A fan state monitoring method and system based on BMC

Abstract

The application relates to the technical field of fan monitoring, in particular to a fan state monitoring method and system based on BMC, which comprises the following steps: multi-parameter acquisition; demand level and energy efficiency modeling; matching coupling determination of abnormalities; abnormal mechanism cross positioning; trend image joint construction; graded abnormality prompt output. Based on the collaborative acquisition and graded coupling analysis mechanism of BMC multi-source operation parameters, the application models and analyzes the parameters such as the whole machine load rate, power input power, fan rotating speed, PWM duty ratio, motor driving current, air inlet and outlet speed and key heating device temperature in a unified energy efficiency response relationship set, greatly improves the sensitivity, accuracy and explainability of abnormal identification, and effectively solves the problems of low precision and slow response speed in timely identification of fan performance degradation caused by the dependence on only a single temperature parameter and fault triggering type control.

Description

Technical Field

[0001] This invention relates to the field of fan monitoring technology, and in particular to a fan status monitoring method and system based on BMC. Background Technology

[0002] With the continued proliferation of data centers, high-performance servers, and high-power electronic devices, the overall operating power and heat density of these systems are constantly increasing. Fans have become one of the key components ensuring the safe and stable operation of these devices, and their working status directly affects the overall heat dissipation capacity and long-term reliability. However, in actual operation, fans operate under complex and variable conditions, with frequent load fluctuations and the airflow environment coupled with the thermal environment. Traditional methods of monitoring status based on single parameters or simple thresholds are insufficient to reflect performance degradation, efficiency decline, and potential failure risks in a timely and accurate manner, easily leading to false alarms, missed alarms, or delayed warnings. How to achieve accurate perception, evolution analysis, and graded early warning of fan operating status in complex operating environments has become a significant challenge in the field of thermal management and reliability assurance for electronic devices.

[0003] Chinese Patent Application Publication No. CN113849056A discloses a fan control method and a server. The method is applied to a server, which includes a Baseboard Management Controller (BMC), a Complex Programmable Logic Device (CPLD), a monitoring point to be cooled, and a fan for cooling the monitoring point. The CPLD controls the output of a first pulse width modulation (PWM) signal generated by the BMC to the fan. The monitoring point is also equipped with a thermistor, which senses temperature changes at the monitoring point and is connected to a PWM module included in the server. The method includes: the PWM module outputting a second PWM signal corresponding to an electrical signal obtained via the thermistor, wherein the electrical signal is related to the resistance of the thermistor as it changes with temperature; when an abnormality is detected in the BMC, the CPLD controls the switching of the PWM signal output to the fan from the first PWM signal to the second PWM signal.

[0004] Therefore, the fan control method has the following problems: the method only uses a single temperature parameter collected by the thermistor for PWM adjustment, which can easily lead to inaccurate judgment of actual heat dissipation requirements when the load fluctuates, power consumption changes, or the air duct conditions are abnormal; the method uses BMC abnormality as the main trigger condition for switching control, and the control strategy relies on the fault triggering mechanism, which can easily lead to the fan abnormality being in a long-term cumulative development state without early warning. Summary of the Invention

[0005] To address this, the present invention provides a fan condition monitoring method and system based on BMC, which overcomes the problems of low accuracy and slow response speed in the prior art due to reliance on a single temperature parameter and fault-triggered control during fan operation by comprehensively analyzing the multi-parameter collaborative relationship and its evolution characteristics.

[0006] To achieve the above objectives, in one aspect, the present invention provides a fan status monitoring method based on BMC, comprising:

[0007] The BMC obtains the fan speed, PWM drive duty cycle, motor drive current, intake speed, exhaust speed and temperature of key heat-generating components in real time, as well as the current overall load rate and active power on the power input side of the device where the fan is located.

[0008] Based on the overall load rate and the active power on the power input side, the theoretical heat dissipation demand intensity level of the fan at the current moment is determined. Based on the mapping relationship between the theoretical heat dissipation demand intensity level and the speed and the PWM drive duty cycle, the energy efficiency response relationship set of the fan is constructed by combining the temperature difference response characteristics and airflow response characteristics of the fan under the corresponding operating conditions.

