Chip hotspot heat dissipation system based on micro robot

Abstract

The application provides a chip hotspot heat dissipation system based on a micro robot, which comprises a magnetic field control device for generating a uniform magnetic field; a chip is placed in the uniform magnetic field; a heat dissipation channel is arranged on the surface of the chip, and a cooling medium for conducting convective heat exchange on a hotspot is circulated in the heat dissipation channel; a robot is made of a magnetic material and is placed in the cooling medium; and a processor is used for controlling the magnetic field control device to generate a magnetic field at a first target rotating frequency, so as to drive the robot to run in the cooling medium to a region where the hotspot of the chip is located; and the processor is also used for controlling the magnetic field control device to generate a magnetic field at a second target rotating frequency, so as to drive the robot to rotate in place in the region where the hotspot of the chip is located, so as to dissipate heat of the hotspot of the chip. The system provided by the application effectively solves the problem that the prior art is difficult to accurately match the dynamic migration characteristics of the hotspot; and the heat dissipation efficiency of the hotspot of the chip is improved.

Description

Technical Field

[0001] This invention relates to the field of thermal management technology for microelectronic devices, and more specifically to a chip hotspot heat dissipation system based on microrobots. Background Technology

[0002] As integrated circuit technology continues to advance towards the nanometer and even angstrom scales, transistor density is increasing exponentially, driving leaps in chip performance in terms of computing speed and integration. However, the high integration of transistors has also led to a sharp increase in chip power consumption density, resulting in an explosive growth in heat generated during chip operation. The silicon substrate material used inside the chip has relatively limited thermal conductivity, making it difficult to quickly and effectively conduct and dissipate large amounts of heat, causing excessive heat accumulation in localized areas of the chip, forming high-temperature hotspots. The presence of high-temperature hotspots seriously threatens the stable operation and performance of the chip. On the one hand, excessively high temperatures accelerate the aging and failure of electronic components inside the chip, significantly reducing the chip's lifespan; on the other hand, high temperatures can cause thermal degradation in chip performance, such as decreased computing speed and increased signal transmission delay, and in severe cases, even chip failure, affecting the reliability and stability of the entire electronic system. It is worth noting that the distribution of chip hotspots exhibits dynamic migration characteristics. On the one hand, due to the differences in circuit structure and operating frequency in different areas of the chip, the heat generation and accumulation in different areas are different; on the other hand, as the chip's operating conditions frequently switch between data processing, graphics rendering, and low-power standby modes, the location and intensity of hot spots will also change dynamically.

[0003] Existing chip cooling technologies mainly include overall passive cooling (such as heat sinks and thermal grease), overall active cooling (such as fans and liquid cooling), and hot spot enhanced cooling (such as local water cooling and thermoelectric cooling).

[0004] Therefore, there is an urgent need for a new type of heat dissipation device that can adapt to the dynamic migration characteristics of chip hotspots in order to effectively solve the problem of dynamic migration of chip hotspots. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a chip hotspot heat dissipation system based on a microrobot.

[0006] A chip hotspot heat dissipation system based on microrobots, comprising:

[0007] A magnetic field control device is used to generate a uniform magnetic field; the chip is placed in the uniform magnetic field.

[0008] A heat dissipation channel is provided on the surface of the chip, and a cooling medium for convective heat transfer to hot spots flows in the heat dissipation channel.

[0009] The robot, made of magnetic material, is placed in the cooling medium;

[0010] The processor is used to control the magnetic field control device to generate a magnetic field at a first target rotation frequency to drive the robot to run in the cooling medium to the hot spot area of ​​the chip;

[0011] It is also used to control the magnetic field control device to generate a magnetic field at a second target rotation frequency, so as to drive the robot to rotate in place in the area where the chip hot spot is located, so as to dissipate heat from the chip hot spot.

[0012] Furthermore, the chip hotspot heat dissipation system based on microrobots as described above also includes: an infrared thermal imaging system and an image acquisition device electrically connected to the processor;

[0013] The infrared thermal imaging system is used to scan the surface of the chip to obtain an infrared thermal imaging image, and transmit the infrared thermal imaging image to the processor;

[0014] The image acquisition device is used to acquire an image of the chip surface and transmit the image of the chip surface to the processor, so that the processor can locate the coordinates of the chip hotspot based on the infrared thermal imaging image and the chip surface image.

[0015] Furthermore, in the microrobot-based chip hotspot heat dissipation system described above, the processor is also used for:

[0016] Determine the driving force and tox drag on the robot in the magnetic field;

[0017] The robot's first self-rotation frequency is determined based on the driving force and the tox drag.

[0018] Based on the robot's first self-rotation frequency, a first target rotation frequency of the magnetic field is determined, and the robot is driven to run in the cooling medium to the hot spot area of ​​the chip at the first target rotation frequency.

