circulation radiator.

[0048] Preferably, the adjustment of the operating parameters of the microchannel pump set includes:

[0049] a) Automatically switch flow regimes based on Reynolds number Re=ρvD / μ (ρ is density kg / m³, v is flow velocity m / s, D is equivalent diameter m, μ is dynamic viscosity Pa·s).

[0050] b) When Re>2300, start the eddy current generator and inject nanobubbles (50-100nm in diameter).

[0051] c) Adjust the pH value in real time to maintain the fluid zeta potential within ±30mV (adjustment accuracy ±0.5pH).

[0052] By adopting the above technical solutions, the flow state can be automatically switched according to the Reynolds number to adapt to different working conditions and improve heat dissipation efficiency; when the Reynolds number is greater than 2300, the eddy current generator is activated and nanobubbles are injected to enhance fluid disturbance and strengthen heat transfer; real-time adjustment of pH value to maintain the fluid zeta potential within a certain range can ensure fluid stability and avoid scaling and clogging of microchannels. In summary, it can better realize the effective adjustment of the working parameters of the microchannel pump group and ensure the stable and efficient operation of the anti-clogging water circulation radiator.

[0053] Preferably, fault tolerance mechanisms are also included:

[0054] a) Detect abnormal frequency components of the pump unit through vibration spectrum analysis (sampling frequency 10kHz);

[0055] b) When a fault > 0.2fN (fN is the rated frequency) is detected, the system automatically switches to the backup flow path;

[0056] c) Initiate the topology reconstruction algorithm to recalculate the optimized velocity distribution map (response time < 200ms).

[0057] By adopting the above technical solution, abnormal frequency components of the pump group can be detected in a timely manner using vibration spectrum analysis. When the abnormal frequency components of the pump group exceed a certain proportion, it can automatically switch to the backup flow channel to avoid the failure affecting the system operation. At the same time, the topology reconstruction algorithm is quickly started to recalculate and optimize the flow velocity distribution map to ensure that the radiator can continuously and efficiently dissipate heat.

[0058] In summary, this application includes at least one of the following beneficial technical effects:

[0059] 1. The filter structure inside the water drain prevents impurities, sediments and other substances from entering the water pump when the liquid is circulating internally, thus preventing blockage of the micro channels and ensuring normal circulation and heat dissipation efficiency of the coolant.

[0060] 2. A closed-loop control system is constructed by a temperature sensor array, an ambient temperature and humidity sensor, a microchannel pump group, an infrared thermal imager, and a vortex cooling device. Combined with control methods, this system can achieve precise regulation of radiator heat dissipation and improve heat dissipation efficiency.

[0061] 3. When the local temperature rise rate is detected to exceed the preset threshold, the phase change material heat storage module is activated to temporarily store heat, and the eddy current heat dissipation device is started simultaneously to implement directional forced heat dissipation to avoid damage to the equipment due to overheating. Attached Figure Description

[0062] Figure 1 is a schematic diagram of the anti-clogging water circulation radiator according to an embodiment of this application.

[0063] Figure 2 is a flowchart of the control method for the anti-clogging water circulation radiator according to an embodiment of this application.

[0064] Explanation of reference numerals in the attached diagram: 1. Radiator body; 11. Hot water inlet; 12. Cold water inlet; 13. Microchannel. Detailed Implementation

[0065] The technical solutions in the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. The described embodiments are only possible technical implementations of the present invention, but are not limited thereto. Other embodiments obtained by those skilled in the art in conjunction with the embodiments of the present invention without creative effort are also within the protection scope of the present invention.

[0066] This application mainly adopts a variety of components and control methods to prevent blockage and improve heat dissipation, which effectively prevents microchannel blockage and improves heat dissipation efficiency. The following is a further detailed description of this application.

