Multi-dimensional data source predictive intelligent temperature control method, device and equipment and storage medium

Abstract

The embodiment of the invention provides a multi-dimensional data source predictive intelligent temperature control method, device and equipment and a storage medium, the method is applied to electronic equipment, the electronic equipment is in communication connection with terminal equipment, the terminal equipment comprises a plurality of functional units and a plurality of fans, each functional unit corresponds to at least one fan, and the fans correspond to the fans. The fan is used for adjusting the temperature of the functional units. The method comprises the steps of obtaining target temperatures of the multiple functional units; according to the multiple target temperatures, an initial control instruction of a fan corresponding to at least one functional unit is determined from the multiple functional units; according to the temperature change rates of the multiple functional units, the initial control instruction of the fan corresponding to at least one functional unit is adjusted, and a target control instruction of the fan corresponding to at least one functional unit is obtained; and controlling the fan corresponding to the at least one functional unit according to the target control instruction of the fan corresponding to the at least one functional unit. According to the embodiment of the invention, prospective and refined intelligent adjustment of the heat dissipation system can be realized.

Description

Technical Field

[0001] This application relates to the field of intelligent control technology, and in particular to a multi-dimensional data source predictive intelligent temperature control method, device, equipment and storage medium. Background Technology

[0002] With the rapid development of high-performance computing and network communication equipment such as enterprise-level servers, core switches, and edge computing units, the integration and power density of their internal components continue to rise. These systems often integrate multiple core heat-generating units, such as high-performance CPUs, ASIC switching chips, FPGAs, and high-speed optical modules. These components generate intense and uneven heat during operation, posing a severe challenge to the stable operation and long-term reliability of the equipment. Currently, such equipment largely relies on traditional air-cooling solutions, and their temperature control effectiveness directly affects system performance and lifespan.

[0003] Traditional temperature-controlled cooling solutions, especially in air-cooled systems, primarily rely on adjusting fan speeds. However, their control strategies are often overly simplistic. Most existing systems monitor a single heat source (typically the CPU core temperature or a single onboard ambient temperature sensor) and establish a simple linear or stepwise function to control all fan speeds. This method only intervenes after the temperature actually reaches a dangerous threshold. For devices with high power transients (such as CPUs under sudden computational loads), this lag can cause temperature peaks to far exceed safety limits, triggering overheat protection mechanisms and leading to system throttling or crashes. Summary of the Invention

[0004] The main objective of this application is to propose a multi-dimensional data source predictive intelligent temperature control method, device, electronic device, and storage medium, which can solve the problem of lag in existing temperature control and heat dissipation control, and realize forward-looking and precise intelligent adjustment of the heat dissipation system.

[0005] To achieve the above objectives, a first aspect of this application proposes a multi-dimensional data source predictive intelligent temperature control method applied to an electronic device. The electronic device is communicatively connected to a terminal device, which includes multiple functional units and multiple fans. Each functional unit corresponds to at least one fan, and the fans are used to regulate the temperature of the functional unit. The method includes: Obtain the target temperature of multiple functional units; Based on the multiple target temperatures, determine the initial control command for the fan corresponding to at least one of the multiple functional units; Based on the temperature change rate of the multiple functional units, the initial control command of the fan corresponding to the at least one functional unit is adjusted to obtain the target control command of the fan corresponding to the at least one functional unit. The fan corresponding to the at least one functional unit is controlled according to the target control command of the fan corresponding to the at least one functional unit.

[0006] In some embodiments, the plurality of said functional units include a central processing unit, a switching chip, an environmental sensor, and a plurality of optical modules; The step of obtaining the target temperature of the plurality of functional units includes: The initial temperature of each functional unit is obtained by polling the central processing unit, the switching chip, the environmental sensor, and the multiple optical modules one by one through the bus interface of the terminal device. Outlier removal and smoothing are performed on the multiple initial temperatures to obtain the target temperatures of the multiple functional units.

[0007] In some embodiments, after performing outlier removal and smoothing on the plurality of initial temperatures to obtain target temperatures for the plurality of functional units, and before determining the initial control command for the fan corresponding to at least one functional unit from the plurality of functional units based on the plurality of target temperatures, the method further includes: Based on the target temperatures of the multiple optical modules, the average temperature and the maximum temperature of the multiple optical modules are obtained, wherein the maximum temperature is the target temperature with the largest value among the target temperatures of the multiple optical modules; The step of determining the initial control command for the fan corresponding to at least one functional unit from among the multiple functional units based on the multiple target temperatures includes: The area containing multiple fans is divided into multiple thermal contribution zones, each of which includes at least one fan; The initial control command for the fan corresponding to at least one functional unit in each thermal contribution zone is calculated based on the highest temperature of the optical module, the average temperature of the optical module, and the target temperature of each functional unit in each thermal contribution zone.

[0008] In some embodiments, dividing the area containing the plurality of fans into a plurality of thermal contribution zones includes: Based on the physical installation locations of the multiple fans and the preset cooling target, the area containing the multiple fans is divided into multiple thermal contribution zones. The multiple thermal contribution zones include a central processing unit thermal contribution zone and a system thermal contribution zone. The central processing unit thermal contribution zone includes at least one first fan for cooling the central processing unit, and the system thermal contribution zone includes at least one second fan for cooling system board-level components.

[0009] In some embodiments, the calculation of the initial control command for the fan corresponding to at least one functional unit in each thermal contribution zone based on the highest temperature of the optical module, the average temperature of the optical module, and the target temperature of each functional unit in each thermal contribution zone includes: For the central processing unit's thermal contribution area, the initial control command for the fan corresponding to at least one functional unit in the central processing unit's thermal contribution area is obtained by weighting the initial temperature of the central processing unit, the average temperature of the optical module, and the ambient temperature with a first set of weight parameters. The first set of weight parameters includes a first weight parameter corresponding to the initial temperature of the central processing unit, a weight parameter corresponding to the average temperature of the optical module, and a first weight parameter corresponding to the ambient temperature. For the system thermal contribution area, the initial control command of the fan corresponding to at least one functional unit in the system thermal contribution area is obtained by weighting the initial temperature of the central processing unit, the maximum temperature of the optical module, and the ambient temperature with the second set of weight parameters. The second set of weight parameters includes the second weight parameter corresponding to the initial temperature of the central processing unit, the weight parameter corresponding to the maximum temperature of the optical module, and the second weight parameter corresponding to the ambient temperature.

[0010] In some embodiments, adjusting the initial control command of the fan corresponding to at least one functional unit based on the temperature change rate of the plurality of functional units to obtain the target control command of the fan corresponding to at least one functional unit includes: Obtain the temperature change rate of multiple functional units; For each functional unit, if the temperature change rate of the functional unit is greater than a preset threshold, a nonlinear control adjustment amount is generated based on the temperature change rate of the functional unit, and the initial control command of the fan corresponding to the functional unit is adjusted based on the nonlinear control adjustment amount to obtain the target control command of the fan corresponding to the functional unit; and, If the temperature change rate of the functional unit is less than or equal to the preset threshold, then the initial control command of the fan corresponding to the functional unit is taken as the target control command of the fan corresponding to the functional unit.

