Circuit board thermal management optimization method combined with quantum computing

Abstract

The invention discloses a circuit board heat management optimization method combined with quantum computing, which comprises the following steps of: predicting heat flow distribution of each heat source on a circuit board through a quantum optimization algorithm, identifying a potential overheat area, allocating cooling resources in advance, and determining whether the circuit board is abnormal or not according to a local temperature abnormal condition. A quantum gate control algorithm is adopted to adjust the conduction path and rate of heat, redundant heat is guided to a low-temperature area so as to avoid heat accumulation, the system monitors the temperature change and power consumption fluctuation of a circuit board in real time, parameters of cooling equipment are dynamically adjusted according to a feedback result, the redundant heat is converted into electric energy through a heat energy recovery device and stored, and the cooling efficiency is improved. When the power consumption is low, the heat flow is released to balance the overall heat flow of the circuit board and prolong the service life of equipment. The invention aims to realize more efficient heat dissipation management through a quantum algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of heat dissipation management of electronic equipment, and in particular to a circuit board thermal management optimization method combined with quantum computing. Background Art

[0002] As circuit boards in high-performance computing and data center servers gradually carry higher power densities, especially when multi-core processors, graphics processors and other high-power devices are running at the same time, a lot of heat will be generated. Traditional heat dissipation methods, such as air cooling and liquid cooling systems, can provide basic heat dissipation functions, but these systems have a slow response speed and cannot respond to local overheating in real time. When the temperature of a certain area on the circuit board suddenly rises, traditional systems often react slowly and cannot quickly adjust the heat dissipation strategy, causing the local overheating problem to be further aggravated. This delayed heat dissipation response will have a negative impact on the long-term reliability and service life of the equipment, and in severe cases may even cause system performance degradation or downtime.

[0003] Existing heat dissipation management systems lack the ability to dynamically adjust the heat conduction path. Traditional heat dissipation methods often rely on fixed heat dissipation paths, that is, heat is conducted from high-power consumption components through thermal conductive materials or cooling media to the radiator or cooling system, and then discharged through fans or liquid cooling. This fixed heat conduction path cannot be dynamically optimized according to actual conditions. When the temperature in certain areas rises, the heat is concentrated in the high-temperature area and cannot be quickly discharged. The cooling resources in the low-temperature area cannot be reasonably utilized.

[0004] Traditional heat dissipation management systems usually rely on preset heat dissipation strategies. The allocation of cooling resources is adjusted based on the average power consumption or global temperature of the device. This preset strategy is too rigid when the device load changes, and it is difficult to flexibly adapt to the temperature fluctuations of various areas on the circuit board. When the temperature of different areas of the circuit board is uneven, the system cannot adjust the cooling requirements of each area in real time, which may result in insufficient cooling of high-temperature areas and excessive cooling of low-temperature areas. Most traditional systems use simple temperature feedback control. This feedback mechanism cannot effectively handle complex heat flow distribution and instantaneous power consumption changes, resulting in uneven allocation of cooling resources, which in turn leads to reduced system efficiency.

[0005] Existing technologies lack an effective heat recovery mechanism. High-power consumption equipment generates a large amount of heat during operation. Traditional systems usually discharge this excess heat directly into the external environment through a radiator or fan without recovering and utilizing the heat. This method not only wastes precious energy, but also increases the energy consumption of the overall system. Although some systems have introduced thermoelectric conversion equipment, most thermal management systems do not closely integrate the heat recovery device with the cooling system, resulting in low utilization of heat energy and failure to effectively reduce the energy consumption of the system.

[0006] Most of the heat dissipation optimization algorithms in the existing technology are based on classical optimization methods, such as finite element analysis in thermal flow analysis or simple feedback control algorithms. These classical algorithms have low processing efficiency when faced with complex interactions among multiple heat sources and nonlinear heat conduction relationships, and it is difficult to quickly find the global optimal solution. In an environment with high power consumption and multiple heat sources, the limitations of this algorithm are particularly prominent, making it difficult for the system to make efficient cooling decisions in a short period of time, affecting the overall effect of heat dissipation management.

[0007] Therefore, how to provide a circuit board thermal management optimization method combined with quantum computing is an urgent problem that technicians in this field need to solve. Summary of the invention

[0008] One object of the present invention is to propose a circuit board thermal management optimization method combined with quantum computing. The present invention makes full use of quantum optimization algorithms and real-time feedback adjustment technology, and describes in detail the global prediction of circuit board heat flow distribution through multi-body entangled state quantum optimization algorithm, the optimization of heat conduction path by quantum gate thermal energy conduction control algorithm, and the intelligent heat dissipation management scheme of dynamically adjusting cooling equipment parameters and heat recovery devices. This method has the advantages of fast response speed, high heat dissipation efficiency, reasonable allocation of cooling resources and high energy utilization.

[0009] A circuit board thermal management optimization method combined with quantum computing according to an embodiment of the present invention includes the following steps:

[0010] S1. Use the multi-body entangled state quantum optimization algorithm to globally predict the heat flow distribution in different areas of the circuit board, capture the nonlinear heat conduction relationship between each heat source, and build a heat flow mapping model. The algorithm can process the interaction of multiple heat sources in parallel, identify potential overheating areas, and pre-allocate cooling resources;

[0011] S2. Based on the constructed heat flow mapping model, when an abnormal temperature occurs in a certain area of ​​the circuit board, the quantum gate heat conduction control algorithm is used to optimize the local heat conduction path, intelligently adjust the heat conduction direction and rate, and guide the excess heat of the identified overheated area to the low-temperature area to avoid heat accumulation and optimize the heat dissipation efficiency;

[0012] S3. According to the heat flow prediction results and heat energy conduction adjustment, dynamically adjust the cooling equipment parameters of different cooling areas, control the cooling requirements of high-heat areas and low-heat areas, realize accurate allocation of cooling resources, optimize the workload of each cooling module, and avoid resource waste;

[0013] S4. By real-time monitoring of temperature changes and power consumption fluctuations in each area, combined with quantum optimization results, adaptively balance the thermal energy, store excess heat in the heat recovery device, and release it when power consumption is low, balance the overall heat flow of the circuit board, and extend the service life of the circuit board;

[0014] S5. Based on real-time temperature feedback and the dynamic release effect of stored thermal energy, the quantum optimized thermal management strategy is updated regularly, and the cooling path is recalculated through the multi-body entangled state quantum optimization algorithm to maintain the dynamic balance of the heat flow mapping model when the circuit board load changes.