[0067] This application provides an anti-clogging water circulation radiator, including a radiator body 1. The radiator body 1 has a built-in filter structure, a temperature sensor array, an ambient temperature and humidity sensor, a microchannel pump group, an infrared thermal imager, and a vortex cooling device. The radiator body 1 is provided with a hot water inlet 11 and a cold water inlet 12, and a water circulation channel is formed between the hot water inlet 11 and the cold water inlet 12, so that the coolant can circulate within the radiator body 1 to carry away heat. Correspondingly, a microchannel 13 is provided inside the radiator body 1 for drainage.

[0068] The filter structure is located inside the water drain and near the inlet of the microchannel pump unit. The straight-line distance from this location to the inlet is a preset value (determined based on the actual design). This arrangement ensures that the coolant is adequately filtered before entering the microchannel pump unit. The water drain is a component in the coolant circulation path, and the filter structure is used to filter the coolant before it enters the microchannel pump unit. The filter structure can be made of metal, such as stainless steel or aluminum, or it can be made of plastic. Its thickness is a preset value (determined based on the actual design), possessing good corrosion resistance and a certain strength, effectively blocking impurities, sediments, and other substances in the water. Simultaneously, it can filter microbubbles with a diameter greater than 0.1mm. If these microbubbles are not removed, they will burst near the worm gear blades during high-speed rotation, generating noise and causing abnormal sounds. Furthermore, after prolonged use, the impact force from the bursting bubbles can cause deformation or damage to the worm gear blades. The filter structure is connected to the water tank via a slot located inside the water tank near the inlet of the microchannel pump unit. The width and height of the slot are matched to the thickness and width of the filter structure, and its length is sufficient to accommodate the filter structure. This facilitates filter installation and ensures the durability of its filtration effect. The filter structure prevents impurities from entering the microchannel pump unit and clogging the microchannels, ensuring smooth coolant circulation and improving the stability and reliability of the heat dissipation system.

[0069] Correspondingly, the temperature sensor array consists of multiple thermocouple temperature sensors, evenly distributed on the microchannel surface of the heat sink body 1, with the spacing between adjacent sensors being a preset value (determined according to the actual design). Thermocouple temperature sensors can measure temperature quickly and accurately, and have a wide temperature measurement range. Multiple temperature sensors are fixed to the microchannel surface by welding, enabling real-time and comprehensive acquisition of multi-dimensional thermal field data on the heat sink surface, more accurately reflecting the temperature status of the heat sink.

[0070] Meanwhile, the ambient temperature and humidity sensor is a capacitive sensor, installed at the external air inlet of the radiator body 1. The air inlet can effectively reflect the temperature and humidity of the external environment. The capacitive temperature and humidity sensor offers high measurement accuracy and stability. It is bolted to a bracket on the outside of the air inlet and can acquire real-time temperature and humidity information from the external environment. This information is crucial for subsequently constructing a three-dimensional heat conduction feature matrix, as ambient temperature and humidity affect the radiator's heat dissipation performance. Taking these factors into account allows for more precise heat dissipation control.

[0071] Furthermore, the microchannel pump unit is connected to the cold water inlet 12 of the radiator body 1 via a flange. The outer diameter, inner diameter, thickness, and other dimensions of the flange are adapted to the connection points of the cold water inlet 12 and the microchannel pump unit. The connection method is bolted, and the bolt specifications and tightening torque are preset values ​​(determined according to the actual design). This connection method has the advantages of good sealing performance and a firm connection. The microchannel pump unit uses a centrifugal pump with a preset flow rate (determined according to the actual design) to meet the needs of coolant circulation. Its function is to provide power for coolant circulation, allowing coolant to flow from the cold water inlet 12 into the radiator body 1, absorb heat, and then flow out from the hot water inlet 11, thus achieving heat dissipation circulation.