[0011] In some embodiments, obtaining the temperature change rate of the plurality of functional units includes: Obtain the historical temperature of multiple functional units; The historical temperature of each of the multiple functional units is calculated using a first-order difference to obtain the temperature change rate of each of the multiple functional units.

[0012] To achieve the above objectives, a second aspect of this application proposes a multi-dimensional data source predictive intelligent temperature control device applied to an electronic device. The electronic device is communicatively connected to a terminal device, which includes multiple functional units and multiple fans. Each functional unit corresponds to at least one fan, and the fans are used to regulate the temperature of the functional unit. The device includes: The acquisition module is used to acquire the target temperature of the multiple functional units; The determining module is used to determine the initial control command of the fan corresponding to at least one functional unit from among the multiple functional units based on the multiple target temperatures; An adjustment module is used to adjust the initial control command of the fan corresponding to at least one functional unit according to the temperature change rate of the multiple functional units, so as to obtain the target control command of the fan corresponding to at least one functional unit. The control module is used to control the fan corresponding to the at least one functional unit according to the target control command of the fan corresponding to the at least one functional unit.

[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] This application proposes a multi-dimensional data source predictive intelligent temperature control method, device, electronic device, and storage medium. It acquires the target temperature of each functional unit, determines the initial control command for the corresponding fan based on this temperature data, adjusts the initial control command based on the temperature change rate of each functional unit to obtain the target control command, and finally regulates the corresponding fan according to the target control command. This technical solution achieves precise and proactive temperature control of the multi-functional units of the terminal device. This application breaks through the limitations of traditional reactive control relying solely on a single heat source. By introducing multi-dimensional temperature sensing and predictive intervention based on temperature change trends, it can not only effectively solve the problem of localized overheating of other key components besides the CPU, but also proactively address sudden heat loads, eliminate control lag, and avoid system frequency reduction or shutdown caused by temperature overshoot. This significantly improves the heat dissipation efficiency, operational stability, and overall performance of electronic devices under complex loads. Attached Figure Description

[0016] Figure 1This is a flowchart illustrating the multi-dimensional data source predictive intelligent temperature control method provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the construction of a multidimensional thermodynamic state vector provided in an embodiment of this application; Figure 3 This is a flowchart illustrating the working principle of the dynamic weighted model provided in the embodiments of this application; Figure 4 This is a flowchart of the predictive rate control logic provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of the multi-dimensional data source predictive intelligent temperature control device provided in the embodiments of this application; Figure 6 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0018] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] In modern high-performance computing and network communication equipment, such as enterprise servers, core switches, and edge computing units, the integration and power density of internal components are increasing at an unprecedented rate. These complex electronic systems typically integrate multiple core heat-generating units, including but not limited to high-performance central processing units (CPUs), application-specific integrated circuit (ASIC) switching chips, field-programmable gate arrays (FPGAs), and numerous high-speed pluggable optical modules (such as SFP+ and QSFP28). These components generate intense and uneven heat under different workloads, posing a severe challenge to the stable operation, performance, and long-term reliability of the system, necessitating efficient heat dissipation solutions to address the complexity of heat distribution.

[0021] Existing temperature control and heat dissipation solutions, especially in air-cooled systems, mainly rely on adjusting fan speeds, but their control strategies are often overly simplistic. Most existing systems control the speeds of all fans by monitoring the readings of a single heat source (usually the CPU core temperature or a single onboard ambient temperature sensor) and establishing a simple linear or stepwise function based on this. In today's increasingly complex hardware architectures, the technical limitations of this "single-point input, unified output" model are becoming increasingly apparent, manifesting in three key technical problems: 1) Hotspot perception blind spots and local overheating risks: This strategy completely ignores other key heat sources in the system besides the CPU. For example, in a network switching platform, when the CPU is under low load and the optical module is operating at high capacity, the optical module temperature rises sharply but cannot be responded to, leading to performance degradation or permanent damage; 2) Lag and thermal overshoot: This strategy is essentially reactive control, which must wait until the temperature actually reaches the dangerous threshold before intervening. For devices with high power transients, the temperature peak far exceeds the safety line, triggering the overheat protection mechanism, causing the system to reduce frequency or crash; 3) Poor scenario adaptability: A fixed temperature control curve is difficult to adapt to the diverse application scenarios of the device. For example, the thermal distribution characteristics of compute-intensive and network I / O-intensive applications are different, and traditional methods cannot dynamically adapt, lacking intelligence and flexibility.

[0022] Based on this, embodiments of this application provide a multi-dimensional data source predictive intelligent temperature control method, device, electronic device, and storage medium. The aim is to provide a predictive intelligent temperature control method based on multi-dimensional data sources, realizing a fundamental shift from a "single-point, reactive" control model to a "multi-dimensional, predictive" control model. It does not rely on any single temperature reading, but instead constructs a "thermal state vector" that can comprehensively reflect the instantaneous thermal state of the equipment. Combined with the analysis of temperature change trends, it achieves forward-looking and refined intelligent adjustment of the heat dissipation system.

[0023] The multidimensional data source predictive intelligent temperature control method, device, electronic device and storage medium provided in the embodiments of this application are specifically described through the following embodiments. First, the multidimensional data source predictive intelligent temperature control method in the embodiments of this application is described.

[0024] The multi-dimensional data source predictive intelligent temperature control method provided in this application relates to the field of intelligent control technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the multi-dimensional data source predictive intelligent temperature control method, but is not limited to the above forms.

[0025] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0026] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0027] Figure 1This is an optional flowchart of a multi-dimensional data source predictive intelligent temperature control method provided in this application embodiment. This application embodiment can be applied to an electronic device, which is communicatively connected to a terminal device. The terminal device includes multiple functional units and multiple fans, with each functional unit corresponding to at least one fan. The fans are used to adjust the temperature of the functional unit. Figure 1 The method may include, but is not limited to, steps S100 to S400.

[0028] Step S100: Obtain the target temperature of the plurality of functional units.

[0029] In this embodiment, the electronic device can be an onboard embedded controller deployed inside a terminal device (such as an industrial control computer, server, or switch), or it can be another device deployed outside the terminal device. The electronic device communicates with the terminal device and can read sensor data from various components inside the terminal device via buses such as I2C and SMBus, and control the fans on the terminal device. The terminal device includes multiple functional units and multiple fans. Each functional unit has at least one fan, which is used to regulate the temperature of the corresponding functional unit. Functional units include, but are not limited to, core heat-generating components such as a central processing unit (CPU), application-specific integrated circuit (ASIC), optical modules (such as SFP+, QSFP28), and switching chips.

