Adaptive Microchannel Heat Exchanger for Non-Uniform Dynamic Heat Distribution and Its Method for Intelligent Perception of Global Temperature

Abstract

The present invention belongs to the technical field of microchannel cooling, and particularly relates to an adaptive microchannel heat exchanger for non-uniform dynamic heat distribution and its global temperature intelligent perception method. The bottom plate (2) is in contact with the components to be cooled, and a global intelligent perception and control system is arranged on the components; the microchannel layer (1) is arranged at the upper end of the bottom plate (2), the cover plate (3) is arranged at the top of the microchannel layer (1), and the four sides of the microchannel layer (1) are respectively provided with a typical working condition inlet I (4), a typical working condition inlet II (5), a typical working condition inlet III (6), and an outlet (7). The outlet (7) is selected such that each typical working condition inlet enters the microchannel and finally converges to the outlet for outflow. The present invention is used to solve the problem of intelligent heat dissipation under spatially non-uniform multi-source heat loads, and can set multiple cooling working fluid inlets according to several typical working heat loads and perform adaptive switching during actual operation to achieve real-time, accurate, and efficient heat dissipation effects.

Description

Technical Field

[0001] The present invention belongs to the technical field of microchannel cooling, and particularly relates to an adaptive microchannel heat exchanger for non-uniform dynamic heat distribution and an intelligent method for globally sensing temperature thereof. Background Art

[0002] In recent years, with the development of advanced manufacturing technologies, electronic devices and components have been developing towards miniaturization and integration, which has led to an exponential increase in their heat generation. This poses higher requirements for the performance of heat sinks in limited spaces. If heat cannot be removed in a timely and effective manner, it will cause the temperature of the components to rise, thereby greatly reducing their reliability and even causing component failures or damages. With the large-scale application of high heat flux density components in various fields, such as very large scale integrated circuits, new energy vehicle battery packs, military lasers, and radar arrays, traditional heat dissipation methods are difficult to meet their heat dissipation requirements. Therefore, there is an urgent need to develop efficient cooling technologies to ensure the safe and reliable operation of high heat flux density components.

[0003] The cooling methods for high heat flux components mainly include forced air cooling, heat pipe phase change cooling, microchannel cooling, porous medium cooling, spray cooling, and jet impingement cooling. Among them, microchannel heat dissipation technology, as a highly efficient thermal control technology with a compact structure, stable operation, and lightweight, has received extensive attention in the development of heat dissipation technologies for compact space electronic devices.

[0004] The integrated development of electronic components has led to the generation of non-uniform dynamic heat distribution. When the workload changes, the cooling solutions originally designed for a single thermal condition will perform poorly, and it is also unrealistic to design multiple different heat sink structures for different workloads. In addition, a globally controllable thermal state is becoming increasingly important for the normal operation of electronic devices, which largely depends on efficient and global spatio-temporal thermal feedback. Considering the overall system design specifications, measurement feasibility, or layout rationality, the number and location of temperature sensors are often limited. The temperature measurement method based on discrete sensors can only obtain the local temperature at the measurement points and is difficult to describe the global state in real time, resulting in thermal control problems such as locality, time lag, and overshoot. For example, in traditional heat sinks, the coolant is usually delivered to all potential heat generation areas of electronic components, resulting in some coolant being dispersed to areas that do not require cooling. Therefore, it is necessary to design an intelligent microchannel heat sink that can reconstruct the global temperature in real time according to the changes in thermal workload and the data of discrete temperature sensors and provide a reliable heat dissipation solution in a timely manner. Summary of the Invention

[0005] The present invention provides an adaptive microchannel heat exchanger for non-uniform dynamic heat distribution and its global temperature intelligent sensing method, which is used to solve the intelligent heat dissipation problem under spatial non-uniform multi-source heat loads. Multiple cooling working fluid inlets can be set according to several typical working heat loads and adaptively switched during actual operation to achieve real-time, accurate and efficient heat dissipation effects.