[0072] Furthermore, the infrared thermal imager is a cooled infrared thermal imager, fixed to the top monitoring area of ​​the radiator body 1. This monitoring area is located at the center of the top of the radiator body 1, and its area and shape are adapted to the monitoring range of the infrared thermal imager. The cooled infrared thermal imager provides clear images and high sensitivity. It is bolted to a bracket in the top monitoring area and can monitor the top monitoring area of ​​the radiator body 1, acquiring real-time images of the temperature field distribution on the radiator surface, facilitating the timely detection of abnormalities such as localized overheating.

[0073] Correspondingly, the vortex cooling device is integrated downstream of the hot water inlet 11 of the radiator body 1. The two are connected by a pipe, the inner diameter, wall thickness, and other dimensions of which are preset values ​​(determined according to the actual design). The pipe connection is sealed with a sealing ring, the material, hardness, and size of which are compatible with the pipe connection to prevent coolant leakage. The vortex cooling device adopts a vortex-type air-cooled cooling system, which accelerates airflow by generating vortices to enhance the heat dissipation effect. It can further dissipate heat from the hot water after it flows out of the radiator body 1, reducing the hot water temperature and improving the efficiency of the entire cooling system.

[0074] Therefore, the temperature sensor array and ambient temperature and humidity sensor are connected to an external central controller via a CAN bus. The CAN bus model and specifications are preset (determined according to the actual design). CAN bus communication features high reliability and strong anti-interference capabilities, ensuring that the data collected by the sensors is accurately transmitted to the central controller. The microchannel pump group and the eddy current cooling device each receive PWM control signals, forming a closed-loop control system. The PWM control signals can precisely adjust the speed of the microchannel pump group and the power of the eddy current cooling device, dynamically adjusting according to the actual temperature and heat dissipation requirements.

[0075] On the other hand, the anti-clogging water circulation radiator uses the following control method to perform heat dissipation:

[0076] S1. A distributed temperature sensor array is used to collect multi-dimensional thermal field data of the radiator surface in real time, and a three-dimensional heat conduction feature matrix is ​​constructed by combining this data with ambient temperature and humidity sensors. The distributed temperature sensor array adopts a honeycomb topology layout, with the sensor spacing d satisfying d=(Q / (kΔT))^(1 / 3), where Q is the rated power of the radiator (unit: W), k is the thermal conductivity of the substrate material (unit: W / m·K), and ΔT is the system's allowable temperature rise threshold (unit: ℃). This layout can more comprehensively cover the radiator surface and collect more accurate temperature data. Combined with information obtained from the ambient temperature and humidity sensors, a three-dimensional heat conduction feature matrix containing multi-dimensional information such as temperature and humidity can be constructed, providing a detailed data foundation for subsequent thermal analysis.

[0077] S2. A dynamic thermal resistance model is established based on the unsteady-state heat conduction equation. The heat dissipation efficiency attenuation gradient is calculated according to the characteristic matrix, and the local heat sink accumulation trend is predicted. The dynamic thermal resistance model is solved discretly using the finite element method. The heat flux density correction coefficient α is obtained through training a convolutional neural network. The training objective function is L=||Tsim - Treal||²+λ|∇α|, where λ is the regularization coefficient (range 0.01 - 0.1), Tsim is the simulated temperature value, and Treal is the measured temperature value. Through this model and calculation method, the heat dissipation performance of the radiator can be understood more accurately, and the local heat sink accumulation trend can be predicted in advance so that timely measures can be taken for adjustment.

[0078] S3. Computational fluid dynamics simulation is used to generate an optimized velocity distribution map of the microchannel array on the radiator surface, and the preferred control region is determined based on the heat sink accumulation trend. The generation of the optimized velocity distribution map includes:

[0079] a) Establish an equivalent flow resistance network model for the microchannel array;

[0080] b) A genetic algorithm is used to solve the multi-objective optimization problem: min[Σ(Ti - Topt)² + βΣvi³]; where vi is the flow velocity of each microchannel (unit: m / s), Topt is the optimal temperature distribution (unit: °C), and β is the energy consumption weighting coefficient (value range: 0.05-0.2). This allows the optimal flow velocity distribution of the coolant in the microchannel array to be found, and combined with the heat sink accumulation trend, the areas that need to be focused on for control can be determined, heat dissipation resources can be rationally allocated, and heat dissipation efficiency can be improved.