[0019] Furthermore, in the microrobot-based chip hotspot heat dissipation system described above, the driving force experienced by the robot in the magnetic field is:

[0020]

[0021] The Stokes drag experienced by the robot in the magnetic field is:

[0022]

[0023] in, The driving force experienced by the robot in the magnetic field. This is the structural shape factor of the robot. This is the first self-rotation frequency of the microrobot. Indicates the dynamic viscosity coefficient of the cooling medium. Let be the robot's equivalent radius; Let v be the toxenary drag force experienced by the robot in the magnetic field, and v be the robot's speed.

[0024] Furthermore, in the microrobot-based chip hotspot heat dissipation system described above, the infrared thermal imaging system is also used to measure the first temperature at the hotspot when the robot has not reached the chip hotspot, and the second temperature at the hotspot after the robot has rotated stably in the area where the chip hotspot is located.

[0025] The processor is also configured to: acquire all chip surface images acquired by the image acquisition device at a preset frequency to determine whether the robot has reached a stable rotation state in the area where the chip hotspot is located; and after reaching a stable state, acquire the second temperature using the infrared thermal imaging system.

[0026] Based on the first temperature and the second temperature, determine the initial heat flux change of the robot at the hot spot;

[0027] Based on the initial heat flux change, the current second spin frequency of the magnetic field control device is continuously adjusted, and the robot is driven to rotate continuously in the area where the chip hot spot is located by the uniform magnetic field corresponding to the second spin frequency, until the current heat flux change meets the expected heat dissipation target.

[0028] Furthermore, in the microrobot-based chip hotspot heat dissipation system described above, determining the initial heat flux change of the robot at the hotspot based on the first temperature and the second temperature includes:

[0029] The change in heat flux of the robot at the hot spot is determined using the following formula:

[0030]

[0031]

[0032] in, The first temperature; The initial temperature of the cooling medium. For the size of the robot, The thermal conductivity of the cooling medium The convective heat transfer coefficient at the hotspot is given before the robot moves to it; Gr is the Grashof number; Pr is the Prandtl number. The dynamic viscosity coefficient, It is thermal conductivity; The second temperature; This is the second spin frequency of the microrobot. is the kinematic viscosity coefficient of the cooling medium.

[0033] Furthermore, the chip hotspot heat dissipation system based on microrobots as described above also includes an inlet temperature sensor, an outlet temperature sensor, and a flow sensor connected to the processor;

[0034] The inlet temperature sensor is located at the inlet of the heat dissipation channel, and the outlet temperature sensor is located at the outlet of the heat dissipation channel; the flow sensor is used to measure the flow rate of the cooling medium in the heat dissipation channel.

[0035] The processor is also used for:

[0036] The initial heat exchange efficiency of the robot at the hot spot is determined based on the temperature data measured by the inlet temperature sensor and the outlet temperature sensor, respectively, and the data measured by the flow sensor.

[0037] Based on the initial heat exchange efficiency, continuously adjust the... The third target rotational frequency is determined until the current heat exchange efficiency meets the expected heat exchange efficiency.

[0038] Furthermore, in the microrobot-based chip hotspot heat dissipation system described above, determining the initial heat exchange efficiency of the robot at the hotspot based on the temperature data measured by the inlet temperature sensor and the outlet temperature sensor, respectively, and the data measured by the flow sensor, includes:

[0039] The initial heat exchange efficiency of the robot at the hot spot is determined using the following formula:

[0040]

[0041]

[0042]

[0043]

[0044] in, The actual heat exchange between the cooling medium and the chip This represents the maximum heat transfer between the cooling medium and the chip. The area through which the cooling medium flows on the chip. The average temperature difference between the chip surface and the cooling medium. This represents the average temperature at the hot spot of the chip. The temperature before the cooling medium enters the chip. The temperature after the cooling medium flows out of the chip. For the cooling medium flow rate, The specific heat capacity of the cooling medium is the flow rate. This represents the highest temperature at the hot spot of the chip.

[0045] Furthermore, in the microrobot-based chip hotspot heat dissipation system described above, the heat dissipation channel is a straight channel, a serpentine channel, a tree-like channel, or a grid-like channel.