[0015] Optionally, the S1 specifically includes:

[0016] S11. Define n heat sources on the circuit board, denoted as H1, H2, …, H n , the heat flux of each heat source varies according to its power P i (t) and position r i =(x i ,y i ,z i ) determines the heat transfer coefficient matrix T ij (t), where the matrix element T ij (t) represents the heat source H i With heat source H j The time-dependent heat transfer coefficient between:

[0017]

[0018] Among them, γ ij represents the attenuation coefficient of heat conduction, α ij represents the initial heat conduction intensity, ω ij is the frequency of thermal fluctuations between the two heat sources;

[0019] S12. Based on the heat source parameters, a multi-body entangled state system of the circuit board is established. The quantum state |ψ> is defined as the entangled state between all heat sources to describe the global correlation of heat flow. The evolution process of the quantum state is used to capture the nonlinear heat conduction relationship between the heat sources and calculate the time evolution of each heat source.

[0020] S13, by solving the Hamiltonian of the system H = ∑ i,j T ij H i H j , simulate the heat flow interaction between various heat sources, identify areas with drastic temperature fluctuations, and mark them as potential overheating areas;

[0021] S14. Construct a heat flow mapping model to map the power, position and heat conduction coefficient of each heat source to the global heat flow field of the model. The model solves the heat conduction equation Where T is the temperature, q is the thermal power density of the heat source, and k is the thermal conductivity, which generates the global heat flux distribution prediction result;

[0022] S15. Use quantum optimization algorithm to update the parameters of heat flow mapping model in real time, including power fluctuation of each heat source, thermal conductivity coefficient T ij Dynamic changes ensure optimal heat flow distribution under different load conditions;

[0023] S16. According to the constructed heat flow mapping model, cooling resources are allocated in advance. The cooling resource vector is represented as C = (C1, C2,, C n ), where each element C i Indicates the allocation to the heat source H i The cooling amount is adjusted dynamically through quantum optimization algorithm to avoid heat concentration;

[0024] S17. During the operation of the circuit board, the heat conduction changes of each heat source are captured and analyzed in real time, the heat flow mapping model is continuously updated, and the cooling resources are dynamically adjusted according to the latest heat flow distribution prediction.

[0025] Optionally, the S2 specifically includes:

[0026] S21. Based on the constructed heat flow mapping model, when an abnormal temperature occurs in a certain area of ​​the circuit board, the temperature gradient of the overheated area is extracted, and the nonlinear heat conduction relationship between heat sources is adjusted using the quantum gate heat energy conduction control algorithm to capture and process areas with drastic temperature fluctuations;

[0027] S22. Based on the extracted temperature gradient and the prediction results of the heat flow mapping model, the quantum state |ψ hot (t)>Describe the heat conduction behavior in the overheating area and dynamically adjust the heat conduction path;

[0028] S23. Quantum gate operation is used to dynamically adjust the heat conduction path. The adjusted conduction coefficient matrix is ​​expressed as:

[0029]

[0030] Among them, T ij (t) is the initial conduction coefficient, ΔT hot is the temperature difference in the overheated area, ω ij is the thermal oscillation frequency between the two heat sources;

[0031] S24, according to the adjustment results, the excess heat Q in the overheating area hot Lead to low temperature area R cool , optimize the heat transfer rate from overheated areas to low-temperature areas by dynamically adjusting cooling resources;

[0032] S25. Based on the adjustment of the heat conduction path, through the real-time feedback mechanism, the quantum gate algorithm is used to continuously optimize the conduction path, monitor the heat flow changes in real time, make the heat conduction direction and rate dynamically adapt to the current heat flow state, and prevent heat from accumulating in the overheating area.

[0033] Optionally, the S3 specifically includes:

[0034] S31. Based on the constructed heat flow mapping model, identify the high heat area R on the circuit board hot and low heat area R cool , respectively calculate the cooling requirements of each area. The heat flux mapping model provides the prediction results of the global heat flux distribution and cooling requirements on the circuit board. For the high heat area, the cooling requirement is:

[0035]

[0036] Among them, q hot is the thermal power density of the high heat area, is the temperature gradient in the high heat area, f adjust (t) is the dynamic adjustment coefficient;

[0037] For low heat areas, the cooling requirement is:

[0038]

[0039] Among them, q cool and are the thermal power density and temperature gradient in the low heat area, g adjust (t) is the cooling adjustment factor for the low heat area;

[0040] S32, dynamically adjust the parameters of the cooling equipment according to the cooling demand forecast, including the fan speed V f (t) and the flow rate F of the liquid cooling system l (t):

[0041] V f (t) = V f (0)+γ·(C hot (t)-C current (t));

[0042] F l (t) = F l (0)+λ·(C cool (t) ΔT cool );

[0043] Among them, γ and λ are adjustment gains respectively;

[0044] S33, based on the adjustment of cooling equipment parameters, calculating the workload of different cooling modules, optimizing the allocation of cooling resources, and optimizing the load of each cooling module;

[0045] S34, monitor the temperature and cooling equipment parameters of each cooling area in real time, use quantum optimization algorithm to dynamically update the allocation strategy of cooling resources, and continuously optimize the allocation of cooling resources to avoid waste of cooling resources and imbalanced workload;

[0046] S35, based on the cooling data feedback and combined with the quantum optimization results, optimize the working efficiency of the cooling module, so that the high-heat area can obtain sufficient cooling resources, while optimizing the cooling resource allocation in the low-heat area to avoid excessive resource consumption.