[0009] A coupled analysis is performed on the matching relationship between heat dissipation demand, airflow response and power input, which are characterized by the energy efficiency response relationship. Based on the analysis results of the continuous mismatch between airflow response and heat dissipation demand, the fan is determined to enter an abnormal working candidate state.

[0010] Based on the abnormal operating candidate states, cross-constraint analysis is performed on the rotational speed, the motor drive current, the air intake speed, the air exhaust speed, and the temperature of the key heat-generating device to determine the abnormal dominant mechanism of the abnormal operating candidate states;

[0011] Based on the aforementioned abnormal dominant mechanism, the decay rate of the rotational speed, the temperature rise acceleration of the key heating device, and the drift trend of the motor drive current are jointly analyzed within a preset evolution monitoring window to form a fault development trend profile corresponding to the abnormal mechanism.

[0012] Based on the fault development trend profile, an abnormal fan operation monitoring alert is output.

[0013] Furthermore, the theoretical heat dissipation demand intensity level is calculated based on a thermal power mapping model constructed from the overall load rate and the active power on the power input side to characterize the equivalent heat dissipation demand level of the equipment under different load conditions.

[0014] Furthermore, the theoretical heat dissipation demand intensity level is divided into several discrete demand level intervals according to a preset grading rule, and each demand level interval corresponds to a set of target coordinated control intervals of fan speed and PWM drive duty cycle.

[0015] Furthermore, the energy efficiency response relationship set is composed of the temperature difference decrease corresponding to a unit increase in fan speed and the airflow gain corresponding to a unit increase in current under different theoretical heat dissipation demand intensity levels.

[0016] Furthermore, the coupling analysis of the matching relationship is determined by comparing the trend consistency and phase synchronization of the changing trends of heat dissipation demand, airflow response, and electrical input.

[0017] Furthermore, when the airflow response change trend and the heat dissipation demand change trend show a directional inconsistency or amplitude imbalance within a preset duration, the fan is determined to enter the abnormal working candidate state.

[0018] Furthermore, the abnormal dominant mechanism is located based on a joint constraint determination of the characteristics of rotational speed change, the characteristics of motor drive current change, and the airflow thermal response coupling relationship between air intake speed, exhaust speed and the temperature of key heat-generating devices.

[0019] Furthermore, the fault development trend profile is composed of the joint trajectory of the speed decay trend curve, the temperature rise acceleration curve of key heat-generating components, and the motor drive current drift trend curve on the same evolution time axis.

[0020] Furthermore, the fan operation anomaly monitoring prompts are output in a graded manner based on the evolution slope and duration of the fault development trend profile, corresponding to three prompt levels: mild anomaly prompts, moderate risk warnings, and severe failure alarms.

[0021] On the other hand, the present invention also provides a fan status monitoring system based on BMC, comprising:

[0022] The parameter acquisition module is used by the BMC to acquire in real time the fan speed, PWM drive duty cycle, motor drive current, intake speed, exhaust speed and temperature of key heat-generating components, as well as the current overall load rate and active power on the power input side of the device where the fan is located.

[0023] The relationship set construction module is used to determine the theoretical heat dissipation demand intensity level of the fan at the current moment based on the overall load rate and the active power of the power input side, and construct the energy efficiency response relationship set of the fan based on the mapping relationship between the theoretical heat dissipation demand intensity level and the speed and the PWM drive duty cycle, combined with the temperature difference response characteristics and airflow response characteristics of the fan under the corresponding operating conditions.

[0024] The state determination module is used to perform coupled analysis on the matching relationship between heat dissipation demand, airflow response and power input, which are represented by the energy efficiency response relationship, and to determine the fan entering an abnormal working candidate state based on the analysis result that there is a continuous mismatch between airflow response and heat dissipation demand.

[0025] The dominant determination module is used to perform cross-constraint analysis on the rotational speed, the motor drive current, the air intake speed, the air exhaust speed, and the temperature of the key heat-generating device based on the abnormal operating candidate state, so as to determine the abnormal dominant mechanism of the abnormal operating candidate state;

[0026] The profile determination module is used to perform joint analysis on the attenuation rate of the rotational speed, the temperature rise acceleration of the key heat-generating device, and the drift trend of the motor drive current within a preset evolution monitoring window based on the abnormal dominant mechanism, so as to form a fault development trend profile corresponding to the abnormal mechanism.