[0046] The system provided by this invention, on the one hand, generates a uniform magnetic field corresponding to a first target rotation frequency by controlling a magnetic field control device through a processor, enabling the robot to accurately reach the hot spot area, effectively solving the problem that existing technologies are unable to accurately match the dynamic migration characteristics of hot spots; on the other hand, it generates a uniform magnetic field corresponding to a second target rotation frequency by controlling a magnetic field control device through a processor, enabling the robot to rotate in place in the area where the chip hot spot is located, thereby agitating the cooling medium, increasing the flow rate of the cooling medium, thereby accelerating the transfer of heat from the chip hot spot area to the cooling medium, and ultimately improving the heat dissipation efficiency of the chip hot spot. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the chip hotspot system structure based on microrobots provided by the present invention;

[0048] Figure 2 This is a block diagram of a chip hotspot system structure based on microrobots provided by the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0050] In today's era of rapid technological advancement, integrated circuit technology has made significant progress, resulting in a substantial increase in transistor density. While this progress has greatly improved chip performance and functionality, it has also brought about a series of serious challenges. Among these, the heat generation and dissipation issues of chips are particularly prominent and have become a key factor restricting further improvements in chip performance.

[0051] With the rapid increase in transistor density, chip power consumption density continues to rise, exacerbating the contradiction between heat generation and dissipation. Due to the limited thermal conductivity of silicon substrates, when a chip generates a large amount of heat, the heat is difficult to dissipate quickly within the chip, resulting in high-temperature hotspots in localized areas. These chip hotspots are not simply due to overheating, but involve complex electro-thermal-mechanical coupling. This complex coupling leads to a series of serious consequences. It accelerates electromigration within the chip, causing atomic migration in metal wires, which may lead to circuit breaks or short circuits; it also generates thermal stress in the chip, which, when exceeding the material's tolerance limit, can cause thermal stress cracking, damaging the chip's structural integrity; simultaneously, localized high temperatures can degrade transistor performance, affecting the chip's normal operating state; furthermore, to protect itself, the chip triggers a frequency reduction mechanism, which undoubtedly directly reduces the chip's computing power, severely impacting its performance in practical applications. From the perspective of the spatial distribution of chip power consumption, it exhibits a high degree of complexity. Chip hotspots are typically distributed discretely on the order of hundreds of micrometers, and their positions and intensities are not fixed but dynamically change with variations in chip operating conditions, exhibiting significant spatial non-uniformity and temporal dynamic variation. In other words, during chip operation, the positions of hotspots are not static but move between different areas over time—a phenomenon known as dynamic migration of hotspot positions. This places extremely high demands on chip heat dissipation technology, requiring it to respond to these changes in real-time and with precision.

[0052] Currently, chip cooling technologies mainly include overall passive cooling, overall active cooling, and hotspot-enhanced cooling. Overall passive and overall active cooling technologies are based on the idea of ​​dissipating heat from the entire chip, making it difficult to precisely match the dynamic migration characteristics of hotspots, resulting in low heat dissipation efficiency and energy waste. While hotspot-enhanced cooling technology can dissipate heat from localized hotspots, its cooling position and intensity cannot be adjusted in real time for dynamically changing hotspots, making it difficult to achieve efficient and continuous heat dissipation. To solve this problem, this application provides a chip hotspot heat dissipation system, such as... Figure 1 , Figure 2 As shown, the system includes: a magnetic field control device capable of generating a uniform magnetic field, with chip 2 placed in the uniform magnetic field; a heat dissipation channel 1 is provided on the surface of chip 2, and a cooling medium for convective heat transfer to hot spots flows within the heat dissipation channel 1; it also includes a processor connected to the magnetic field control device, which is used to control the magnetic field control device to generate a magnetic field at a first target rotation frequency to drive the robot to run in the cooling medium to the area where the hot spot of the chip is located; and is also used to control the magnetic field control device to generate a magnetic field at a second target rotation frequency to drive the robot to rotate in place in the area where the hot spot of the chip is located to dissipate heat from the hot spot of the chip.

[0053] In this embodiment of the invention, the microrobot is a magnetic particle with chiral characteristics, typically in the micrometer or nanometer range, preferably with a diameter of 20 μm to 100 μm. The asymmetric structure of the chiral microrobot is key to its chiral characteristics and specific functions; common asymmetric structures include helical shapes and tadpole-shaped tails. Taking the helical structure as an example: the helical structure is a typical asymmetric structure, similar to the double helix of DNA or the thread of a screw. This structure has a continuous, helical extension characteristic, and its pitch, helical radius, and other parameters affect the performance of the microrobot. When the helical chiral microrobot is placed in a cooling medium and driven by an external magnetic field, it will rotate like a propeller. Due to the chirality of the helical structure, the rotational motion is converted into linear motion. For example, driven by a uniform magnetic field, the magnetic helical microrobot can move linearly to a local hot spot on the chip.