[0047] Optionally, the S4 specifically includes:

[0048] S41, extract the temperature feedback parameter ΔT by real-time monitoring of the temperature changes and power consumption fluctuations of each heat source on the circuit board i (t) and power consumption fluctuation parameter P i (t), using quantum optimization algorithm to determine the excess heat of each heat source, and convert the excess heat Q excess Converted into electrical energy;

[0049] S42. The converted excess heat is stored in a heat recovery device, which includes a plurality of heat storage units, using thermal batteries and phase change materials for storage, and the excess heat is stored in the heat storage unit to avoid local overheating:

[0050] Q store (t)=∑ i (Q i (t)-Q threshold );

[0051] Among them, Q store (t) is the stored heat, Q threshold is the activation threshold of the thermal energy storage unit;

[0052] S43. When the power consumption of the circuit board is lower than a preset threshold, the heat recovery device starts to release the stored heat, and gradually releases the heat when the power consumption is low to balance the overall heat flow of the circuit board:

[0053]

[0054] Among them, T release is the heat release cycle, f release (t) is a dynamic release function;

[0055] S44. Adjust the heat storage and release process through quantum optimization algorithm, adjust the thermoelectric conversion efficiency and release frequency in combination with the circuit board load conditions, cope with the heat accumulation under high load and the release demand under low load, and achieve stable distribution of heat flow;

[0056] S45. Adjust the working parameters of the heat recovery device based on the feedback mechanism, continuously optimize the operating state of the heat recovery device, balance the temperature of each area, and extend the service life of the circuit board.

[0057] Optionally, the S5 specifically includes:

[0058] S51, through the real-time temperature feedback system, monitor the temperature changes of each heat source of the circuit board, based on the temperature feedback parameter ΔT i (t) and power consumption feedback parameter P i (t), and combined with the dynamic release effect of the heat recovery device, calculate the temperature fluctuation of each heat source and generate feedback data;

[0059] S52. Based on the feedback data, the quantum optimization algorithm sets the frequency of updating the thermal management strategy. The frequency of updating the strategy is updated through a timed update mechanism, and the thermal management strategy is updated every Δt time interval. When the temperature fluctuation amplitude of any heat source is ΔT i (t) exceeds the preset fluctuation threshold ΔT threshold Trigger update when

[0060] S53, the feedback data sources include the temperature sensor, the power consumption monitoring device and the release data of the heat recovery device, and generate the comprehensive feedback data F total (t):

[0061] F total (t) = α·f feedback (t)+β·P monitor 9t)+γ·Q release 9t);

[0062] Among them, P monitor (t) is the power consumption monitoring data, Q release 9t) is the heat release data of the heat recovery device, α, β, γ are the feedback weight coefficients;

[0063] S54, based on comprehensive feedback data F total 9t), the quantum optimization algorithm recalculates the cooling path through multi-body entangled state quantum computing:

[0064] |ψ(t)>=∑ i c i (t)|H i +∑ i,j c ij (t)|H i Hj ;

[0065] Among them, c i (t) represents the heat flux contribution of a single heat source, c ij (t) is the entanglement effect between multiple heat sources;

[0066] S55, combining the quantum optimization calculation results and the load changes of the circuit board, dynamically adjust the cooling path, and optimize the heat transfer coefficient T ij (t) and cooling equipment parameters to achieve a dynamic balance of heat flow:

[0067]

[0068] Among them, ω ij is the thermal oscillation frequency between the two heat sources;

[0069] S56, by regularly evaluating the feedback parameter F total Based on the changes in (t) and the results of quantum optimization calculations, the thermal management strategy of the circuit board is updated to maintain the dynamic balance of heat flow, avoid local overheating and optimize the heat dissipation effect.

[0070] The beneficial effects of the present invention are:

[0071] The present invention uses a multi-body entangled state quantum optimization algorithm to globally predict the heat flux distribution of multiple heat sources on a circuit board, capturing the nonlinear heat conduction relationship between the heat sources. The algorithm can process the heat flux interactions of multiple heat sources in parallel, accurately identify potential overheating areas, and allocate cooling resources in advance, avoiding local overheating problems caused by reaction lags in traditional thermal management systems, and significantly improving the response speed and efficiency of the heat dissipation system. The quantum gate thermal energy conduction control algorithm further optimizes the heat conduction path. By dynamically adjusting the direction and rate of heat conduction, the present invention can effectively reduce the accumulation of heat in high power consumption areas and quickly guide excess heat in overheating areas to low temperature areas. This innovative dynamic heat conduction control method solves the problems of fixed heat conduction paths and uneven utilization of cooling resources in the prior art, greatly improving the heat dissipation efficiency and the flexibility of the thermal management system.