[0081] S4. A multi-level frequency conversion control signal is generated based on a fuzzy neural network algorithm to dynamically adjust the operating parameters of the microchannel pump group in the priority control region. The fuzzy neural network contains a 5-layer structure:

[0082] The input layer contains heat sink accumulation rate, ambient temperature and humidity, and historical control records;

[0083] The fuzzification layer uses a Gaussian membership function to divide the fuzziness into three levels;

[0084] The rule base stores 27 control rules generated from expert experience;

[0085] The inference layer uses the Mamdani method for fuzzy inference;

[0086] The deblurring layer uses the centroid method to output the frequency conversion control signal. Through this complex and intelligent algorithm, the operating parameters of the microchannel pump unit can be dynamically adjusted according to the actual situation, achieving precise heat dissipation control.

[0087] S5. The surface temperature field changes of the heat sink are monitored in real time using an infrared thermal imager, and the boundary condition parameters of the dynamic thermal resistance model are corrected accordingly. Boundary condition correction includes:

[0088] a) Extract the isotherm curvature feature C = |∇²T| (unit: 1 / m²) from thermal imaging data;

[0089] b) When C>0.8Cmax (Cmax is the design maximum curvature), the contact thermal resistance parameters of the thermal resistance model are automatically adjusted;

[0090] c) A Kalman filter is introduced to calibrate the model parameters in real time (sampling period 50ms). Through continuous feedback correction, the dynamic thermal resistance model can be made more accurate, and the heat dissipation control can be made more scientific and reasonable.

[0091] S6. When the local temperature rise rate exceeds a preset threshold, the phase change material thermal storage module is activated for temporary heat storage, and the eddy current cooling device is simultaneously activated to implement directional forced cooling. The phase change material thermal storage module includes a graded activation mechanism:

[0092] Primary thermal storage: Paraffin-based phase change materials undergo phase change in the range of 45-50℃ (latent heat value 180-200kJ / kg);

[0093] Secondary thermal storage: Metal-organic framework materials adsorb and release heat in the 55-60℃ range (specific surface area 3000m² / g).

[0094] Three-stage thermal storage: Nanocapsule-encapsulated salt water solution undergoes phase separation above 65℃ (particle size distribution 50-200nm). This staged activation mechanism can store heat in a targeted manner according to different temperature conditions. Combined with directional forced heat dissipation from a vortex cooling device, it can effectively address local overheating issues.

[0095] In addition, the following steps are included after step S6:

[0096] A1: Thermal response time constant τ = RC (R is thermal resistance, C is heat capacity); this parameter reflects the response speed of the heat sink to temperature changes.

[0097] A2: Heat dissipation efficiency per unit power consumption η = ΔP / (W·t) (ΔP is the heat dissipation, W is the power consumption, and t is the time); by calculating this efficiency, the energy-saving performance of the heat sink can be evaluated.

[0098] A3: Temperature uniformity index U = 1 - [max(Ti) - min(Ti)] / Tavg; When any index deteriorates by more than 15%, the system self-test program is triggered. The system self-test program can check the working status of each component, promptly identify potential problems, and repair them.

[0099] The operating parameters of the microchannel pump set are adjusted as follows:

[0100] a) Automatically switch the flow state based on the Reynolds number Re = ρvD / μ (ρ is density kg / m³, v is flow velocity m / s, D is equivalent diameter m, μ is dynamic viscosity Pa·s); by judging the Reynolds number, the flow state of the coolant can be determined so as to take appropriate adjustment strategies.

[0101] b) When Re>2300, start the eddy current generator and inject nanobubbles (50-100nm in diameter); the eddy current generator and nanobubbles can enhance the mixing and heat transfer of the coolant.