[0030] Specifically, the electronic device synchronously collects raw temperature data of each functional unit at a preset cycle (e.g., 1 second / time) using sensors built into the terminal device. The collected raw temperature data undergoes preprocessing, including range verification (e.g., filtering valid data within the range of -40°C to 125°C), outlier removal (removing invalid data such as NaN values ​​and fixed-value jitter), amplitude limiting, and noise reduction and smoothing, to obtain the target temperature for each functional unit. All collected temperature data is organized into a structured "thermal state vector." This vector not only contains the raw temperature values ​​but can also include derived values, such as the average and maximum temperatures of all optical modules, providing standardized data input for subsequent complex decisions.

[0031] In this embodiment, to ensure the comprehensiveness and reliability of temperature data, a circular buffer queue of length W can be maintained for each functional unit to store the temperature data after the most recent W preprocessing, for subsequent data verification and trend analysis. If the percentage of valid sensor readings of a certain functional unit is lower than a preset threshold (e.g., p_min=80%), a rollback strategy of "holding the previous valid value + slow decay" is adopted, and a safety warning mechanism is triggered at the same time.

[0032] Step S200: Based on the multiple target temperatures, determine the initial control command for the fan corresponding to at least one functional unit from the multiple functional units.

[0033] In this embodiment, the electronic device calculates the initial control command for the fan corresponding to each functional unit based on the target temperature of each functional unit and a preset dynamic weighted model. The initial control command is the fan's pulse width modulation (PWM) duty cycle signal, used to control the fan's base speed.

[0034] Specifically, for each functional unit, a weighted coefficient is assigned to its target temperature based on its heat dissipation characteristics, importance in the terminal device, and physical relationship with the corresponding fan. The initial PWM duty cycle of the fan corresponding to that functional unit is then calculated using a weighted average. For example, the control command for the system fan can be calculated using a weighted average of the CPU temperature, the maximum temperature of the optical module, and the ambient temperature; the control command for the CPU fan can be calculated using a weighted average of the CPU temperature, the average temperature of the optical module, and the ambient temperature. The weighted coefficients can be pre-configured or dynamically adjusted based on the device's heat dissipation structure and application scenario.

[0035] Step S300: Based on the temperature change rate of the multiple functional units, adjust the initial control command of the fan corresponding to the at least one functional unit to obtain the target control command of the fan corresponding to the at least one functional unit.

[0036] In this embodiment, the electronic device analyzes the temperature change trends of each functional unit and predictively adjusts the initial control commands to actively suppress temperature peaks. The system maintains a time-series buffer for key functional units, storing the most recent temperature data. By calculating the rate of temperature change (dT / dt), it determines whether a rapid temperature rise trend exists. If the rate of temperature change of a functional unit exceeds a preset rapid temperature rise threshold, the initial control command for its corresponding fan is immediately nonlinearly enhanced to provide stronger heat dissipation capacity in advance and suppress temperature peaks.

[0037] Step S400: Control the fan corresponding to the at least one functional unit according to the target control command of the fan corresponding to the at least one functional unit.

[0038] In this embodiment, the electronic device converts the adjusted target control command into a PWM signal or other control signal to drive the corresponding fan, thereby achieving precise and dynamic adjustment of the internal temperature of the terminal device. During the control process, the electronic device receives the fan speed feedback signal and the temperature feedback data of the functional unit in real time, and performs closed-loop calibration of the target control command: if the feedback temperature still shows a rapid upward trend, the PWM duty cycle increment is further increased; if the feedback temperature tends to stabilize or decrease, the PWM duty cycle is gradually reduced to a reasonable range to ensure a balance between heat dissipation efficiency and fan energy consumption.

[0039] The flow chart for constructing the multi-dimensional thermal state vector in this embodiment is as shown in Figure 2 the following figure. Through the periodic data acquisition task implemented in the baseboard management controller (RISC-V MCU), a comprehensive perception of the thermal distribution of the entire system is established. The steps include: (1) Sampling scheduling: The timer triggers an acquisition every Δt; if the previous acquisition is not completed, this cycle is skipped to prevent bus congestion.

[0040] (2) Channel polling: Access in the order of CPU → SFP / QSFP (via I2CMux) → switching chip → environmental sensor; set the retry count Nretry and timeout ttimeout for each channel.

[0041] (3) Reading and verification: Perform range checks (such as -40°C ≤ T ≤ 125°C), outlier rejection (NaN, fixed value jitter), clamping, and unit normalization on each reading.

[0042] (4) Circular buffer: Maintain a circular queue with a length of W for each temperature measurement point to store the last W readings for differential and smoothing use.

[0043] (5) Denoising and smoothing: Apply median filtering (window k = 3) and first-order exponential smoothing (EMA, coefficient αT) to the single-point sequence to obtain Tfilt.

[0044] (6) Derived quantity calculation: Based on all optical modules, obtain T_sfp_avg, T_sfp_max (optional T_sfp_p95); the ambient temperature compensated temperature difference ΔThot = Thot - T_ambient can be calculated.

[0045] (7) Missing / anomaly fallback: If the proportion of valid readings of a certain type of sensor is < p_min, adopt the fallback of "maintaining the previous valid value + slow decay" and trigger the safety upper limit fan strategy.

[0046] (8) Assemble the vector: Construct V_thermal = {T_cpu, T_sfp_avg, T_sfp_max, T_switch, T_ambient,...} and attach the timestamp t, and deliver it to the subsequent module.

[0047] This embodiment acquires the target temperatures of multiple heterogeneous functional units in the system to build a comprehensive thermal state perception, solving the problem of focusing only on CPU temperature while failing to perceive local high temperatures in other components such as optical modules, thus avoiding potential hardware damage risks. By monitoring the rate of temperature change rather than relying solely on absolute temperature values, the system can intervene in advance. When a rapid temperature rise is detected, even if the current temperature has not exceeded the limit, it can proactively and non-linearly increase the fan speed to suppress the temperature peak in advance, preventing overheat protection from being triggered and ensuring that the system continues to operate stably at its highest performance state. Based on a configurable weighted initial control instruction calculation model, it allows for the preset or dynamic adjustment of the influence weights of different heat sources according to the product's physical design (such as network-intensive or compute-intensive), realizing refined and zoned management of the thermal environment of complex systems.

[0048] In some embodiments, the plurality of functional units may include, but are not limited to, a central processing unit, a switching chip, an environmental sensor, and a plurality of optical modules, and step S100 may include, but is not limited to, steps S110 to S120: Step S110: Poll the central processing unit, the switching chip, the environmental sensor and the multiple optical modules one by one through the bus interface of the terminal device to obtain the initial temperature of each functional unit; Step S120: Outlier removal and smoothing are performed on the multiple initial temperatures to obtain the target temperatures of the multiple functional units.