[0054] Before dissipating heat from the chip hotspots, preparatory work is first performed: the chip is placed within the area covered by a magnetic field control device capable of generating a uniform magnetic field, while ensuring that the heat dissipation channels (microchannels) on the chip surface are unobstructed and filled with a cooling medium for convective heat transfer to the hotspots. A robot made of magnetic material is placed in the cooling medium. Because the magnetic field exerts a magnetic force on the robot, it can move along the direction of the magnetic field. This invention dissipates heat from the chip hotspots in two stages: In the first stage, the processor controls the magnetic field control device to generate a uniform magnetic field at a first target rotation frequency. Under the influence of this magnetic field at the first target rotation frequency, the magnetic robot flows in the cooling medium and moves to the area where the chip hotspots are located. In the second stage, the processor controls the magnetic field control device to generate a uniform magnetic field at the second target rotation frequency. Under the action of the magnetic field at the second target rotation frequency, the magnetic robot rotates in place in the area where the chip hot spot is located. When the robot rotates, the cooling medium around it is agitated and its flow state changes, thereby enhancing the convective heat transfer between the cooling medium and the chip hot spot area. This allows the heat of the chip hot spot to be quickly carried away by the cooling medium, thus ultimately achieving heat dissipation of the chip hot spot.

[0055] The magnetic field control device provided in this application includes a uniform magnetic field composed of three sets of orthogonal Helmholtz coils. This magnetic field's strength and direction are adjusted by controlling the magnitude and direction of the current in the coils to generate a spatially uniform rotating magnetic field. The robot's speed and direction are then adjusted based on the strength and direction of the rotating magnetic field. The rotating magnetic field... for:

[0056]

[0057] in, The magnitudes of the magnetic field in each direction are given. The rotational angular frequency (JFF) of a rotating magnetic field describes how fast the field rotates. It determines the rate at which the magnetic field direction changes with time, influencing the motion of a microrobot within the magnetic field, such as its rotational frequency. When the JFF changes, the magnetic torque acting on the microrobot changes, thus altering its rotation and speed. Therefore, the JFF is essentially the magnetic field generated by the Helmholtz coil, characterizing a key property of the magnetic field's rotational motion. The JFF can be converted into the rotational frequency of the magnetic field. . This initial phase represents the state of the rotating magnetic field at the start. When a Helmholtz coil generates a rotating magnetic field, this initial state influences the subsequent changes in the magnetic field, as well as the forces and motion tendencies of the microrobot within the magnetic field at the initial moment. For example, different initial phases may cause the microrobot to begin rotating motion from different directions.

[0058] here, The calculation formula is as follows:

[0059]

[0060] in, Let be the radius of the Helmholtz coil. Let be the distance from a point on the axis to the axis center. The number of turns of the coil. The current intensity passing through the coil, It is the permeability in vacuum, which is .

[0061] In this embodiment of the invention, the heat dissipation channel is specifically a microchannel disposed on the chip surface. A microchannel is a channel structure with a tiny size (typically at the micrometer level). In the field of chip heat dissipation, it provides a specific flow path for the cooling medium. The cooling medium flows within the microchannel and carries away the heat generated by the chip through convective heat exchange with the chip surface, thereby achieving the purpose of heat dissipation.

[0062] The heat dissipation channels can be straight channels, serpentine channels, tree-like channels, or grid-like channels.

[0063] Linear channels are distributed in a straight line on the chip surface (e.g.) Figure 1 (As shown). In some cases where heat dissipation requirements are not particularly complex and the chip structure is relatively regular, linear microchannels can meet basic heat dissipation needs. They allow the cooling medium to flow quickly across the chip surface, carrying away heat.

[0064] The serpentine microchannel, with its meandering shape, increases the flow path of the cooling medium across the chip surface. For chips with high power consumption and concentrated hotspots, the serpentine microchannel can better cover the hotspot areas, enhancing heat dissipation.

[0065] Tree-like channels resemble the branching structure of a tree, with a main channel and multiple branch channels. They can distribute the cooling medium more evenly across different areas of the chip surface, ensuring effective heat dissipation in each area. Especially when the chip area is large and different areas have different heat dissipation requirements, tree-like microchannels can achieve more precise heat dissipation control.

[0066] Mesh-like channels consist of multiple intersecting channels forming a grid-like structure. They offer excellent heat dissipation uniformity, simultaneously cooling multiple areas of the chip surface and preventing localized overheating. For chips with extremely high requirements for heat dissipation uniformity, such as high-performance computing chips, mesh-like microchannels are an ideal choice.

[0067] Heat dissipation channels can be made of metal or silicone. Metal materials include copper and aluminum. Because metals like copper and aluminum have excellent thermal conductivity, they can quickly conduct heat generated by the chip away. Therefore, using metal materials to create microchannels can improve the heat dissipation efficiency of chip hotspots. Heat dissipation channels made of silicone have a similar coefficient of thermal expansion to the chip, resulting in better compatibility and reducing the risk of damage caused by thermal stress.