[0102] c) Adjust the pH value in real time to maintain the fluid zeta potential within ±30mV (adjustment accuracy ±0.5pH); maintaining a suitable pH value and zeta potential can prevent impurities in the coolant from agglomerating and reduce the risk of blockage.

[0103] This anti-clogging water circulation radiator also includes a fault-tolerant mechanism:

[0104] a) Abnormal frequency components of the pump unit can be detected by vibration spectrum analysis (sampling frequency 10kHz); abnormal vibration of the pump unit can be detected in a timely manner.

[0105] b) When a fault > 0.2fN (fN is the rated frequency) is detected, the system automatically switches to the backup flow path; the backup flow path can ensure that the coolant can still circulate normally when the main flow path has a problem.

[0106] c) Activate the topology reconfiguration algorithm to recalculate and optimize the flow velocity distribution map (response time <200ms); the topology reconfiguration algorithm can quickly adjust the structure and parameters of the heat dissipation system to adapt to new operating conditions.

[0107] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A clog-resistant water circulation radiator, characterized in that, The system includes a radiator body (1), which incorporates a filter structure, a temperature sensor array, an ambient temperature and humidity sensor, a microchannel pump group, an infrared thermal imager, and a vortex heat dissipation device. The radiator body (1) is equipped with a hot water inlet (11) and a cold water inlet (12). The filter structure is located inside the water drain and adjacent to the inlet of the microchannel pump group. The filter structure is made of stainless steel, aluminum, or plastic. The filter structure filters out other metal scraps and air bubbles circulating in the liquid within the water drain chamber, reducing the accumulation of copper and causing electrochemical oxidation corrosion, thus maintaining stable flow and heat exchange. The temperature sensor array is evenly distributed within the radiator body. (1) The microchannel surface, the ambient temperature and humidity sensor is installed at the air inlet outside the radiator body (1), the microchannel pump group is connected to the cold water inlet (12) of the radiator body (1) through the flange, the infrared thermal imager is fixed at the top monitoring area of ​​the radiator body (1), and the vortex heat dissipation device is integrated downstream of the hot water inlet (11) of the radiator body (1); the temperature sensor array and the ambient temperature and humidity sensor are connected to the external central controller through the CAN bus, the microchannel pump group and the vortex heat dissipation device respectively receive PWM control signals to form a closed-loop control system; the anti-clogging water circulation radiator uses the following control method to perform heat dissipation: S1. Real-time collection of multi-dimensional thermal field data on the radiator surface through the distributed temperature sensor array, combined with the ambient temperature and humidity sensor to construct a three-dimensional heat conduction feature matrix; S2. Establish a dynamic thermal resistance model based on the unsteady heat conduction equation, calculate the heat dissipation efficiency attenuation gradient according to the characteristic matrix, and predict the local heat sink accumulation trend; S3. Use computational fluid dynamics simulation to generate an optimized flow velocity distribution map of the microchannel array on the radiator surface, and determine the priority control area in combination with the heat sink accumulation trend; S4. Generate multi-level frequency conversion control signals based on fuzzy neural network algorithm to dynamically adjust the operating parameters of the microchannel pump group in the priority control area; S5. Monitor the temperature field change of the radiator surface in real time using an infrared thermal imager and correct the boundary condition parameters of the dynamic thermal resistance model; S6. When the local temperature rise rate is detected to exceed the preset threshold, activate the phase change material heat storage module to temporarily store heat and simultaneously start the eddy current heat dissipation device to implement directional forced heat dissipation.

2. The anti-clogging water circulation radiator according to claim 1, characterized in that, In step S1, the distributed temperature sensor array adopts a honeycomb topology layout, and the sensor spacing d satisfies d=(Q / (kΔT))^(1 / 3), where Q is the rated power of the heat sink (unit: W), k is the thermal conductivity of the substrate material (unit: W / m·K), and ΔT is the system's allowable temperature rise threshold (unit: ℃).