[0049] After step S120 and before step S200, the following steps may also be included, but are not limited to: Based on the target temperatures of the multiple optical modules, the average temperature and the maximum temperature of the multiple optical modules are obtained, wherein the maximum temperature is the target temperature with the largest value among the target temperatures of the multiple optical modules.

[0050] In this embodiment, the electronic device acquires temperature data from multiple heterogeneous heat sources in the system in real time and synchronously through periodic data acquisition tasks implemented by firmware.

[0051] Specifically, the electronic device polls the aforementioned functional units one by one through the terminal device's bus interface (e.g., I2C bus). To access every critical heat-generating unit in the system, especially the numerous optical modules, the controller can obtain the precise internal temperature values ​​of the optical modules by reading their Digital Diagnostic Monitoring (DDM) interface. During the polling process, access can be performed in a preset order, such as sequentially accessing: CPU, SFP / QSFP optical modules (via I2C Mux), switching chips, environmental sensors, etc. Furthermore, a retry count and timeout are set for each channel to ensure stable system operation when the bus is busy or devices are unresponsive. This polling process obtains the initial temperature data for each functional unit.

[0052] In this embodiment, due to potential electromagnetic interference or instantaneous sensor errors in the acquisition environment, the directly obtained initial temperature may contain noise. Therefore, it is necessary to perform outlier removal and smoothing processing on multiple initial temperatures to obtain the accurate target temperature.

[0053] Specifically, an initial temperature range check is performed on each functional unit. The preset effective temperature range is -40°C ≤ T ≤ 125°C, and invalid data exceeding this range is removed. Simultaneously, abnormal data such as NaN values ​​and fixed-value jitter in the original data are removed. If the percentage of valid readings for a functional unit is lower than a preset threshold p_min (e.g., 80%), a rollback strategy of "holding the previous valid value + slow decay" is adopted, and a safety upper limit fan warning is triggered. The valid data after outlier removal is clamped to ensure the data is within a reasonable physical range and to avoid the impact of extreme values ​​on subsequent calculations. A circular buffer queue of length W (e.g., W=5) is maintained for each functional unit to store the most recent W valid data after outlier removal and clamping, providing data support for smoothing processing. The median filter (window k=3) and the first-order exponential smoothing (EMA, coefficient αT, such as 0.3) algorithm are applied sequentially to the data in the circular buffer queue to eliminate random fluctuations in the data and obtain smoothed temperature data. Among them, the median filter is used to remove isolated noise points, and the first-order exponential smoothing is used to weaken short-term fluctuations in the data and improve data stability.

[0054] After the above processing, the target temperatures of the central processing unit (T_cpu), the switching chip (T_switch), the environmental sensor (T_ambient), and each optical module (T_sfp_i_filt) are obtained. The average target temperature (T_sfp_avg) and the maximum target temperature (T_sfp_max) of the multiple optical modules are then calculated. These target temperatures are organized into a structured thermodynamic state vector, such as V_thermal = {T_cpu, T_sfp_avg, T_sfp_max, T_switch, T_ambient, ...}, with a timestamp t, and submitted to subsequent modules.

[0055] Since the system contains multiple optical modules, statistical characteristic values ​​are calculated based on the target temperatures of the optical modules to more accurately characterize the overall heat load level and local hotspot risk of the optical module group. These statistical characteristic values ​​include the average temperature (T_sfp_avg) and the maximum temperature (T_sfp_max) of all optical modules. The maximum temperature refers to the highest value among the target temperatures of all optical modules within the current sampling period, representing the hottest optical module in the system; the average temperature reflects the overall heat distribution level of the optical modules.

[0056] Specifically, T_sfp_avg(t) = mean{ T_sfp_i_filt(t)}, i = 1..N; T_sfp_max(t) = max { T_sfp_i_filt(t)}, i = 1..N.

[0057] For example, Δt = 1s, N = 6. Acquired (approximately equal to the filtered output): Tcpu = 72.4°C, Tswitch = 68.0°C, T_ambient = 32.0°C. Temperatures of each optical module: [58.2, 61.0, 47.1, 63.5, 55.0, 59.3]°C. Calculation: T_sfp_avg = (58.2+61.0+47.1+63.5+55.0+59.3) / 6 = 57.35°C; Tsfpmax = 63.5°C. Therefore, V_thermal(t) for this period = {72.4, 57.35, 63.5, 68.0, 32.0}.

[0058] This embodiment avoids bus congestion and data loss by employing an ordered polling strategy, communication retry, and timeout control, ensuring the complete acquisition of raw temperature data. Combined with range checking, outlier removal, median filtering, and exponential smoothing, it effectively eliminates data interference and fluctuations, obtaining accurate and stable target temperatures, providing reliable data support for the accurate generation of subsequent control commands.

[0059] In some embodiments, step S200 may include, but is not limited to, steps S210 to S220: Step S210: Divide the area containing the multiple fans into multiple thermal contribution zones, each of the thermal contribution zones including at least one fan; Step S220: Calculate the initial control command for the fan corresponding to at least one functional unit in each thermal contribution area based on the highest temperature of the optical module, the average temperature of the optical module, and the target temperature of each functional unit in each thermal contribution area.

[0060] In this embodiment, to achieve refined management of the complex thermal environment, the concept of a "thermal contribution zone" is introduced. The system divides an area containing multiple fans into multiple thermal contribution zones, each including at least one fan and logically or physically corresponding to a specific heat dissipation area. For example, the fans can be divided into a "CPU fan zone" and a "system fan zone," with the former primarily responsible for heat dissipation around the central processing unit, and the latter responsible for heat dissipation of the overall environment, switching chips, and optical module areas.

[0061] For each thermal contribution zone, the system employs an independent, pre-defined weighted calculation formula to calculate the initial control command (typically expressed as the base value of the PWM duty cycle) for the fans in that zone based on the target temperature of the functional units related to heat dissipation in that zone (including the aggregated index of the optical module). In each formula, the temperature of different functional units is assigned different weighting coefficients to accurately reflect their contribution to the thermal environment of that zone.

[0062] Specifically, the working principle of the dynamic weighted model is as follows: Figure 3 As shown, different weighting formulas are set according to the different heat dissipation targets of the thermal contribution zone.

[0063] The core heat dissipation goal of the CPU heat dissipation area (corresponding to the CPU fan) is to ensure the stability of the CPU temperature. At the same time, taking into account the overall heat generation of the optical module group on the internal environment of the device, the weighted formula is: CPU_Fan_PWM=W_c1*T_cpu+W_s1*T_sfp_avg+W_a1*T_ambient.