[0027] The anomaly alert module is used to output anomaly monitoring alerts for fan operation based on the fault development trend profile.

[0028] Compared with existing technologies, the beneficial effects of this invention are as follows: Based on the collaborative acquisition and hierarchical coupling analysis mechanism of BMC multi-source operating parameters, it incorporates parameters such as overall load rate, power input power, fan speed, PWM duty cycle, motor drive current, intake and exhaust air speed, and temperature of key heat-generating components into a unified energy efficiency response relationship set for modeling and analysis. This achieves continuous characterization of the dynamic matching relationship between heat dissipation demand, airflow response, and power input. By comprehensively judging the trend direction, amplitude ratio, phase synchronization, and correlation among the three, it can identify the problem before the fan experiences significant performance degradation. It avoids potential risks such as delayed airflow response, decreased drive efficiency, or abnormal heat dissipation channels; further, it combines multi-dimensional evolution characteristics such as speed decay, temperature rise acceleration, and current drift to construct a fault development trend profile, realizing the full-process characterization and graded prompting of fan anomalies from "incipient stage - evolution stage - failure stage". This significantly improves the sensitivity, accuracy, and interpretability of anomaly identification while ensuring the temporal consistency and physical correlation of monitoring results. It effectively solves the problem of low accuracy and slow response speed in timely identification of fan performance degradation due to relying solely on a single temperature parameter and fault-triggered control.

[0029] Furthermore, by normalizing the overall load rate and the active power on the power input side, and constructing an equivalent thermal power characterization value in a weighted manner, the computational load intensity and power consumption level of the equipment are uniformly mapped into a quantifiable heat dissipation demand intensity index. This allows for a true reflection of the actual pressure on the heat dissipation system caused by changes in the output heat of the heat source under different operating conditions. Simultaneously, by dividing the continuously changing equivalent thermal power into multiple discrete heat dissipation demand levels and configuring target collaborative control ranges for fan speed and PWM duty cycle for each level, a stable magnitude correspondence is formed between the fan output airflow and the equipment heat intensity. This avoids excessive heat dissipation and ineffective energy consumption under low loads, and also prevents the risk of heat accumulation caused by insufficient heat dissipation under high loads. Overall, it achieves a coordinated match between heat dissipation demand intensity, drive power allocation, and fan output capacity, improving the response consistency and energy efficiency of heat dissipation control.

[0030] Furthermore, by simultaneously introducing temperature difference response and airflow gain under different theoretical heat dissipation demand intensity levels, and further constructing the temperature difference response coefficient corresponding to unit speed change and the airflow efficiency coefficient corresponding to unit current change, this embodiment unifies the fan speed regulation, motor current input, airflow change, and temperature change of key heat-generating components into the same energy efficiency evaluation framework. This transforms the response relationship between heat dissipation performance and power input from discrete empirical judgment into a quantifiable proportional relationship and efficiency feature vector expression. Through this energy efficiency response relationship set composed of temperature difference response coefficient and airflow efficiency coefficient, it is possible not only to distinguish the marginal contribution difference of fan speed increase to actual cooling effect under different load conditions, but also to effectively identify whether current increase is truly converted into effective airflow output. This provides stable, consistent, and condition-distinguishing basic characteristic parameters for subsequent heat dissipation demand matching analysis, abnormal candidate state judgment, and fault dominant mechanism identification, significantly improving the accuracy, interpretability, and engineering applicability of fan state evaluation.

[0031] Furthermore, by introducing time-synchronous modeling and trend coupling analysis of the equivalent thermal power characterization value, airflow response characteristic quantity, and motor drive current, this embodiment integrates the changes in fan heat dissipation demand, airflow output capacity, and electrical energy input under load changes into the same dynamic analysis framework for comprehensive judgment. This not only characterizes the consistency of dynamic response among the three from multiple dimensions such as the direction of change, amplitude of change, and phase synchronization relationship, but also identifies potential performance mismatch states in advance when the airflow response to heat dissipation demand is lagging, the amplitude is insufficient, or it loses coordination with energy input. This effectively distinguishes between "normal power increase caused by load increase" and abnormal operating conditions such as "fan performance degradation, drive efficiency decline, and abnormal airflow supply," avoiding misjudgment or omission caused by relying solely on a single parameter threshold. This ensures that the judgment of abnormal operating candidate states simultaneously possesses the rationality of thermal load, the verifiability of airflow response, and the traceability of electrical energy input, significantly improving the accuracy, stability, and engineering applicability of fan anomaly identification.