[0006] The present invention is realized through the following technical solutions:

[0007] An adaptive microchannel heat exchanger for non-uniform dynamic heat distribution, the adaptive microchannel heat exchanger includes a microchannel layer, a bottom plate, a cover plate, typical working condition inlet I, typical working condition inlet II, typical working condition inlet III, an outlet and a global intelligent sensing and control system. The bottom plate is in contact with the components that need heat dissipation, and the global intelligent sensing and control system is arranged on the components.

[0008] The microchannel layer is arranged at the upper end of the bottom plate, the cover plate is arranged at the top of the microchannel layer, the four sides of the microchannel layer are respectively provided with typical working condition inlet I, typical working condition inlet II, typical working condition inlet III and the outlet. The selection of the outlet is that the fluid entering the microchannel from each typical working condition inlet will finally converge at the outlet and flow out.

[0009] Further, the global intelligent sensing and control system includes temperature sensors. A plurality of the temperature sensors are arranged on the components through heat characteristics. The heat characteristics find the optimal set of sensor positions through an optimization algorithm and a judgment criterion to determine the positions of each sensor.

[0010] Further, the optimization algorithm includes a greedy algorithm or a genetic algorithm;

[0011] The judgment criterion includes maximizing the determinant, minimizing the condition number and minimizing the mean square error.

[0012] A global temperature intelligent sensing method for an adaptive microchannel heat exchanger for non-uniform dynamic heat distribution, the global temperature intelligent sensing method includes the following steps:

[0013] Step 1: Obtain the heat characteristics of different loads to obtain a heat characteristic set;

[0014] Step 2: Select the most important heat characteristics based on the heat characteristic set in Step 1;

[0015] Step 3: Based on the most important heat characteristics in Step 2, determine the layout of the temperature sensors;

[0016] Step 4: Based on the layout of the sensors in Step 3, combine each heat characteristic according to the real-time data of the discrete temperature sensors and the working load information to obtain the reconstruction of the global temperature field, that is, perform the reconstruction of the global temperature field.

[0017] Specifically, the global temperature intelligent perception is realized based on the process of capturing the physical state characteristics of in-service components with multiple degrees of freedom from limited data, constructing a low-order temperature estimation model, and dynamically adjusting the corresponding coefficients. The multiple degrees of freedom mean that the geometric forms, physical characteristics, etc. of the temperature-sensing components can be diverse.

[0018] Further, step 1 is specifically to obtain the prior temperature field data Φ of the electronic device under different loads in the working condition in advance through experimental or simulation methods m×n , and use feature learning or feature decomposition methods to obtain the thermal feature sets of different loads

[0019] Further, step 2 is specifically to select r most important thermal features from the final thermal feature set which can effectively filter out the interference caused by noise in the prior data Φ to the thermal features. m×n

[0020] Further, step 3 is specifically to find the optimal set of sensor positions through an optimization algorithm and a decision criterion to determine the positions of each sensor.

[0021] Further, step 4 is specifically to transform the reconstruction into an objective optimization, that is, to minimize the Euclidean distance between the real temperature field T and the approximate temperature field T r to obtain the optimal low-dimensional representation coefficient μ r ,

[0022]

[0023] where, is the real-time discrete temperature measurement data set; is the Mth temperature measurement data of the discrete measurement point set S = {s1, s2,..., s M}.

[0024] Further, step 4 is specifically to use an accurate and fast matrix operation solver to perform global temperature field reconstruction based on the following formula and the real-time discrete temperature measurement data;

[0025]

[0026] where,

[0027] Furthermore, the global intelligent perception and control system controls the opening / closing of each inlet of the microchannel heat exchanger and the flow rate of the cooling working medium according to the working heat load distribution, global temperature distribution, and heat flux density of the components until the working heat load of the components changes.