[0068] The system provided by this invention, on the one hand, generates a uniform magnetic field corresponding to a first target rotation frequency by controlling a magnetic field control device through a processor, enabling the robot to accurately reach the hot spot area, effectively solving the problem that existing technologies are unable to accurately match the dynamic migration characteristics of hot spots; on the other hand, it generates a uniform magnetic field corresponding to a second target rotation frequency by controlling a magnetic field control device through a processor, enabling the robot to rotate in place in the area where the chip hot spot is located, thereby agitating the cooling medium, increasing the flow rate of the cooling medium, thereby accelerating the transfer of heat from the chip hot spot area to the cooling medium, and ultimately improving the heat dissipation efficiency of the chip hot spot.

[0069] Furthermore, such as Figure 2 As shown, the system provided by the present invention further includes an infrared thermal imaging system and an image acquisition device electrically connected to the processor; the infrared thermal imaging system is used to scan the surface of the chip to obtain an infrared thermal imaging image and transmit the infrared thermal imaging image to the processor; the image acquisition device is used to acquire an image of the chip surface and transmit the chip surface image to the processor, so that the processor can determine the coordinates of the hot spot positions on the chip based on the infrared thermal imaging image and the chip surface image.

[0070] Specifically, this application employs a high-precision infrared thermal imaging system to perform full-field temperature measurement on the chip surface to obtain the surface temperature distribution of the chip. This system can capture infrared radiation energy at different locations on the chip surface, converting it into temperature data to form a temperature distribution image of the chip surface. Furthermore, the system is equipped with a macro optical lens, enabling... The system achieves high spatial resolution (level 1) and temperature resolution (0.1K) to ensure accurate capture of hotspot distributions at the hundred-micron level. By acquiring the surface temperature distribution of the chip, it can precisely capture temperature changes at different locations on the chip surface, providing fundamental data for subsequent hotspot localization and thus improving the accuracy of hotspot location.

[0071] Furthermore, this application calculates the heat flux density distribution on the chip surface based on the power consumption distribution and heat conduction theory of the chip, using Fourier's law. Heat flux density reflects the intensity of heat transfer at various points on the chip surface and is an important basis for locating hot spots. This heat flux density distribution is calculated using the following formula:

[0072]

[0073] Where k is the thermal conductivity of the chip material.

[0074] This application sets a hotspot temperature threshold. and heat flux density threshold The surface temperature distribution of the chip is respectively With preset temperature threshold Comparison and heat flux density distribution With heat flux density threshold Compare; in and In this case, determine the surface temperature distribution The area in question is a hotspot; based on this hotspot area, the location coordinates of the hotspot are obtained using a coordinate positioning algorithm. .

[0075] In this invention, the location coordinates of the hotspot can be obtained using a coordinate positioning algorithm. The specific method is as follows:

[0076] First, the processor establishes a correspondence between the chip surface image captured by the image acquisition device and the infrared thermal imaging image obtained by scanning the infrared thermal imaging image:

[0077] The processor calibrates the infrared thermal imaging system, determining the pixel coordinates of the infrared thermal image and the actual size of the chip based on the chip surface image. Then, it establishes a proportional relationship between the hotspot pixel coordinates and the actual chip size. Assuming the number of pixels in the x-direction is Nx, and the corresponding actual chip length is Lx, then the actual length represented by each pixel in the x-direction is Δx = Lx / Nx. Similarly, in the y-direction, if the number of pixels in the image is Ny, and the corresponding actual chip width is Ly, then the actual length represented by each pixel in the y-direction is Δy = Ly / Ny. For the z-coordinate, since infrared thermal imaging typically only measures the temperature distribution on the object's surface, the z-coordinate is generally assumed to be the plane containing the chip surface, i.e., z = 0.

[0078] Determine the hot spot pixel coordinates: The processor finds the point with the highest temperature in the infrared thermal imaging image, i.e., the hot spot, and records its pixel coordinates (i,j) in the image.

[0079] Calculate the actual coordinates of the hotspot: Based on the previously established correspondence, convert the pixel coordinates into actual coordinates. The actual coordinates of the hotspot in the x-direction are x = i * Δx, and the actual coordinates in the y-direction are y = j * Δy, z = 0. Thus, the chip hotspot location coordinates are calculated. .

[0080] This application utilizes an image acquisition device to monitor the microrobot's position information in real time and compares it with the coordinates of a specified location above the hotspot to determine its position. Assuming the specified location coordinates above the hotspot are (x, y, z), the current real-time position coordinates of the microrobot obtained through monitoring are: Calculate the distance between the two. Set an allowable error range Δd. When d ≤ Δd, the microrobot is considered to have moved to the hot spot position.

[0081] Furthermore, in the system provided by the present invention, the processor is also configured to: determine the driving force and torpedo drag experienced by the robot in the magnetic field; determine a first self-spinning frequency of the robot based on the driving force and torpedo drag; determine a first target rotation frequency of the magnetic field based on the first self-spinning frequency of the robot; and drive the robot to run in the cooling medium to the hot spot area of ​​the chip at the first target rotation frequency.