[0032] Furthermore, by simultaneously introducing multi-parameter coupling constraints between rotational speed, motor drive current, and the inlet and outlet air speeds and temperatures of key heat-generating components, a complete closed-loop correlation is formed between the fan's "driving behavior—airflow output—actual heat exchange results." On one hand, the consistency between rotational speed and airflow speed reflects whether the fan's mechanical output is truly converted into effective air delivery; on the other hand, the correspondence between drive current and rotational speed directly characterizes the matching state between the motor's electromagnetic driving force and the mechanical load; simultaneously, the linkage between inlet and outlet airflow and the temperature of heat-generating components can intuitively reflect whether the airflow truly participates in the effective heat exchange process. Through continuous constraints and cross-verification of the above multi-channel parameters on the same time scale, it is possible not only to distinguish different failure sources such as motor drive anomalies, mechanical load anomalies, and airflow supply anomalies, but also to avoid misjudgment based on a single parameter, thereby significantly improving the accuracy, stability, and engineering applicability of locating the dominant anomaly mechanism.

[0033] Furthermore, by jointly modeling and analyzing the rotational speed decay rate, the temperature rise acceleration of key heat-generating components, and the drift trend of motor drive current on the same evolution time axis, a stable correspondence is formed between changes in mechanical output capability, changes in electromagnetic drive state, and the heat load accumulation process in the time dimension. This allows for the simultaneous characterization of the causal transmission chain between "drive-rotation-heat exchange-temperature rise," which not only avoids the risk of misjudgment caused by fluctuations in a single parameter but also identifies the key inflection point in the evolution of anomalies from mild degradation to rapid failure in advance through the linkage characteristics between the change rates and directions of different parameters. This improves the interpretability, traceability, and early warning capability for complex composite faults.

[0034] Furthermore, by compressing the combined change trajectory consisting of the speed decay trend, the acceleration of temperature rise of key heat-generating components, and the drift trend of motor drive current into two core characterization quantities—the comprehensive evolution slope and the duration of abnormality—the three types of change processes—drive capability decay, heat dissipation capacity decline, and heat load accumulation—are quantified on a unified scale. This not only distinguishes between transient anomalies caused by short-term fluctuations and failure evolution caused by continuous degradation, but also enables graded warning control from mild to severe based on the inherent progressive relationship between change "speed—intensity—duration." This transforms maintenance decisions from experience-based judgment to dynamic judgment based on evolutionary characteristics, significantly improving the time accuracy of anomaly warnings, the rationality of risk stratification, and the foresight and safety boundary assurance capabilities of operation and maintenance.

[0035] Furthermore, using the BMC as a unified data acquisition and scheduling entry point, multi-source parameters such as load, power input, airflow response, and thermal response are correlated and modeled. Through coupled analysis of "demand-response-input," continuous judgment of fan anomalies from early mismatch to failure risk is achieved. Furthermore, a fault development trend profile is formed through multi-parameter joint evolution, which improves anomaly identification from single-point threshold judgment to process evolution judgment. This not only improves the sensitivity and accuracy of anomaly identification, but also enables early warning of fault types and risk levels, thereby effectively ensuring the long-term stable operation and safe heat dissipation of equipment. Attached Figure Description

[0036] Figure 1 This is a flowchart of the fan status monitoring method based on BMC in this embodiment;

[0037] Figure 2 This is a flowchart of S3 in this embodiment;

[0038] Figure 3 This is a flowchart of S4 in this embodiment;

[0039] Figure 4 This is a schematic diagram of the fan status monitoring system based on BMC in this embodiment. Detailed Implementation

[0040] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0041] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0042] Please see Figure 1The diagram shows a flowchart of the fan status monitoring method based on BMC in this embodiment. On one hand, this embodiment provides a fan status monitoring method based on BMC, including:

[0043] S1. The BMC obtains the fan speed, PWM drive duty cycle, motor drive current, intake speed, exhaust speed and temperature of key heat-generating components in real time, as well as the current overall load rate and active power on the power input side of the device where the fan is located.