[0064] Wherein, W_c1 is the weighting coefficient for the CPU target temperature; W_s1 is the weighting coefficient for the average temperature of the optical module; W_a1 is the weighting coefficient for the ambient temperature, used to adapt to changes in the overall ambient temperature of the device; T_cpu is the CPU target temperature; T_sfp_avg is the average temperature of the optical module; and T_ambient is the ambient temperature.

[0065] The core heat dissipation goal of the system heat dissipation area (corresponding to the system fan) is to balance the local high temperature of the optical module and the heat generation of the switching chip. The weighted formula is: SYS_Fan_PWM=W_c2*T_cpu+W_s2*T_sfp_max+W_a2*T_ambient.

[0066] Where W_s2 is the weighting coefficient for the highest temperature of the optical module; W_c2 is the weighting coefficient for the CPU temperature; W_a2 is the weighting coefficient for the ambient temperature, used for ambient temperature compensation; T_cpu is the target temperature of the CPU; T_sfp_max is the highest temperature of the optical module; and T_ambient is the ambient temperature.

[0067] Weighting coefficients (such as Wc1, Ws1, etc.) are the core of the dynamic weighted model. These coefficients can be preset according to the physical design of the product and the expected application scenario. For example, in a network-intensive device, the weights (Ws1, Ws2) related to the temperature of the optical module will be set higher (the weights can be modified by changing the configuration file, and the new configuration content is applied by reading the configuration file upon power-on). Furthermore, these weighting coefficients can be designed to be dynamically adjusted by upper-level management software (by checking the configuration file for changes, and if changes are found, the new configuration values ​​are read and applied), giving the system extremely high scenario adaptability.

[0068] This embodiment divides the heat contribution zone, ensuring that the fan in each zone only responds to the temperature changes of the corresponding functional unit. This avoids a one-size-fits-all control logic. For example, the CPU fan prioritizes CPU cooling, while the system fan focuses on the heat generated by the optical module and switching chip, reducing waste of heat dissipation resources and improving heat dissipation efficiency. Based on configurable and dynamically adjustable weight coefficients, the influence weight of temperature parameters can be flexibly adjusted according to the physical design of the device and the application scenario. For example, the weight related to the optical module is reduced in network I / O-intensive scenarios, and the CPU weight is increased in computation-intensive scenarios, so that the heat dissipation strategy is accurately matched with the device workload characteristics, solving the problem of poor scenario adaptability of traditional methods.

[0069] In some embodiments, step S210 may include, but is not limited to, step S211: Step S211: Based on the physical installation location of the multiple fans and the preset cooling target, the area containing the multiple fans is divided into multiple thermal contribution areas. The multiple thermal contribution areas include a central processing unit thermal contribution area and a system thermal contribution area. The central processing unit thermal contribution area includes at least one first fan for cooling the central processing unit, and the system thermal contribution area includes at least one second fan for cooling system board-level components.

[0070] In this embodiment, to achieve refined allocation of heat dissipation resources, the area containing multiple fans is divided into multiple thermal contribution zones, which can be determined based on the physical installation location of the multiple fans within the terminal device and the preset cooling target. These multiple thermal contribution zones are mainly divided into two core areas: the central processing unit thermal contribution zone and the system thermal contribution zone.

[0071] Specifically, the central processing unit (CPU) thermal contribution area primarily corresponds to the core component with the highest computational density in the device. According to a preset cooling target, the task of this area is to maintain the CPU within its operating temperature range. Physically, this thermal contribution area includes at least one first fan, which is typically physically mounted above or near the CPU heatsink, dedicated to cooling the CPU and creating a directional airflow for heat dissipation.

[0072] The system thermal contribution area primarily corresponds to system board-level components other than the CPU. These board-level components typically include ASICs, FPGAs, dual in-line memory modules (DIMMs), and numerous high-speed pluggable optical modules (such as SFP+ and QSFP28). Based on the preset cooling goals, the task of this area is to maintain a uniform temperature throughout the entire chassis and eliminate localized hotspots. Physically, this thermal contribution area includes at least one secondary fan, typically configured as a system exhaust fan or chassis fan, to cool the system board-level components by dissipating heat through an overall airflow system within the chassis.

[0073] The thermal contribution area division in this embodiment is not a fixed architecture, but rather a general framework based on "heat dissipation execution unit + weighted linear combination," which has strong scalability and allows for the addition of new thermal contribution areas according to the heat source type of the device. For example, a "power supply fan area" can be added specifically for the temperature of the power supply module, or a "storage fan area" for NVMe hard drive groups, or an "I / O fan area" for specific expansion cards. The generation of fan control commands for any newly added area follows the same weighted linear combination paradigm.

[0074] This embodiment divides the area by combining the physical installation location of the fan, so that the fan airflow can directly act on the cooling target, reducing heat exchange loss and solving the problem of inefficient heat dissipation caused by the mismatch between the fan and the heat dissipation object in the prior art; at the same time, the clear cooling target makes the fan only respond to the temperature change of the core heat dissipation object, avoiding interference from irrelevant temperature data.

[0075] In some embodiments, step S220 may include, but is not limited to, steps S221 to S222: Step S221: For the thermal contribution area of ​​the central processing unit, the initial control command of the fan corresponding to at least one functional unit in the thermal contribution area of ​​the central processing unit is obtained by weighting the initial temperature of the central processing unit, the average temperature of the optical module, and the ambient temperature with the first set of weight parameters. The first set of weight parameters includes the first weight parameter corresponding to the initial temperature of the central processing unit, the weight parameter corresponding to the average temperature of the optical module, and the first weight parameter corresponding to the ambient temperature. Step S222: For the system thermal contribution area, the initial control command of the fan corresponding to at least one functional unit in the system thermal contribution area is obtained by weighting the initial temperature of the central processing unit, the maximum temperature of the optical module, and the ambient temperature with the second set of weight parameters. The second set of weight parameters includes the second weight parameter corresponding to the initial temperature of the central processing unit, the weight parameter corresponding to the maximum temperature of the optical module, and the second weight parameter corresponding to the ambient temperature.

[0076] In this embodiment, for the central processing unit's thermal contribution area (CPU fan area), the input parameters include the target temperature of the central processing unit (CPU) (T_cpu), the average temperature of multiple optical modules (T_sfp_avg), and the ambient temperature (T_ambient). The average temperature of the optical modules is used because the CPU fan is typically responsible for building the basic airflow within the chassis; responding to the average thermal level avoids frequent fan speed adjustments due to instantaneous fluctuations in the temperature of a single optical module, thus maintaining stable operation. The first set of weighted parameters includes a first weighted parameter (W_c1) corresponding to the target temperature of the central processing unit (CPU), a weighted parameter (W_s1) corresponding to the average temperature of the optical modules, and a first weighted parameter (W_a1) corresponding to the ambient temperature.