[0072] The present invention also realizes precise heat dissipation management of different cooling areas by real-time monitoring of temperature changes and power consumption fluctuations in various areas of the circuit board, combining with quantum optimization algorithms, and dynamically adjusting the operating parameters of the cooling equipment. The cooling requirements of high-heat areas and low-heat areas can be accurately matched, optimizing the allocation of cooling resources, and avoiding the problem of over-allocation or insufficient cooling resources in traditional technologies, thereby reducing the waste of cooling resources and improving the energy utilization rate of the system. The introduction of a heat recovery device is also one of the important innovations of the present invention, which converts excess heat into electrical energy for storage and releases it when power consumption is low to balance the overall heat flow of the circuit board. The present invention solves the problem of heat waste in the prior art, further improves energy utilization efficiency, and effectively extends the service life of circuit boards and related electronic equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0074] Figure 1 A flow chart of a circuit board thermal management optimization method combined with quantum computing proposed by the present invention;

[0075] Figure 2 Schematic diagram of the quantum gate heat conduction control algorithm in the present invention dynamically adjusting the heat conduction path. DETAILED DESCRIPTION

[0076] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0077] refer to Figure 1-2 , a circuit board thermal management optimization method combined with quantum computing, comprising the following steps:

[0078] S1. Use the multi-body entangled state quantum optimization algorithm to globally predict the heat flow distribution in different areas of the circuit board, capture the nonlinear heat conduction relationship between each heat source, and build a heat flow mapping model. The algorithm can process the interaction of multiple heat sources in parallel, identify potential overheating areas, and pre-allocate cooling resources;

[0079] S2. Based on the constructed heat flow mapping model, when an abnormal temperature occurs in a certain area of ​​the circuit board, the quantum gate heat conduction control algorithm is used to optimize the local heat conduction path, intelligently adjust the heat conduction direction and rate, and guide the excess heat of the identified overheated area to the low-temperature area to avoid heat accumulation and optimize the heat dissipation efficiency;

[0080] S3. According to the heat flow prediction results and heat energy conduction adjustment, dynamically adjust the cooling equipment parameters of different cooling areas, control the cooling requirements of high-heat areas and low-heat areas, realize accurate allocation of cooling resources, optimize the workload of each cooling module, and avoid resource waste;

[0081] S4. By real-time monitoring of temperature changes and power consumption fluctuations in each area, combined with quantum optimization results, adaptively balance the thermal energy, store excess heat in the heat recovery device, and release it when power consumption is low, balance the overall heat flow of the circuit board, and extend the service life of the circuit board;

[0082] S5. Based on real-time temperature feedback and the dynamic release effect of stored thermal energy, the quantum optimized thermal management strategy is updated regularly, and the cooling path is recalculated through the multi-body entangled state quantum optimization algorithm to maintain the dynamic balance of the heat flow mapping model when the circuit board load changes.

[0083] In this implementation manner, the S1 specifically includes:

[0084] S11. Define n heat sources on the circuit board, denoted as H1, H2, …, H n , the heat flux of each heat source varies according to its power P i (t) and position r i =(x i ,y i ,z i ) determines the heat transfer coefficient matrix T ij (t), where the matrix element T ij (t) represents the heat source H i With heat source H j The time-dependent heat transfer coefficient between:

[0085]

[0086] Among them, γ ij represents the attenuation coefficient of heat conduction, α ij represents the initial heat conduction intensity, ω ij is the frequency of thermal fluctuations between the two heat sources;

[0087] S12. Based on the heat source parameters, a multi-body entangled state system of the circuit board is established. The quantum state |ψ> is defined as the entangled state between all heat sources to describe the global correlation of heat flow. The evolution process of the quantum state is used to capture the nonlinear heat conduction relationship between the heat sources and calculate the time evolution of each heat source.

[0088] S13, by solving the Hamiltonian of the system H = ∑ i,j T ij H i H j, simulate the heat flow interaction between various heat sources, identify areas with drastic temperature fluctuations, and mark them as potential overheating areas;

[0089] S14. Construct a heat flow mapping model to map the power, position and heat conduction coefficient of each heat source to the global heat flow field of the model. The model solves the heat conduction equation Where T is the temperature, q is the thermal power density of the heat source, and k is the thermal conductivity, which generates the global heat flux distribution prediction result;

[0090] S15. Use quantum optimization algorithm to update the parameters of heat flow mapping model in real time, including power fluctuation of each heat source, thermal conductivity coefficient T ij Dynamic changes ensure optimal heat flow distribution under different load conditions;

[0091] S16. According to the constructed heat flow mapping model, cooling resources are allocated in advance. The cooling resource vector is represented as C = (C1, C2,, C n ), where each element C i Indicates the allocation to the heat source H i The cooling amount is adjusted dynamically through quantum optimization algorithm to avoid heat concentration;

[0092] S17. During the operation of the circuit board, the heat conduction changes of each heat source are captured and analyzed in real time, the heat flow mapping model is continuously updated, and the cooling resources are dynamically adjusted according to the latest heat flow distribution prediction.

[0093] In this implementation, S2 specifically includes:

[0094] S21. Based on the constructed heat flow mapping model, when an abnormal temperature occurs in a certain area of ​​the circuit board, the temperature gradient of the overheated area is extracted, and the nonlinear heat conduction relationship between heat sources is adjusted using the quantum gate heat energy conduction control algorithm to capture and process areas with drastic temperature fluctuations;

[0095] S22. Based on the extracted temperature gradient and the prediction results of the heat flow mapping model, the quantum state |ψ hot (t)>Describe the heat conduction behavior in the overheating area and dynamically adjust the heat conduction path;

[0096] S23. Quantum gate operation is used to dynamically adjust the heat conduction path. The adjusted conduction coefficient matrix is ​​expressed as:

[0097]

[0098] Among them, T ij (t) is the initial conduction coefficient, ΔT hot is the temperature difference in the overheated area, ω ij is the thermal oscillation frequency between the two heat sources;

[0099] S24, according to the adjustment results, the excess heat Q in the overheating area hot Lead to low temperature area R cool , optimize the heat transfer rate from overheated areas to low-temperature areas by dynamically adjusting cooling resources;

[0100] S25. Based on the adjustment of the heat conduction path, through the real-time feedback mechanism, the quantum gate algorithm is used to continuously optimize the conduction path, monitor the heat flow changes in real time, make the heat conduction direction and rate dynamically adapt to the current heat flow state, and prevent heat from accumulating in the overheating area.