[0082] Specifically, in this embodiment, the microrobot is a magnetic particle with a tail. When the microrobot contains magnetic material inside or on its surface, it generates driving force and resistance under the influence of a rotating magnetic field.

[0083] Here, the driving force experienced by the robot in the magnetic field. for:

[0084]

[0085] in, This is the structural shape factor of the robot. Indicates the dynamic viscosity coefficient of the cooling medium. The self-rotation frequency of the microrobot Let be the equivalent radius of the microrobot.

[0086] Stokes drag on the robot in a magnetic field for:

[0087]

[0088] Where v is the robot's running speed and L is the robot's equivalent radius.

[0089] When a chiral microrobot's asymmetric structure (such as a spiral or tadpole-shaped tail) rotates under a uniform rotating magnetic field, the interaction between the structural asymmetry and the fluid generates a propulsive force in a specific direction. In this application, since the rotational Reynolds number is much less than 1, the applicability conditions of Stokes' drag theory are met. The propulsive force generated in a specific direction by the interaction between the structural asymmetry and the fluid when the chiral microrobot rotates under a uniform rotating magnetic field is essentially the reaction force of the fluid on the rotating structure. In this case, Stokes' drag theory can be used to approximate the calculation of this propulsive force.

[0090] when At that moment, the robot accelerates forward.

[0091] when When the robot moves forward at a constant speed v, it will move forward at a constant speed v.

[0092] when When the robot rotates in place, it will remain stationary.

[0093] In this case, in order for the robot to move to the hot spot under the influence of the magnetic field, it is necessary to ensure that: .

[0094] when At that time, there were:

[0095]

[0096] The critical self-rotation frequency of the robot can then be determined according to the following formula. for:

[0097]

[0098] Then: Assume the robot's running speed is Then the first self-rotation frequency of the robot is Assume the robot's running speed is Then the first self-rotation frequency of the robot is To ensure Then it is necessary to ensure the robot's first self-rotation frequency. .

[0099] Furthermore, when the magnetic robot is placed in a magnetic field, its rotational frequency will eventually approach the rotational frequency of the magnetic field. In this case, it is assumed that the rotational frequency of the magnetic robot... Equal to the rotation frequency of the magnetic field ,Right now Therefore, only the rotation frequency of the magnetic field... The robot will only operate when... At this time, the robot will rotate in place. Therefore, this application determines the robot's rotation frequency by setting a speed, and then determines the corresponding rotation frequency of the magnetic field based on the robot's rotation frequency. Thus, this application only needs to ensure the rotation frequency of the magnetic field, indirectly determined by the robot's running speed. That's it. Once the rotation frequency of the magnetic field is determined, it can be substituted into the above formula to determine the driving force and resistance experienced by the robot in the magnetic field, and then the robot's running speed can be determined based on the driving force and resistance experienced by the robot in the magnetic field.

[0100] This application uses the following formula to control the robot to move to the area where the chip hotspot is located:

[0101]

[0102] in, The target position for the microrobot. This is the current position of the microrobot. Let be the initial velocity of the microrobot. For the movement time of the microrobot, Let be the acceleration of the microrobot. When the robot moves forward at a constant speed, the acceleration is 0.

[0103] Here, the acceleration of the microrobot in the magnetic field It can be obtained through the following methods:

[0104]

[0105]

[0106] in, For the density of microrobots, For the volume of the microrobot, Let F be the acceleration of the microrobot in the magnetic field; and let F be the net force acting on the robot in the magnetic field. The driving force experienced by the robot in the magnetic field; This represents the resistance experienced by the robot in the rotating magnetic field.

[0107] Furthermore, the system provided in this application includes an infrared thermal imaging system that measures a first temperature at the hot spot of the chip before the robot reaches it, and a second temperature at the hot spot after the robot has rotated stably in the area where the chip hot spot is located. The processor is also used to: acquire all chip surface images acquired by the image acquisition device at a preset frequency to determine whether the robot has reached a stable rotation state in the area where the chip hot spot is located; after reaching a stable state, acquire the second temperature using the infrared thermal imaging system; and determine the initial heat flux change of the robot at the hot spot based on the first temperature and the second temperature; finally, continuously adjust the current second self-rotation frequency of the magnetic field control device based on the initial heat flux change, and drive the robot to rotate continuously in the area where the chip hot spot is located with the uniform magnetic field corresponding to the second self-rotation frequency until the current heat flux change meets the expected heat dissipation target.