[0028] The beneficial effects of the present invention are as follows:

[0029] By real-time sensing the heat source and temperature distribution, the present invention adjusts the opening and closing states of each inlet of the microchannel radiator, and timely switches the inlet positions according to the change of the working heat load and the real-time global heat distribution. At the same time, the flow rate of the cooling working medium is adjusted according to the change of the heat flux density in the area to be cooled, thereby enhancing the heat dissipation effect of the microchannel heat exchanger and making the temperature of the components meet the range of their working temperature zones.

[0030] Since the present invention can adjust the key cooling area of the heat exchanger in real time according to the hot spots and global temperature distribution, it can solve the problem that the traditional single-inlet microchannel heat exchanger has poor cooling effect on the non-uniform distribution of multiple hot spots, realize the efficient utilization of the cooling flux, and thus save energy.

[0031] Taking the genetic algorithm as the optimization technique and the mean square error as the judgment criterion. Randomly initialize the sensor combination, take the minimization of the mean square error of the reconstruction result as the objective function, continuously adjust the sensor set through genetic algorithm optimization, perform multiple temperature reconstruction processes, and finally find a better sensor layout. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is the main structural view of the present invention.

[0033] Figure 2 is the three-dimensional schematic diagram of the present invention

[0034] Figure 3 is the schematic diagram of the microchannel structure of the present invention.

[0035] Figure 4 is the flow chart of the implementation method of the global intelligent perception of the invention.

[0036] Figure 5 is the working schematic diagram of the microchannel under several typical thermal loads, where (a) is the schematic diagram of the cooling working medium exchanging heat according to the designed flow channel when the heat source 1 is turned on, (b) is the schematic diagram of the cooling working medium exchanging heat according to the designed flow channel when the heat source 2 is turned on, and (c) is the schematic diagram of the cooling working medium exchanging heat according to the designed flow channel when the heat source 3 is turned on. DETAILED DESCRIPTION OF THE INVENTION

[0037] In the following description, specific details such as specific system architectures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0038] It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0039] It should also be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0040] The following combines the appended Figures 1-5 , and clearly and completely describes the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0041] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present application. Therefore, the present application is not limited by the specific embodiments disclosed below.

[0042] Embodiment 1

[0043] This embodiment provides an adaptive microchannel heat exchanger for non-uniform dynamic heat distribution. As Figures 1-3 shown, the adaptive microchannel heat exchanger includes a microchannel layer 1, a bottom plate 2, a cover plate 3, a typical working condition inlet I 4, a typical working condition inlet II 5, a typical working condition inlet III 6, an outlet 7, and a global intelligent perception and control system. The bottom plate 2 is in contact with the components that need to dissipate heat, and the global intelligent perception and control system is arranged on the components;

[0044] The thermal load distribution of the components determines the operating conditions of the heat exchanger; each inlet of the microchannel is a typical operating condition, and the inlets correspond to the respective typical operating conditions. Different cooling working fluid inlets are switched according to different working thermal loads; the outlet is located on the side of the heat exchanger where there is no distribution of the cooling working fluid inlet.

[0045] A microchannel layer 1 is provided at the upper end of the bottom plate 2, a cover plate 3 is provided at the top of the microchannel layer 1, and typical operating condition inlets I 4, typical operating condition inlets II 5, typical operating condition inlets III 6, and an outlet 7 are respectively provided on the four side surfaces of the microchannel layer 1. The selection of the outlet 7 is such that the working fluid entering through each typical operating condition inlet will finally converge at the outlet and flow out.

[0046] The global intelligent perception and control system controls the opening of the corresponding inlet by sensing the position of the thermal load, so that the cooling working fluid is mainly distributed in the high heat flux density area, realizing efficient cooling and improving the working efficiency.

[0047] The microchannel heat exchanger has a plurality of cooling working fluid inlets, which are distributed on different side surfaces of the heat exchanger.

[0048] The structural design method of the microchannel is uniform topology optimization design, that is, an optimization algorithm for minimizing the objective function. Considering the cooling requirements of each typical thermal operating condition, and under the condition of specifying the total inlet pressure of the radiator, the weighted sum of each working load thermal resistance is used as the multi-objective cost function to be optimized, and the microchannel structure is designed by minimizing the objective function.