[0101] In this implementation, S3 specifically includes:

[0102] S31. Based on the constructed heat flow mapping model, identify the high heat area R on the circuit board hot and low heat area R cool , respectively calculate the cooling requirements of each area. The heat flux mapping model provides the prediction results of the global heat flux distribution and cooling requirements on the circuit board. For the high heat area, the cooling requirement is:

[0103]

[0104] Among them, q hot is the thermal power density of the high heat area, is the temperature gradient in the high heat area, f adjust (t is the dynamic adjustment coefficient;

[0105] For low heat areas, the cooling requirement is:

[0106]

[0107] Among them, q cool and are the thermal power density and temperature gradient in the low heat area, g adjust (t) is the cooling adjustment factor for the low heat area;

[0108] S32, dynamically adjust the parameters of the cooling equipment according to the cooling demand forecast, including the fan speed V f (t) and the flow rate F of the liquid cooling system l (t):

[0109] V f (t) = V f (0)+γ·(C hot (t)-C current (t));

[0110] F l (t) = F l(0)+λ·(C cool (t) ΔT cool );

[0111] Among them, γ and λ are adjustment gains respectively;

[0112] S33, based on the adjustment of cooling equipment parameters, calculating the workload of different cooling modules, optimizing the allocation of cooling resources, and optimizing the load of each cooling module;

[0113] S34, monitor the temperature and cooling equipment parameters of each cooling area in real time, use quantum optimization algorithm to dynamically update the allocation strategy of cooling resources, and continuously optimize the allocation of cooling resources to avoid waste of cooling resources and imbalanced workload;

[0114] S35, based on the cooling data feedback and combined with the quantum optimization results, optimize the working efficiency of the cooling module, so that the high-heat area can obtain sufficient cooling resources, while optimizing the cooling resource allocation in the low-heat area to avoid excessive resource consumption.

[0115] In this implementation, S4 specifically includes:

[0116] S41, extract the temperature feedback parameter ΔT by real-time monitoring of the temperature changes and power consumption fluctuations of each heat source on the circuit board i (t) and power consumption fluctuation parameter P i (t), using quantum optimization algorithm to determine the excess heat of each heat source, and convert the excess heat Q excess Converted into electrical energy;

[0117] S42. The converted excess heat is stored in a heat recovery device, which includes a plurality of heat storage units, using thermal batteries and phase change materials for storage, and the excess heat is stored in the heat storage unit to avoid local overheating:

[0118] Q store (t)=∑ i (Q i (t)-Q threshold );

[0119] Among them, Q store (t) is the stored heat, Q threshold is the activation threshold of the thermal energy storage unit;

[0120] S43. When the power consumption of the circuit board is lower than a preset threshold, the heat recovery device starts to release the stored heat, and gradually releases the heat when the power consumption is low to balance the overall heat flow of the circuit board:

[0121]

[0122] Among them, Trelease is the heat release cycle, f release (t) is a dynamic release function;

[0123] S44. Adjust the heat storage and release process through quantum optimization algorithm, adjust the thermoelectric conversion efficiency and release frequency in combination with the circuit board load conditions, cope with the heat accumulation under high load and the release demand under low load, and achieve stable distribution of heat flow;

[0124] S45. Adjust the working parameters of the heat recovery device based on the feedback mechanism, continuously optimize the operating state of the heat recovery device, balance the temperature of each area, and extend the service life of the circuit board.

[0125] In this implementation manner, S5 specifically includes:

[0126] S51, through the real-time temperature feedback system, monitor the temperature changes of each heat source of the circuit board, based on the temperature feedback parameter ΔT i (t) and power consumption feedback parameter P i (t), and combined with the dynamic release effect of the heat recovery device, calculate the temperature fluctuation of each heat source and generate feedback data;

[0127] S52. Based on the feedback data, the quantum optimization algorithm sets the frequency of updating the thermal management strategy. The frequency of updating the strategy is updated through a timed update mechanism, and the thermal management strategy is updated every Δt time interval. When the temperature fluctuation amplitude of any heat source is ΔT i (t) exceeds the preset fluctuation threshold ΔT threshold Trigger update when

[0128] S53, the feedback data sources include the temperature sensor, the power consumption monitoring device and the release data of the heat recovery device, and generate the comprehensive feedback data F total (t):

[0129] F total (t) = α·f feedback (t)+β·P monitor (t)+γ·Q release (t);

[0130] Among them, P monitor (t) is the power consumption monitoring data, Q release (t) is the heat release data of the heat recovery device, α, β, γ are the feedback weight coefficients;

[0131] S54, based on comprehensive feedback data F total (t), the quantum optimization algorithm recalculates the cooling path through multi-body entangled state quantum computing:

[0132] |ψ(t)>=∑ i ci (t)|H i +∑ i,j c ij (t)|H i H j ;

[0133] Among them, c i (t) represents the heat flux contribution of a single heat source, c ij (t) is the entanglement effect between multiple heat sources;

[0134] S55, combining the quantum optimization calculation results and the load changes of the circuit board, dynamically adjust the cooling path, and optimize the heat transfer coefficient T ij (t) and cooling equipment parameters to achieve a dynamic balance of heat flow:

[0135]

[0136] Among them, ω ij is the thermal oscillation frequency between the two heat sources;

[0137] S56, by regularly evaluating the feedback parameter F total Based on the changes in (t) and the results of quantum optimization calculations, the thermal management strategy of the circuit board is updated to maintain the dynamic balance of heat flow, avoid local overheating and optimize the heat dissipation effect.