[0108] Specifically, the change in heat flux is the difference in heat transferred per unit area and per unit time at the chip hotspot before the microrobot moves to the hotspot and after the robot has stabilized at the hotspot. Once the robot reaches the hot spot, it rotates in place within the area to agitate the cooling medium. This allows the heat from the hot spot to be quickly transferred to the cooling medium, and then carried away by the flow of the medium, thus achieving effective heat dissipation for the chip hot spot. However, if the calculated change in heat flux does not achieve the expected heat dissipation effect, the rotation frequency of the magnetic field can be adjusted to indirectly regulate the robot's rotation frequency. This allows the robot to rotate faster, enabling it to more efficiently remove heat from the hot spot through the cooling medium.

[0109] This application determines the initial heat flux change of the robot at the hot spot based on the following formula. :

[0110]

[0111]

[0112]

[0113]

[0114]

[0115]

[0116]

[0117] in, The first temperature; The initial temperature of the cooling medium. For the size of the robot, The thermal conductivity of the cooling medium The convective heat transfer coefficient at the hotspot is given before the robot moves to it; Gr is the Grashof number; Pr is the Prandtl number. The dynamic viscosity coefficient, It is thermal conductivity; The second temperature; The kinematic viscosity coefficient of the cooling medium. This is the second spin frequency of the microrobot. Once the robot stabilizes in the hot spot... , is the second target rotation frequency of the magnetic field.

[0118] Furthermore, the system provided in this application also includes an inlet temperature sensor, an outlet temperature sensor, and a flow sensor connected to the processor; the inlet temperature sensor is located at the inlet of the heat dissipation channel, and the outlet temperature sensor is located at the outlet of the heat dissipation channel; the flow sensor is used to measure the flow rate of the cooling medium in the heat dissipation channel; the processor provided in this application is also used for:

[0119] Based on the temperature data measured by the inlet and outlet temperature sensors, and the data measured by the flow sensor, the initial heat exchange efficiency of the robot at the hot spot is determined; based on this initial heat exchange efficiency, the efficiency is continuously adjusted. The third target rotational frequency is determined until the current heat exchange efficiency meets the expected heat exchange efficiency.

[0120] Specifically, this application determines the initial heat exchange efficiency of the robot at the hot spot according to the following formula:

[0121]

[0122]

[0123]

[0124]

[0125] in, This refers to the actual heat exchange between the cooling medium and the chip. This represents the maximum heat transfer between the cooling medium and the chip. The area through which the cooling medium flows on the chip. The average temperature difference between the chip surface and the cooling medium. This represents the average temperature at the hot spot of the chip. The temperature before the cooling medium enters the chip. The temperature after the cooling medium flows out of the chip. For the cooling medium flow rate, The specific heat capacity of the cooling medium is the flow rate. This represents the highest temperature at the hot spot of the chip.

[0126] Specifically, heat transfer efficiency refers to the efficiency with which heat is transferred from a high-temperature object (here, the chip hotspot) to a low-temperature object (cooling medium), reflecting the thoroughness of the heat exchange process. In essence, when the cooling medium flows past the chip hotspot, heat is transferred from the chip to the cooling medium. Higher heat transfer efficiency means more heat is transferred from the chip hotspot to the cooling medium per unit time, resulting in a faster temperature drop at the chip hotspot. Conversely, lower heat transfer efficiency means less efficient heat transfer, making it difficult to effectively reduce the chip hotspot temperature. When the calculated heat transfer efficiency does not achieve the expected heat dissipation effect, the rotation frequency of the magnetic field can be adjusted to indirectly regulate the robot's rotation frequency, thereby increasing the robot's rotation speed and further improving heat transfer efficiency, ultimately achieving a more efficient heat dissipation effect.

[0127] It should be noted that when adjusting the rotation frequency of the magnetic field, it is necessary to ensure that the rotation frequency of the magnetic field is less than the critical rotation frequency of the magnetic field, while the critical rotation frequency of the magnetic field is equal to the critical self-rotation frequency of the robot. .

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A micro-robot-based chip hotspot heat dissipation system, characterized in that, The chip hotspot heat dissipation system comprises: a magnetic field control device for generating a uniform magnetic field; the chip is placed in the uniform magnetic field; a heat dissipation channel arranged on the surface of the chip, and a cooling medium for convection heat exchange of the hotspot is circulated in the heat dissipation channel; a robot made of magnetic material and placed in the cooling medium; a processor for controlling the magnetic field control device to generate a magnetic field at a first target rotation frequency to drive the robot to run in the cooling medium to the hotspot area of the chip; and for controlling the magnetic field control device to generate a magnetic field at a second target rotation frequency to drive the robot to rotate in place in the hotspot area of the chip to dissipate heat from the hotspot of the chip; the processor is further configured to: determine the driving force and the toks resistance of the robot in the magnetic field, respectively; determine the first self-rotation frequency of the robot according to the driving force and the toks resistance; determine the first target rotation frequency of the magnetic field according to the first self-rotation frequency of the robot, and drive the robot to run in the cooling medium to the hotspot area of the chip at the first target rotation frequency.