3. The anti-clogging water circulation radiator according to claim 2, characterized in that, In step S2, the dynamic thermal resistance model is solved by discretization using the finite element method. The heat flux density correction coefficient α is obtained by training a convolutional neural network. The training objective function is L=||Tsim-Treal||²+λ|∇α|, where λ is the regularization coefficient (ranging from 0.01 to 0.1), Tsim is the simulated temperature value, and Treal is the measured temperature value.

4. The anti-clogging water circulation radiator according to claim 2, characterized in that, The generation of the optimized flow velocity distribution map in step S3 includes: a) establishing an equivalent flow resistance network model of the microchannel array; b) using a genetic algorithm to solve the multi-objective optimization problem: min[Σ(Ti-Topt)² + βΣvi³]; where vi is the flow velocity of each microchannel (unit: m / s), Topt is the optimal temperature distribution (unit: ℃), and β is the energy consumption weight coefficient (value range: 0.05-0.2).

5. The anti-clogging water circulation radiator according to claim 1, characterized in that, In step S4, the fuzzy neural network comprises a 5-layer structure: Input layer: heat sink accumulation rate, ambient temperature and humidity, and historical control records; Fuzzification layer: using Gaussian membership functions to divide the fuzziness into 3 levels; Rule base: storing 27 control rules generated by expert experience; Inference layer: using the Mamdani method for fuzzy inference; Defuzzification layer: using the centroid method to output the frequency conversion control signal.

6. The anti-clogging water circulation radiator according to claim 5, characterized in that, The boundary condition correction in step S5 includes: a) extracting the isotherm curvature feature C=|∇²T| (unit: 1 / m²) from thermal imaging data; b) automatically adjusting the contact thermal resistance parameters of the thermal resistance model when C>0.8Cmax (Cmax is the design maximum curvature); c) introducing a Kalman filter to calibrate the model parameters in real time (sampling period 50ms).

7. The anti-clogging water circulation radiator according to claim 1, characterized in that, In step S6, the phase change material thermal storage module includes a graded activation mechanism: Level 1 thermal storage: paraffin-based phase change material undergoes phase change in the 45-50℃ range (latent heat value 180-200kJ / kg); Level 2 thermal storage: metal-organic framework material adsorbs and releases heat in the 55-60℃ range (specific surface area 3000m² / g); Level 3 thermal storage: nanoencapsulated salt water solution undergoes phase separation above 65℃ (particle size distribution 50-200nm).

8. The anti-clogging water circulation radiator according to claim 7, characterized in that, After step S6, Includes the following steps: A1: Thermal response time constant τ=RC (R is thermal resistance, C is heat capacity); A2: Heat dissipation efficiency per unit power consumption η=ΔP / (W·t) (ΔP is heat dissipation, W is power consumption, t is time); A3: Temperature uniformity index U=1-[max(Ti)-min(Ti)] / Tavg; When any index deteriorates by more than 15%, the system self-test program is triggered.

9. A clog-resistant water circulation radiator according to claim 7, characterized in that, The microchannel pump set operating parameters are adjusted as follows: a) Automatically switch the flow state according to the Reynolds number Re=ρvD / μ (ρ is density kg / m³, v is flow velocity m / s, D is equivalent diameter m, μ is dynamic viscosity Pa·s); b) When Re>2300, start the eddy current generator and inject nanobubbles (50-100nm in diameter); c) Adjust the pH value in real time to maintain the fluid zeta potential within ±30mV (adjustment accuracy ±0.5pH).

10. A clog-resistant water circulation radiator according to claim 7, characterized in that, It also includes fault tolerance mechanisms: a) Detecting abnormal frequency components of the pump group through vibration spectrum analysis (sampling frequency 10kHz); b) When fault > 0.2fN (fN is the rated frequency), automatically switching to the backup flow channel; c) Starting the topology reconstruction algorithm to recalculate and optimize the flow velocity distribution map (response time < 200ms).

Images