[0049] The microchannels in the microchannel layer include main channels and secondary channels. The change in the inlet position under different thermal operating conditions mainly depends on the cooling area of the main channels, and the secondary channels are cross-shared under each operating condition.

[0050] Further, prior data of the temperature field of the electronic device under different loads in certain operating conditions is obtained in advance, and feature learning or feature decomposition methods (including but not limited to SVD, dictionary learning, etc.) are used to obtain the thermal features of different loads.

[0051] The global intelligent perception and control system includes temperature sensors. A plurality of the temperature sensors are arranged on the components according to the thermal features. The optimal set of sensor positions is found through an optimization algorithm and a decision criterion to determine the positions of each sensor.

[0052] Specifically, the main thermal features are selected. Through optimization techniques (such as greedy algorithm, genetic algorithm, etc.) and the given decision criterion, the optimal set S = {s1, s2, …, s M} of sensor positions is found, so as to maximize the reconstruction performance.

[0053] Further, the optimization algorithm includes a greedy algorithm, a genetic algorithm, etc.;

[0054] The decision criteria include maximizing the determinant, minimizing the condition number, minimizing the mean square error, etc.

[0055] Specifically, taking the greedy algorithm as the optimization technique, maximizing the determinant of is used as the decision criterion. Among them, is related to The corresponding pruning matrix is given by the following formula:

[0056]

[0057] Among them, the elements in the i-th row of the mask matrix M are all 0, and only s i is 1, satisfying the idempotency.

[0058] The set of sensor positions that maximizes the determinant of is found through the greedy algorithm. Specifically, starting from the empty sensor position set S, at the t-th iteration step (t = 1,..., k), when each unselected sensor position is added to the set S, the corresponding determinant is calculated, and the position that results in the maximum determinant is selected as the t-th item of the set S until k sensors are determined.

[0059] Taking the genetic algorithm as the optimization technique and the mean square error as the decision criterion. The sensor combination is randomly initialized, with the minimization of the mean square error of the reconstruction result as the objective function. The sensor set is continuously adjusted through genetic algorithm optimization, and multiple temperature reconstruction processes are carried out to finally find a better sensor layout.

[0060] Embodiment 2

[0061] This embodiment provides a global temperature intelligent sensing method for an adaptive microchannel heat exchanger for non-uniform dynamic heat distribution. This embodiment uses an adaptive microchannel heat exchanger for non-uniform dynamic heat distribution as described in Embodiment 1. The implementation of the global temperature intelligent sensing is divided into two parts: offline preparation and online reconstruction; as Figure 4 shown,

[0062] The offline preparation is to obtain the temperature distribution results under different loads and environmental conditions through experiments / numerical simulations in advance, utilize the acquired prior heat characteristics, and determine the temperature sensor layout;

[0063] The online reconstruction part is to perform fast and accurate iterations in actual temperature monitoring by combining the discrete temperature sensor data to generate the global temperature distribution that conforms to the temperature measurement results in real time;

[0064] The global temperature intelligent sensing method includes the following steps, ​

[0065] Step 1: Obtain the thermal characteristics of different loads to get a set of thermal characteristics;

[0066] Step 2: Select the most important thermal characteristics based on the set of thermal characteristics in Step 1;

[0067] Step 3: Based on the most important thermal characteristics in Step 2, determine the layout of temperature sensors;

[0068] Step 4: Based on the layout of the sensors in Step 3, combine each thermal characteristic according to the real-time data of discrete temperature sensors and the workload information to obtain the reconstruction of the global temperature field, that is, perform the reconstruction of the global temperature field.

[0069] Specifically, the global temperature intelligent perception is realized based on the process of capturing the physical state characteristics of in-service components with multiple degrees of freedom from limited data, constructing a low-order temperature estimation model and dynamically adjusting the corresponding coefficients. The multiple degrees of freedom mean that the geometric forms, physical characteristics, etc. of the temperature-sensing components can be diverse.