[0077] For the system thermal contribution area (system fan area), the input parameters include the target temperature of the central processing unit (CPU) (T_cpu), the maximum temperature of multiple optical modules (T_sfp_max), and the ambient temperature (T_ambient). The use of the maximum temperature of the optical modules ensures that even if only one optical module experiences a rapid temperature rise, the system fan can respond quickly, preventing localized overheating and hardware damage, thus protecting against the "weakest link" effect. The second set of weighted parameters includes a second weighted parameter corresponding to the CPU's target temperature (e.g., W_c2), a weighted parameter corresponding to the maximum temperature of the optical modules (e.g., W_s2), and a second weighted parameter corresponding to the ambient temperature (e.g., W_a2).

[0078] The first and second sets of weight parameters are pre-configured and independent of each other. Their specific values ​​are optimized based on the device's physical heat dissipation design, airflow layout, and expected typical workload. These weight parameters are typically stored in the device's configuration file, loaded when the system powers on, and can be dynamically adjusted via a management interface, thus giving the cooling system a high degree of scenario adaptability.

[0079] This embodiment improves the accuracy of initial control commands by designing dedicated temperature parameter combinations and weight parameter sets for two thermal contribution areas. The central processing unit thermal contribution area focuses on CPU temperature stability, while the system thermal contribution area focuses on suppressing local overheating of the optical module. The preset configuration and dynamic adjustment capabilities of the two sets of weight parameters enable the solution to flexibly adapt to different application scenarios such as compute-intensive and network I / O-intensive applications, as well as terminal devices with different structural layouts. Without reconstructing the core computing logic, only the weight parameters need to be adjusted to achieve scenario adaptation, solving the problem of poor scenario adaptability of traditional fixed control logic.

[0080] In some embodiments, step S300 may include, but is not limited to, steps S310 to S330: Step S310: Obtain the temperature change rate of the multiple functional units; Step S320: For each functional unit, if the temperature change rate of the functional unit is greater than a preset threshold, a nonlinear control adjustment amount is generated based on the temperature change rate of the functional unit, and the initial control command of the fan corresponding to the functional unit is adjusted based on the nonlinear control adjustment amount to obtain the target control command of the fan corresponding to the functional unit; and, Step S330: If the temperature change rate of the functional unit is less than or equal to the preset threshold, then the initial control command of the fan corresponding to the functional unit is taken as the target control command of the fan corresponding to the functional unit.

[0081] In this embodiment, the electronic device maintains a small time-series buffer (e.g., a circular queue) in memory for each key functional unit (or its corresponding temperature measurement point) to store the most recently acquired temperature readings. By performing first-order difference calculations on these time-series data, the current temperature change slope of the functional unit can be obtained, i.e., the temperature change rate (dT / dt). The temperature change rate directly characterizes the rising or falling trend of the unit's temperature and its severity. The predictive rate control logic flowchart is as follows: Figure 4 As shown.

[0082] For each monitored functional unit, the system compares its calculated real-time temperature change rate with a preset rapid temperature rise threshold. This threshold can be configured independently based on the thermal characteristics of different functional units and is typically stored in a configuration file.

[0083] When the system detects that the rate of temperature change of a certain functional unit (especially devices with low thermal inertia and rapid temperature rise, such as optical modules) exceeds a preset rapid temperature rise threshold, the system will determine that an overheating event is about to occur in that area, even if the current absolute temperature value of the functional unit is still within the safe range. At this time, the system enters predictive intervention mode.

[0084] In this mode, the system generates a nonlinear control adjustment based on the temperature change rate of the functional unit. Here, "nonlinear" means that the increment of the fan PWM duty cycle and the temperature change rate are not a simple linear relationship, but rather a nonlinear mapping relationship (e.g., an exponential function or a higher-order polynomial function). The purpose of this design is to ensure that when a rapid temperature spike is detected, the fan speed increases far beyond the increase in the temperature change rate itself. By providing excessive airflow in the early stages of a temperature spike, the impending temperature peak is suppressed in advance.

[0085] Finally, the system performs superposition or compound calculation on the initial control command based on the aforementioned nonlinear control adjustment amount to obtain the target control command for the fan corresponding to the functional unit.

[0086] When the rate of temperature change is within the normal range, it indicates that the system's thermal state is relatively stable and there are no sudden thermal shocks. At this time, to maintain stable fan operation and avoid unnecessary speed fluctuations, the system does not perform predictive intervention. The system directly uses the initial control command of the fan corresponding to the functional unit as the target control command for the fan corresponding to that functional unit.

[0087] This embodiment achieves proactive identification of rapid temperature rise by monitoring the rate of temperature change rather than relying solely on the current temperature value. Even if the current temperature of the functional unit is within a safe range, it can intervene and adjust in advance, solving the temperature overshoot problem caused by traditional reactive control that adjusts only when the temperature reaches the threshold. The design of the nonlinear control adjustment ensures that the faster the temperature rises, the greater the increase in fan speed, providing sufficient airflow in the shortest time, effectively smoothing out temperature peaks, avoiding equipment frequency reduction and downtime, and ensuring service continuity.

[0088] In some embodiments, step S310 may include, but is not limited to, steps S311 to S312: Step S311: Obtain the historical temperature of the multiple functional units; Step S312: Perform first-order difference calculation on the historical temperature of each of the plurality of functional units to obtain the temperature change rate of each of the plurality of functional units.

[0089] In this embodiment, the system allocates a fixed-length First-In-First-Out (FIFO) buffer or circular queue in memory (such as the RAM of the embedded controller) for each key functional unit that needs to monitor trends (e.g., CPU, various optical modules, switching chips, etc.). This buffer is used to store the smoothed "target temperature" values ​​of the most recent W consecutive periods of the functional unit in chronological order. Each new temperature sample value is pushed into the queue after calculation, while the oldest value is removed, thus always maintaining a latest time series of length W.

[0090] After acquiring historical temperature data, the system performs differential calculations on the time-series data of each functional unit. Specifically, by calculating the difference between the current temperature value and the temperature value of the previous (or several previous) time periods, the current temperature change slope (i.e., the rate of temperature change dT / dt) of that functional unit is obtained. This first-order differential calculation result intuitively reflects the rate and direction of temperature change (heating or cooling), enabling the control algorithm to keenly capture sudden changes in heat load and providing a quantitative basis for subsequent judgment on whether to trigger predictive intervention.

[0091] This embodiment avoids calculation distortion caused by interference from original data or missing data by clearly defining the source of historical temperature (target temperature after preprocessing), validity verification rules, and anomaly handling strategies, ensuring that the rate of change can truly reflect the temperature change trend of the functional unit. It adopts a first-order difference algorithm, which has low computational complexity, is compatible with the firmware operating environment of the embedded controller, does not increase the computational burden of the device, and can meet the real-time requirements of temperature control.