[0138] Embodiment 1:

[0139] In order to verify the feasibility of the present invention in implementation, the present invention is applied to the actual application of a large data center. The processor and graphics processor of the server generate a large amount of heat under high load conditions, resulting in frequent overload of the server's cooling system. Although traditional air cooling systems and liquid cooling systems can solve certain heat dissipation problems, as the computing load increases, local overheating of the system becomes more and more serious, resulting in frequent processor frequency reduction and a significant reduction in overall computing efficiency. Since traditional cooling systems cannot achieve dynamic adjustment, cooling resources are often unevenly distributed, high-heat areas cannot be cooled in a timely and effective manner, while low-heat areas are over-cooled, wasting precious cooling resources. The data center decided to adopt the circuit board thermal management optimization method combined with quantum computing proposed in the present invention to improve the server's thermal management system.

[0140] In this actual scenario, the present invention uses a multi-body entangled state quantum optimization algorithm to make a global prediction of the heat flow distribution inside the server, captures the nonlinear heat conduction relationship between the processor and other high-power devices in real time, and establishes a high-precision heat flow mapping model. Under high-load operation, the quantum optimization algorithm can process the interaction of multiple heat sources in parallel, identify potential overheating areas in advance, and reasonably allocate cooling resources, avoiding the problem of response lag in traditional systems. During a high-load period in the data center, the temperature of the processor area in the server cluster rises rapidly. The traditional system needs to wait for the global temperature to reach a preset value before starting to increase the fan speed or liquid cooling flow. The system of the present invention predicts in real time through the quantum optimization algorithm that the temperature of the area is about to exceed the safe range, adjusts the fan speed and liquid cooling flow in advance, and ensures that the temperature of the area is controlled within a reasonable range. Specific data show that after adopting the quantum optimization system of the present invention, the average temperature of the processor area drops from the original 85°C to 75°C, and the peak temperature drops from 95°C to 82°C. At the same time, due to the reasonable allocation of cooling resources, the overall cooling energy consumption of the system is reduced by about 20%.

[0141] The graphics processor cluster in the data center works under high load, and the heat generated is highly concentrated in a specific area. The traditional heat dissipation system cannot export the heat in time, causing the temperature in the area to continue to rise, even exceeding 90°C. The present invention uses a quantum gate heat conduction control algorithm to optimize the local heat conduction path, adjust the heat conduction direction and rate in real time, and guide the excess heat from the overheated area to the area with lower temperature, thereby optimizing the heat conduction path. In this process, the quantum gate operation dynamically adjusts the heat conductivity coefficient, so that the heat in the high-heat area can be quickly guided to the low-heat area, reducing heat accumulation. After applying this technology, the temperature of the graphics processor area is stabilized below 80°C, effectively avoiding the performance degradation problem caused by local overheating.

[0142] The heat recovery device of the present invention also plays an important role in these scenarios. The real-time monitoring system will capture the temperature changes and power consumption fluctuations of each heat source, calculate the excess heat of each heat source, and convert this heat into electrical energy storage through a thermoelectric conversion device. Taking a certain data center as an example, during high-load periods, the system can recover up to 300 watts of thermal energy in each high-load cycle and convert it into approximately 60 watts of electrical energy, which is stored in the heat recovery device. During the period when the server load is reduced, the system gradually releases the stored heat to maintain the balance of the internal temperature of the server, and the overall energy consumption of the data center is further reduced by approximately 15%.

[0143] The present invention dynamically adjusts the parameters of cooling equipment through a quantum optimization algorithm to ensure that the cooling needs of high-heat areas and low-heat areas are accurately met. In a nighttime load test, the fan speed and liquid cooling system flow of the data center are intelligently adjusted according to the real-time load and temperature fluctuations of the server cluster. In the high-heat area, the fan speed is increased from the default 1000RPM to 1500RPM, while in the low-heat area, the fan speed is maintained at a low level of 900RPM. The flow of the liquid cooling system is also adjusted accordingly based on the real-time heat flow prediction model. The test results show that during the entire nighttime high-load period, the system's cooling energy consumption is reduced by about 25%, while avoiding the problem of over-cooling of low-heat areas in traditional systems.

[0144] After a month-long comparison of operating data, the technical team of the data center found that after adopting the quantum optimized thermal management system of the present invention, the overall heat dissipation efficiency of the server was improved by about 30%, local overheating was significantly reduced, the average operating temperature of the system was about 10°C lower than that of the traditional system, and the energy consumption of the cooling system was reduced by about 20%. The heat recovery mechanism further reduced the overall energy consumption of the data center, and the energy utilization efficiency was improved by more than 15%. In multiple high-load tests, the system was able to stably maintain the temperature of each area within a safe range, effectively extending the service life of the server and reducing the failure rate of equipment.

[0145] Table 1 Test data of quantum optimization thermal management method

[0146]

[0147] It can be seen from the table data that the quantum optimized thermal management system of the present invention has shown significant optimization effects in multiple test scenarios. In terms of temperature control in the processor area, the average temperature of the traditional system is 85°C, and the peak temperature is as high as 95°C, while the system of the present invention reduces the average temperature to 75°C and the peak temperature to 82°C, reducing the cooling energy consumption by about 20%. During the heat conduction process in the graphics processor area, the temperature of the traditional system is 90°C, and the present invention reduces the temperature of this area to 80°C, avoiding overheating. In terms of heat recovery, the heat recovery device of the present invention can recover 300 watts of heat energy per cycle and convert it into 60 watts of electrical energy, which improves the energy utilization rate by about 15%. The fan speed and liquid cooling system flow are dynamically adjusted through the quantum optimization algorithm, and the overall cooling energy consumption is reduced by 25%, which optimizes the allocation of cooling resources, so that the cooling needs of high-heat areas and low-heat areas are accurately met, further improving the stability and energy utilization efficiency of the system.