[0070] Further, Step 1 is specifically to obtain the prior data Φ of the temperature field of the electronic device under different loads in the working condition in advance through experimental or simulation methods m×n , and use feature learning or feature decomposition methods to obtain a set of thermal characteristics of different loads

[0071] Further, Step 2 is specifically to select r most important thermal characteristics from the final set of thermal characteristics This process can not only achieve the low-order estimation of the temperature field with less computational effort, but also effectively filter out the interference caused by the noise in the prior data Φ to the thermal characteristics. m×n

[0072] Further, Step 3 is specifically to find the optimal set of sensor positions through an optimization algorithm and a decision criterion to determine the positions of each sensor.

[0073] Further, Step 4 is specifically to transform the reconstruction into an objective optimization, that is, to minimize the Euclidean distance between the real temperature field T and the approximate temperature field T r to obtain the optimal low-dimensional representation coefficient μ r ,

[0074]

[0075] where is the real-time discrete temperature measurement data set; is the Mth temperature measurement data of the discrete measurement point set S = {s1, s2,..., s M}.

[0076] Further, step 4 specifically is to use an accurate and fast matrix operation solver to reconstruct the global temperature field based on the following formula and real-time discrete temperature measurement data;

[0077]

[0078] wherein,

[0079] Here, the acquisition of the low-dimensional representation coefficient μ r is only for example, and other optimization or interpolation methods can also be used to solve the target optimization problem in step 3 to obtain it.

[0080] Further, the global intelligent perception and control system controls the opening / closing of each inlet of the microchannel heat exchanger and the flow rate of the cooling working medium according to the working heat load distribution, global temperature distribution and heat flux density of the components until the working heat load of the components changes.

[0081] Specifically, as Figure 1 shown, an adaptive microchannel heat exchanger with multiple inlets, the microchannel heat exchanger includes a bottom plate, microchannels, a cover plate, an outlet and multiple inlets. The heat load generated by the electronic components acts on the lower surface of the bottom plate of the microchannel heat exchanger. The opening / closing of each inlet is controlled according to the heat load position, and the flow rate of the cooling working medium is controlled according to the heat load magnitude.

[0082] As Figure 5 (a) shown, when the working load of the electronic component is heat source 1, the new adaptive heat exchanger enables the cooling working medium inlet corresponding to the microchannels densely distributed at heat source 1 through the control system, that is, inlet Ⅰ4, and at the same time closes inlet Ⅱ5 and inlet Ⅲ6, so that the cooling working medium flows and exchanges heat according to the designed microchannels; at the same time, the heat characteristics when the load is heat source 1 are enabled, and the global temperature field is reconstructed based on the real-time measurement information of the discrete temperature sensors, and the working medium with the best flow rate is matched by integrating the existing heat field information, so that a larger flow rate is used for heat dissipation, effectively improving the working efficiency of the radiator

[0083] Based on the adaptive microchannel heat exchanger, when the working load of the electronic component is heat source 2, as Figure 5 (b) shown, the heat exchanger enables the cooling working medium inlet corresponding to the microchannels densely distributed at heat source 2 through the control system, that is, inlet Ⅱ5, and at the same time closes inlet Ⅰ4 and inlet Ⅲ6, so that the cooling working medium exchanges heat according to the designed channels; at the same time, the heat characteristics when the load is heat source 2 are enabled, and the global temperature field is reconstructed based on the real-time measurement information of the discrete temperature sensors, and the working medium with the best flow rate is matched by integrating the existing heat field information, so that a larger flow rate is used for heat dissipation, greatly improving the working efficiency of the radiator.

[0084] Based on the adaptive microchannel heat exchanger, when the working load of the electronic component is heat source 3, as Figure 5 (c) shows, the heat exchanger opens the cooling working fluid inlets corresponding to the microchannels that are more densely distributed at heat source 3 through the control system, that is, inlet III 6, and at the same time closes inlet I 4 and inlet II 5, so that the cooling working fluid exchanges heat according to the designed flow channels; at the same time, the thermal characteristics when the load is heat source 3 are enabled, and the global temperature field is reconstructed based on the real-time measurement information of the discrete temperature sensors, and the working fluid with the optimal flow rate is matched by integrating the existing thermal field information, so that a larger flow rate is used for heat dissipation, greatly improving the working efficiency of the radiator.