2. The micro-robot-based chip hot-spot heat dissipation system of claim 1, wherein, further comprising an infrared thermal imaging system and an image acquisition device electrically connected to the processor; the infrared thermal imaging system is used for scanning the surface of the chip to obtain an infrared thermal imaging image, and transmitting the infrared thermal imaging image to the processor; the image acquisition device is used for acquiring a chip surface image and transmitting the chip surface image to the processor, so that the processor locates the hotspot coordinates of the chip according to the infrared thermal imaging image and the chip surface image.

3. The micro-robot-based chip hot-spot cooling system of claim 1, wherein, the driving force of the robot in the magnetic field is: ； the toks resistance of the robot in the magnetic field is: ； wherein, is the driving force experienced by the robot in the magnetic field, is the structural shape factor of the robot, is the first spin rotation frequency of the micro robot, denotes the dynamic viscosity coefficient of the cooling medium, is the equivalent radius of the robot; is the drag force experienced by the robot in the magnetic field, v is the running speed of the robot.

4. The micro-robot-based chip hotspot heat dissipation system according to claim 2, wherein the infrared thermal imaging system is further configured to measure a first temperature at the hotspot when the robot does not reach the hotspot of the chip, and a second temperature at the hotspot after the robot stably rotates in the hotspot area of the chip; the processor is further configured to acquire all chip surface images collected by the image acquisition device at a preset frequency to determine whether the robot reaches a stable rotation state in the hotspot area of the chip; after reaching the stable state, the second temperature is obtained by the infrared thermal imaging system; determine the initial heat flux change of the robot at the hotspot according to the first temperature and the second temperature; according to the initial heat flux change, continuously adjust the current second self-rotation frequency of the magnetic field control device, and drive the robot to continuously rotate in the hotspot area of the chip at the uniform magnetic field corresponding to the second self-rotation frequency until the current heat flux change meets the expected heat dissipation target.

5. The micro-robot-based chip hot-spot heat dissipation system of claim 4, wherein, the determination of the initial heat flux change of the robot at the hotspot according to the first temperature and the second temperature comprises: determine the heat flux change of the robot at the hotspot according to the following formula: ； ； ； ； ； ； ； wherein, is a first temperature; is an initial temperature of the cooling medium, is a size of the robot, is a thermal conductivity of the cooling medium, is a convective heat transfer coefficient at the hot spot before the robot moves to the hot spot; Gr is a Grashof number; Pr is a Prandtl number, is a dynamic viscosity coefficient, is a thermal conductivity; is a second temperature; is a second spin rotation frequency of the micro robot, is a kinematic viscosity coefficient of the cooling medium; is a specific heat capacity of the cooling medium.

6. The micro-robot-based chip hot-spot heat dissipation system of claim 1, wherein, further comprising an inlet temperature sensor, an outlet temperature sensor and a flow sensor connected to the processor; The inlet temperature sensor is arranged at an inlet of the heat dissipation channel, and the outlet temperature sensor is arranged at an outlet of the heat dissipation channel; and the flow sensor is used for measuring the flow of the cooling medium in the heat dissipation channel. The processor is further configured to: determine the initial heat exchange efficiency of the robot at the hotspot according to the temperature data measured by the inlet temperature sensor and the outlet temperature sensor respectively and the data measured by the flow sensor; According to the initial heat exchange efficiency, the third target rotational frequency is constantly adjusted until the current heat exchange efficiency meets the expected heat exchange efficiency. of the third target rotational frequency until the current heat exchange efficiency meets the expected heat exchange efficiency.

7. The micro-robot-based chip hot-spot heat dissipation system of claim 6, wherein, determining the initial heat exchange efficiency of the robot at the hotspot according to the temperature data measured by the inlet temperature sensor and the outlet temperature sensor respectively and the data measured by the flow sensor includes: determining the initial heat exchange efficiency of the robot at the hotspot according to the following formula: ； ； ； ； wherein, Q is the actual heat exchange between the cooling medium and the chip, Qmax is the maximum heat exchange between the cooling medium and the chip, A is the area through which the cooling medium flows over the chip, ΔT is the average temperature difference between the chip surface and the cooling medium, T is the average temperature at the hot spots of the chip, Tin is the temperature of the cooling medium before entering the chip, Tout is the temperature of the cooling medium after leaving the chip, F is the flow rate of the cooling medium, Cv is the specific heat capacity of the cooling medium, Thot is the maximum temperature at the hot spots of the chip.

8. The micro-robot-based chip hot spot heat dissipation system according to any one of claims 1-7, wherein, The heat dissipation channel is a straight channel, a serpentine channel, a tree-shaped channel or a grid-shaped channel.

Images