[0148] In summary, by applying the circuit board thermal management optimization method combined with quantum computing proposed in the present invention, the data center has successfully solved the problems of response lag, rigid heat conduction path, uneven distribution of cooling resources and energy waste in the traditional heat dissipation system, significantly improved the system's heat dissipation efficiency and energy utilization, and effectively extended the service life of the equipment.

[0149] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A circuit board thermal management optimization method combined with quantum computing, characterized in that: The steps include: S1. Use the multi-body entangled state quantum optimization algorithm to globally predict the heat flow distribution in different areas of the circuit board, capture the nonlinear heat conduction relationship between each heat source, and build a heat flow mapping model. The algorithm can process the interaction of multiple heat sources in parallel, identify potential overheating areas, and pre-allocate cooling resources; S2. Based on the constructed heat flow mapping model, when an abnormal temperature occurs in a certain area of ​​the circuit board, the quantum gate heat conduction control algorithm is used to optimize the local heat conduction path, intelligently adjust the heat conduction direction and rate, and guide the excess heat of the identified overheated area to the low-temperature area to avoid heat accumulation and optimize the heat dissipation efficiency; S3. According to the heat flow prediction results and heat energy conduction adjustment, dynamically adjust the cooling equipment parameters of different cooling areas, control the cooling requirements of high-heat areas and low-heat areas, realize accurate allocation of cooling resources, optimize the workload of each cooling module, and avoid resource waste; S4. By real-time monitoring of temperature changes and power consumption fluctuations in each area, combined with quantum optimization results, adaptively balance the thermal energy, store excess heat in the heat recovery device, and release it when power consumption is low, balance the overall heat flow of the circuit board, and extend the service life of the circuit board; S5. Based on real-time temperature feedback and the dynamic release effect of stored thermal energy, the quantum optimized thermal management strategy is updated regularly, and the cooling path is recalculated through the multi-body entangled state quantum optimization algorithm to maintain the dynamic balance of the heat flow mapping model when the circuit board load changes.

2. The method for optimizing thermal management of a circuit board combined with quantum computing according to claim 1, characterized in that: The S1 specifically includes: S11. Define n heat sources on the circuit board, denoted as H1, H2, …, H n , the heat flux of each heat source varies according to its power P i (t) and position r i =(x i ,y i ,z i ) determines the heat transfer coefficient matrix T ij (t), where the matrix element T ij (t) represents the heat source H i With heat source H j The time-dependent heat transfer coefficient between: Among them, γ ij represents the attenuation coefficient of heat conduction, α ij represents the initial heat conduction intensity, ω ij is the frequency of thermal fluctuations between the two heat sources; S12. Based on the heat source parameters, a multi-body entangled state system of the circuit board is established. The quantum state |ψ> is defined as the entangled state between all heat sources to describe the global correlation of heat flow. The evolution process of the quantum state is used to capture the nonlinear heat conduction relationship between the heat sources and calculate the time evolution of each heat source. S13, by solving the Hamiltonian of the system H = ∑ i,j T ij H i H j , simulate the heat flow interaction between various heat sources, identify areas with drastic temperature fluctuations, and mark them as potential overheating areas; S14. Construct a heat flow mapping model to map the power, position and heat conduction coefficient of each heat source to the global heat flow field of the model. The model solves the heat conduction equation Where T is the temperature, q is the thermal power density of the heat source, and k is the thermal conductivity, which generates the global heat flux distribution prediction result; S15. Use quantum optimization algorithm to update the parameters of heat flow mapping model in real time, including power fluctuation of each heat source, thermal conductivity coefficient T ij Dynamic changes ensure optimal heat flow distribution under different load conditions; S16. According to the constructed heat flow mapping model, cooling resources are allocated in advance. The cooling resource vector is represented as C = (C1, C2,, C n ), where each element C i Indicates the allocation to the heat source H i The cooling amount is adjusted dynamically through quantum optimization algorithm to avoid heat concentration; S17. During the operation of the circuit board, the heat conduction changes of each heat source are captured and analyzed in real time, the heat flow mapping model is continuously updated, and the cooling resources are dynamically adjusted according to the latest heat flow distribution prediction.

3. The method for optimizing thermal management of a circuit board combined with quantum computing according to claim 1, characterized in that: The S2 specifically includes: S21. Based on the constructed heat flow mapping model, when an abnormal temperature occurs in a certain area of ​​the circuit board, the temperature gradient of the overheated area is extracted, and the nonlinear heat conduction relationship between heat sources is adjusted using the quantum gate heat energy conduction control algorithm to capture and process areas with drastic temperature fluctuations; S22. Based on the extracted temperature gradient and the prediction results of the heat flow mapping model, the quantum state |ψ hot (t)>Describe the heat conduction behavior in the overheating area and dynamically adjust the heat conduction path; S23. Quantum gate operation is used to dynamically adjust the heat conduction path. The adjusted conduction coefficient matrix is ​​expressed as: Among them, T ij (t) is the initial conduction coefficient, ΔT hot is the temperature difference in the overheated area, ω ij is the thermal oscillation frequency between the two heat sources; S24, according to the adjustment results, the excess heat Q in the overheating area hot Lead to low temperature area R cool , optimize the heat transfer rate from overheated areas to low-temperature areas by dynamically adjusting cooling resources; S25. Based on the adjustment of the heat conduction path, through the real-time feedback mechanism, the quantum gate algorithm is used to continuously optimize the conduction path, monitor the heat flow changes in real time, make the heat conduction direction and rate dynamically adapt to the current heat flow state, and prevent heat from accumulating in the overheating area.