[0092] This application's embodiments, by collecting and preprocessing temperature data from all core functional units, overcome the limitations of traditional methods that only focus on the temperature of a single functional unit (such as the CPU). This achieves comprehensive perception of the internal thermal distribution of the terminal device, effectively preventing localized overheating issues in easily overlooked functional units such as optical modules and switching chips, thus reducing the risk of hardware damage. Based on a dynamic weighted model, initial control commands are generated for the fans corresponding to each functional unit. The weighting coefficients can be flexibly configured according to the device design and application scenarios, ensuring precise matching of fan speed with the thermal characteristics of the functional unit. This achieves refined zoned heat dissipation and improves heat dissipation efficiency. By analyzing the rate of temperature change, overheating risks are predicted in advance. A nonlinear adjustment strategy is used to proactively increase fan speed, effectively suppressing temperature peaks and avoiding problems such as temperature overshoot, device frequency reduction, or downtime caused by traditional reactive control, ensuring stable operation and service continuity of the terminal device. The weighting coefficients support dynamic adjustment, adapting to different application scenarios such as compute-intensive and network I / O-intensive applications. Simultaneously, a closed-loop calibration mechanism balances heat dissipation effect and fan energy consumption, reducing device power consumption while ensuring stable device temperature.

[0093] Please see Figure 5 This application also provides a multi-dimensional data source predictive intelligent temperature control device 500, which can be applied to electronic devices. The electronic devices are communicatively connected to terminal devices. The terminal devices include multiple functional units and multiple fans. Each functional unit corresponds to at least one fan. The fans are used to adjust the temperature of the functional unit, thereby realizing the above-mentioned multi-dimensional data source predictive intelligent temperature control method. The device includes: Acquisition module 10 is used to acquire the target temperature of the plurality of functional units; The determining module 20 is used to determine the initial control command of the fan corresponding to at least one functional unit from among the multiple functional units based on the multiple target temperatures; The adjustment module 30 is used to adjust the initial control command of the fan corresponding to the at least one functional unit according to the temperature change rate of the multiple functional units, so as to obtain the target control command of the fan corresponding to the at least one functional unit. The control module 40 is used to control the fan corresponding to the at least one functional unit according to the target control command of the fan corresponding to the at least one functional unit.

[0094] In some embodiments, the plurality of functional units include a central processing unit, a switching chip, an environmental sensor, and a plurality of optical modules; the acquisition module 10 may include: The first acquisition submodule is used to poll the central processing unit, the switching chip, the environmental sensor and the multiple optical modules one by one through the bus interface of the terminal device to obtain the initial temperature of each functional unit; The processing submodule is used to perform outlier removal and smoothing on the multiple initial temperatures to obtain the target temperatures of the multiple functional units.

[0095] In some implementations, the acquisition module 10 may further include: The first calculation submodule is used to obtain the average temperature and the maximum temperature of the multiple optical modules based on the target temperatures of the multiple optical modules, wherein the maximum temperature is the target temperature with the largest value among the target temperatures of the multiple optical modules; Module 20 may include: A partitioning submodule is used to divide the area containing multiple fans into multiple thermal contribution zones, each of the thermal contribution zones including at least one fan; The second calculation submodule is used to calculate the initial control command of the fan corresponding to at least one functional unit in each thermal contribution area based on the highest temperature of the optical module, the average temperature of the optical module, and the target temperature of each functional unit in each thermal contribution area.

[0096] In some implementations, dividing into sub-modules may include: A partitioning unit is used to divide the area containing the multiple fans into multiple thermal contribution zones according to the physical installation positions of the multiple fans and a preset cooling target. The multiple thermal contribution zones include a central processing unit thermal contribution zone and a system thermal contribution zone. The central processing unit thermal contribution zone includes at least one first fan for cooling the central processing unit, and the system thermal contribution zone includes at least one second fan for cooling system board-level components.

[0097] In some implementations, the second computing submodule may include: The first computing unit is used to perform a weighted calculation on the thermal contribution area of ​​the central processing unit based on the initial temperature of the central processing unit, the average temperature of the optical module, and the ambient temperature, and a first set of weighting parameters to obtain the initial control command of the fan corresponding to at least one functional unit in the thermal contribution area of ​​the central processing unit. The first set of weighting parameters includes a first weighting parameter corresponding to the initial temperature of the central processing unit, a weighting parameter corresponding to the average temperature of the optical module, and a first weighting parameter corresponding to the ambient temperature. The second calculation unit is used to perform a weighted calculation on the thermal contribution area of ​​the system based on the initial temperature of the central processing unit, the maximum temperature of the optical module, and the ambient temperature, and a second set of weighting parameters to obtain the initial control command of the fan corresponding to at least one functional unit in the thermal contribution area of ​​the system. The second set of weighting parameters includes a second weighting parameter corresponding to the initial temperature of the central processing unit, a weighting parameter corresponding to the maximum temperature of the optical module, and a second weighting parameter corresponding to the ambient temperature.

[0098] In some implementations, the adjustment module 30 may include: The second acquisition submodule is used to acquire the temperature change rate of the multiple functional units; The first adjustment submodule is used to, for each of the functional units, if the temperature change rate of the functional unit is greater than a preset threshold, generate a nonlinear control adjustment amount based on the temperature change rate of the functional unit, and adjust the initial control command of the fan corresponding to the functional unit based on the nonlinear control adjustment amount to obtain the target control command of the fan corresponding to the functional unit. The second adjustment submodule is used to take the initial control command of the fan corresponding to the functional unit as the target control command of the fan corresponding to the functional unit if the temperature change rate of the functional unit is less than or equal to the preset threshold.

[0099] In some implementations, the second acquisition submodule may include: An acquisition unit is used to acquire the historical temperature of multiple functional units; The third calculation unit is used to perform first-order difference calculation on the historical temperature of each of the multiple functional units to obtain the temperature change rate of each of the multiple functional units.

[0100] The specific implementation of this multidimensional data source predictive intelligent temperature control device is basically the same as the specific embodiment of the multidimensional data source predictive intelligent temperature control method described above, and will not be repeated here.

[0101] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned multi-dimensional data source predictive intelligent temperature control method. This electronic device can be any intelligent terminal, including tablet computers, in-vehicle computers, etc.

[0102] Please see Figure 6 , Figure 6 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 801 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 802 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 802 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 802 and called and executed by the processor 801 using the multi-dimensional data source predictive intelligent temperature control method of this application embodiment. The 803 input / output interface is used to implement information input and output. The communication interface 804 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 805 transmits information between various components of the device (e.g., processor 801, memory 802, input / output interface 803, and communication interface 804); The processor 801, memory 802, input / output interface 803, and communication interface 804 are connected to each other within the device via bus 805.

[0103] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described multidimensional data source predictive intelligent temperature control method.