Claims

1. An adaptive microchannel heat exchanger for non-uniform dynamic heat distribution, characterized in that: The adaptive microchannel heat exchanger comprises a microchannel layer (1), a base plate (2), a cover plate (3), a typical working condition inlet I (4), a typical working condition inlet II (5), a typical working condition inlet III (6), an outlet (7) and a global intelligent sensing and control system, wherein the base plate (2) is in contact with components that require heat dissipation, and the global intelligent sensing and control system is arranged on the components; A microfluidic layer (1) is arranged at the upper end of the base plate (2), a cover plate (3) is arranged at the top end of the microfluidic layer (1), and four sides of the microfluidic layer (1) are respectively provided with a typical working condition inlet I (4), a typical working condition inlet II (5), a typical working condition inlet III (6) and an outlet (7), wherein the outlet (7) is selected so that each typical working condition inlet entering the microfluidic channel will finally converge at the outlet for outflow; The global intelligent perception and control system includes a temperature sensor, and a plurality of the temperature sensors are arranged on components through thermal characteristics. The thermal characteristics find the optimal set of sensor positions through an optimization algorithm and a judgment criterion to determine the position of each sensor; The global intelligent perception and control system senses the position of the heat load and controls the opening of the corresponding inlet, so that the cooling medium is mainly distributed in the high heat flux density area.

2. The adaptive microchannel heat exchanger according to claim 1, characterized in that: The optimization algorithm includes a greedy algorithm or a genetic algorithm; The decision criteria include maximizing the determinant, minimizing the condition number, and minimizing the mean square error.

3. A global temperature intelligent sensing method for an adaptive microchannel heat exchanger for non-uniform dynamic heat distribution according to any one of claims 1-2, characterized in that: The global temperature intelligent sensing method comprises the following steps: Step 1: Obtain thermal characteristics of different loads to obtain a thermal characteristic set; Step 2: Select the most important thermal features based on the thermal feature set in step 1; Step 3: Based on the most important thermal characteristics of step 2, establish the temperature sensor layout; Step 4: Based on the layout of the sensors in step 3, the thermal features are combined according to the real-time data and workload information of the discrete temperature sensors to obtain the reconstruction of the global temperature field, that is, the global temperature field is reconstructed; Specifically, step 3 is to find the optimal set of sensor positions through an optimization algorithm and a decision criterion to determine the position of each sensor; Specifically, the step 4 is to use an accurate and fast matrix operation solver to reconstruct the global temperature field based on the following formula and real-time discrete temperature measurement data; in, ; The global intelligent perception and control system controls the opening / closing of each inlet of the microchannel heat exchanger and the flow rate of the cooling medium according to the working heat load distribution of the components, the global temperature distribution and the heat flux density until the working heat load of the components changes.

4. The global temperature intelligent sensing method according to claim 3 is characterized in that: The step 1 specifically involves obtaining the temperature field priori data of the electronic device under different loads under working conditions in advance through experiments or simulation methods. , using feature learning or feature decomposition methods to obtain thermal feature sets of different loads .

5. The global temperature intelligent sensing method according to claim 4 is characterized in that: The step 2 is specifically as follows: choose r The main thermal characteristics , which can effectively filter out prior data The interference caused by medium noise to thermal characteristics.

6. The global temperature intelligent sensing method according to claim 3 is characterized in that: Specifically, step 4 is to transform the reconstruction into a target optimization, that is, to minimize the difference between the true temperature field T and the approximate temperature field T r The Euclidean distance between them is used to obtain the optimal low-dimensional representation coefficient , in, It is a real-time discrete temperature measurement data set; A set of discrete measurement points No. M Temperature measurement data.

Images