4. The method for optimizing thermal management of a circuit board combined with quantum computing according to claim 1, characterized in that: The S3 specifically includes: S31. Based on the constructed heat flow mapping model, identify the high heat area R on the circuit board hot and low heat area R cool , respectively calculate the cooling requirements of each area. The heat flux mapping model provides the prediction results of the global heat flux distribution and cooling requirements on the circuit board. For the high heat area, the cooling requirement is: Among them, q hot is the thermal power density of the high heat area, is the temperature gradient in the high heat area, f adjust (t is the dynamic adjustment coefficient; For low heat areas, the cooling requirement is: Among them, q cool and are the thermal power density and temperature gradient in the low heat area, g adjust (t) is the cooling adjustment factor for the low heat area; S32, dynamically adjust the parameters of the cooling equipment according to the cooling demand forecast, including the fan speed V f (t) and the flow rate F of the liquid cooling system l (t): V f (t)=V f (0)+γ·(C hot (t)-C current (t)); F l (t)=F l (0)+λ·(C cool (t)·ΔT cool ); Among them, γ and λ are adjustment gains respectively; S33, based on the adjustment of cooling equipment parameters, calculating the workload of different cooling modules, optimizing the allocation of cooling resources, and optimizing the load of each cooling module; S34, real-time monitoring of the temperature and cooling equipment parameters of each cooling area, using quantum optimization algorithm to dynamically update the allocation strategy of cooling resources, and continuously optimize the allocation of cooling resources to avoid waste of cooling resources and imbalanced workload; S35, based on the cooling data feedback and combined with the quantum optimization results, optimize the working efficiency of the cooling module, so that the high-heat area can obtain sufficient cooling resources, while optimizing the cooling resource allocation in the low-heat area to avoid excessive resource consumption.

5. The method for optimizing thermal management of a circuit board combined with quantum computing according to claim 1, characterized in that: The S4 specifically includes: S41, extract the temperature feedback parameter ΔT by real-time monitoring of the temperature changes and power consumption fluctuations of each heat source on the circuit board i (t) and power consumption fluctuation parameter P i (t), using quantum optimization algorithm to determine the excess heat of each heat source, and convert the excess heat Q excess Converted into electrical energy; S42. The converted excess heat is stored in a heat recovery device, which includes a plurality of heat storage units, using thermal batteries and phase change materials for storage, and the excess heat is stored in the heat storage unit to avoid local overheating: Q store (t)=∑ i (Q i (t)-Q threshold ); Among them, Q store (t) is the stored heat, Q threshold is the activation threshold of the thermal energy storage unit; S43. When the power consumption of the circuit board is lower than a preset threshold, the heat recovery device starts to release the stored heat, and gradually releases the heat when the power consumption is low to balance the overall heat flow of the circuit board: Among them, T release is the heat release cycle, f release (t) is a dynamic release function; S44. Adjust the heat storage and release process through quantum optimization algorithm, adjust the thermoelectric conversion efficiency and release frequency in combination with the circuit board load conditions, cope with the heat accumulation under high load and the release demand under low load, and achieve stable distribution of heat flow; S45. Adjust the working parameters of the heat recovery device based on the feedback mechanism, continuously optimize the operating state of the heat recovery device, balance the temperature of each area, and extend the service life of the circuit board.

6. The method for optimizing thermal management of a circuit board combined with quantum computing according to claim 1, characterized in that: The S5 specifically includes: S51, through the real-time temperature feedback system, monitor the temperature changes of each heat source of the circuit board, based on the temperature feedback parameter ΔT i (t) and power consumption feedback parameter P i (t), and combined with the dynamic release effect of the heat recovery device, calculate the temperature fluctuation of each heat source and generate feedback data; S52. Based on the feedback data, the quantum optimization algorithm sets the frequency of updating the thermal management strategy. The frequency of updating the strategy is updated through a timed update mechanism, and the thermal management strategy is updated every Δt time interval. When the temperature fluctuation amplitude of any heat source is ΔT i (t) exceeds the preset fluctuation threshold ΔT threshold Trigger update when S53, the feedback data sources include the temperature sensor, the power consumption monitoring device and the release data of the heat recovery device, and generate the comprehensive feedback data F total (t): F total (t)=α·f feedback (t)+β·P monitor (t)+γ·Q release (t); Among them, P monitor (t) is the power consumption monitoring data, Q release (t) is the heat release data of the heat recovery device, α, β, γ are the feedback weight coefficients; S54, based on comprehensive feedback data F total (t), the quantum optimization algorithm recalculates the cooling path through multi-body entangled state quantum computing: |ψ(t)>=∑ i c i (t)|H i +∑ i,j c ij (t)|H i H j ; Among them, c i (t) represents the heat flux contribution of a single heat source, c ij (t) is the entanglement effect between multiple heat sources; S55, combining the quantum optimization calculation results and the load changes of the circuit board, dynamically adjust the cooling path, and optimize the heat transfer coefficient T ij (t) and cooling equipment parameters to achieve a dynamic balance of heat flow: Among them, ω ij is the thermal oscillation frequency between the two heat sources; S56, by regularly evaluating the feedback parameter F total Based on the changes in (t) and the results of quantum optimization calculations, the thermal management strategy of the circuit board is updated to maintain the dynamic balance of heat flow, avoid local overheating and optimize the heat dissipation effect.

Images