[0104] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0105] The multi-dimensional data source predictive intelligent temperature control method, multi-dimensional data source predictive intelligent temperature control device, electronic device, and storage medium provided in this application embodiment acquire the target temperature of each functional unit, determine the initial control command of the corresponding fan based on the temperature data, then adjust the initial control command according to the temperature change rate of each functional unit to obtain the target control command, and finally regulate the corresponding fan according to the target control command. This technical solution achieves precise and forward-looking control of the temperature of the multi-functional unit of the terminal device. This application breaks through the limitations of traditional reactive control that relies solely on a single heat source. By introducing multi-dimensional temperature sensing and predictive intervention based on temperature change trends, it can not only effectively solve the problem of local overheating of other key components besides the CPU, but also respond to sudden heat loads in advance, eliminate control lag, and avoid system frequency reduction or shutdown caused by temperature overshoot. This significantly improves the heat dissipation efficiency, operational stability, and overall performance of electronic devices under complex loads.

[0106] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0107] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0108] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0109] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0110] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0111] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0112] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0113] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0114] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0115] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0116] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A predictive intelligent temperature control method based on multi-dimensional data sources, characterized in that, The method is applied to an electronic device that is communicatively connected to a terminal device, the terminal device including multiple functional units and multiple fans, each functional unit corresponding to at least one fan, the fans being used to regulate the temperature of the functional unit, the method comprising: Obtain the target temperature of multiple functional units; Based on the multiple target temperatures, determine the initial control command for the fan corresponding to at least one of the multiple functional units; Based on the temperature change rate of the multiple functional units, the initial control command of the fan corresponding to the at least one functional unit is adjusted to obtain the target control command of the fan corresponding to the at least one functional unit. The fan corresponding to the at least one functional unit is controlled according to the target control command of the fan corresponding to the at least one functional unit.

2. The method according to claim 1, characterized in that, The multiple functional units include a central processing unit, a switching chip, an environmental sensor, and multiple optical modules; The step of obtaining the target temperature of the plurality of functional units includes: The initial temperature of each functional unit is obtained by polling the central processing unit, the switching chip, the environmental sensor, and the multiple optical modules one by one through the bus interface of the terminal device. Outlier removal and smoothing are performed on the multiple initial temperatures to obtain the target temperatures of the multiple functional units.

3. The method according to claim 2, characterized in that, After performing outlier removal and smoothing on the multiple initial temperatures to obtain the target temperatures of the multiple functional units, and before determining the initial control command for the fan corresponding to at least one functional unit from the multiple functional units based on the multiple target temperatures, the method further includes: Based on the target temperatures of the multiple optical modules, the average temperature and the maximum temperature of the multiple optical modules are obtained, wherein the maximum temperature is the target temperature with the largest value among the target temperatures of the multiple optical modules; The step of determining the initial control command for the fan corresponding to at least one functional unit from among the multiple functional units based on the multiple target temperatures includes: The area containing multiple fans is divided into multiple thermal contribution zones, each of which includes at least one fan; The initial control command for the fan corresponding to at least one functional unit in each thermal contribution area is calculated based on the highest temperature of the optical module, the average temperature of the optical module, and the target temperature of each functional unit in each thermal contribution area.

4. The method according to claim 3, characterized in that, The division of the area containing the multiple fans into multiple heat contribution zones includes: Based on the physical installation locations of the multiple fans and the preset cooling target, the area containing the multiple fans is divided into multiple thermal contribution zones. The multiple thermal contribution zones include a central processing unit thermal contribution zone and a system thermal contribution zone. The central processing unit thermal contribution zone includes at least one first fan for cooling the central processing unit, and the system thermal contribution zone includes at least one second fan for cooling system board-level components.

5. The method according to claim 4, characterized in that, The initial control command for the fan corresponding to at least one functional unit in each thermal contribution zone is calculated based on the highest temperature of the optical module, the average temperature of the optical module, and the target temperature of each functional unit in each thermal contribution zone, including: For the central processing unit's thermal contribution area, the initial control command for the fan corresponding to at least one functional unit in the central processing unit's thermal contribution area is obtained by weighting the initial temperature of the central processing unit, the average temperature of the optical module, and the ambient temperature with a first set of weight parameters. The first set of weight parameters includes a first weight parameter corresponding to the initial temperature of the central processing unit, a weight parameter corresponding to the average temperature of the optical module, and a first weight parameter corresponding to the ambient temperature. For the system thermal contribution area, the initial control command of the fan corresponding to at least one functional unit in the system thermal contribution area is obtained by weighting the initial temperature of the central processing unit, the maximum temperature of the optical module, and the ambient temperature with the second set of weight parameters. The second set of weight parameters includes the second weight parameter corresponding to the initial temperature of the central processing unit, the weight parameter corresponding to the maximum temperature of the optical module, and the second weight parameter corresponding to the ambient temperature.

6. The method according to claim 1, characterized in that, The step of adjusting the initial control command of the fan corresponding to at least one functional unit based on the temperature change rate of the plurality of functional units to obtain the target control command of the fan corresponding to at least one functional unit includes: Obtain the temperature change rate of multiple functional units; For each functional unit, if the temperature change rate of the functional unit is greater than a preset threshold, a nonlinear control adjustment amount is generated based on the temperature change rate of the functional unit, and the initial control command of the fan corresponding to the functional unit is adjusted based on the nonlinear control adjustment amount to obtain the target control command of the fan corresponding to the functional unit; and, If the temperature change rate of the functional unit is less than or equal to the preset threshold, then the initial control command of the fan corresponding to the functional unit is taken as the target control command of the fan corresponding to the functional unit.

7. The method according to claim 6, characterized in that, The step of obtaining the temperature change rate of the multiple functional units includes: Obtain the historical temperature of multiple functional units; The historical temperature of each of the multiple functional units is calculated using a first-order difference to obtain the temperature change rate of each of the multiple functional units.

8. A multi-dimensional data source predictive intelligent temperature control device, characterized in that, An electronic device communicatively connected to a terminal device, the terminal device including multiple functional units and multiple fans, each functional unit corresponding to at least one fan, the fans being used to regulate the temperature of the functional unit, the device comprising: The acquisition module is used to acquire the target temperature of the multiple functional units; The determining module is used to determine the initial control command of the fan corresponding to at least one functional unit from among the multiple functional units based on the multiple target temperatures; An adjustment module is used to adjust the initial control command of the fan corresponding to at least one functional unit according to the temperature change rate of the multiple functional units, so as to obtain the target control command of the fan corresponding to at least one functional unit. The control module is used to control the fan corresponding to the at least one functional unit according to the target control command of the fan corresponding to the at least one functional unit.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the multidimensional data source predictive intelligent temperature control method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the multidimensional data source predictive intelligent temperature control method according to any one of claims 1 to 7.

Images