[0044] S2. Based on the overall load rate and the active power on the power input side, determine the theoretical heat dissipation demand intensity level of the fan at the current moment, and based on the mapping relationship between the theoretical heat dissipation demand intensity level and the speed and the PWM drive duty cycle, construct the energy efficiency response relationship set of the fan in combination with the temperature difference response characteristics and airflow response characteristics of the fan under the corresponding operating conditions.

[0045] S3. Perform a coupled analysis on the matching relationship between heat dissipation demand, airflow response and power input, which are represented by the energy efficiency response relationship, and determine the fan to enter an abnormal working candidate state based on the analysis results of the continuous mismatch between airflow response and heat dissipation demand.

[0046] S4. Based on the abnormal working candidate state, perform cross-constraint analysis on the rotation speed, the motor drive current, the air intake speed, the air exhaust speed, and the temperature of the key heat-generating device to determine the abnormal dominant mechanism of the abnormal working candidate state;

[0047] S5. Based on the aforementioned abnormal dominant mechanism, the decay rate of the rotational speed, the temperature rise acceleration of the key heating device, and the drift trend of the motor drive current are jointly analyzed within a preset evolution monitoring window to form a fault development trend profile corresponding to the abnormal mechanism.

[0048] S6. Based on the fault development trend profile, output a monitoring prompt for abnormal fan operation.

[0049] In this embodiment, the fan is deployed within the heat dissipation channel of the server, communication equipment, or industrial control equipment chassis to provide forced air cooling for key heat-generating components such as the motherboard, power module, and processing unit. The fan speed is obtained by real-time pulse signals output from a Hall effect speed sensor inside the fan and counted via the speed acquisition interface of the Baseboard Management Controller (BMC). The PWM drive duty cycle is output by the BMC to the fan drive module through the PWM control port and simultaneously read back and stored as control parameters. The motor drive current is acquired by a current sampling resistor or Hall effect current sensor located in the fan power supply circuit and uploaded to the BMC via an analog acquisition module. The intake air speed is acquired by a miniature wind speed sensor located in the fan intake side duct, and the exhaust air speed is acquired by a corresponding wind speed sensor located in the fan exhaust side duct. The temperature of key heat-generating components is acquired in real-time by temperature sensors located on the processor, power devices, or heatsink base. The overall load rate is calculated by the BMC based on processor utilization, memory occupancy, and the operating status of each functional module. The active power on the power input side is measured in real-time by the power detection circuit built into the power module and fed back to the BMC via the power management bus. By combining direct sampling from the aforementioned hardware sensors with bus data readback, synchronous, continuous, and traceable acquisition of various operating parameters is achieved, providing a unified and reliable data foundation for subsequent energy efficiency relationship modeling and anomaly detection.

[0050] The preset evolution monitoring window is a fixed time length used to perform joint evolution analysis on fan speed, temperature rise acceleration of key heat-generating components, and motor drive current drift trend. It depends on the fan thermal inertia characteristics, equipment load change rate, and temperature conduction delay characteristics, and is usually set between 30 seconds and 5 minutes. In this embodiment, it is set to 120 seconds, which can stably reflect the real change process of fault development trend while taking into account both short-term fluctuation suppression and medium- and long-term abnormal evolution capture.

[0051] Based on the collaborative acquisition and hierarchical coupling analysis mechanism of BMC multi-source operating parameters, parameters such as overall load rate, power input power, fan speed, PWM duty cycle, motor drive current, intake and exhaust air speed, and temperature of key heat-generating components are incorporated into a unified energy efficiency response relationship set for modeling and analysis. This enables continuous characterization of the dynamic matching relationship between heat dissipation demand, airflow response, and power input. By comprehensively judging the trend direction, amplitude ratio, phase synchronization, and correlation among the three, potential risks such as airflow response lag, decreased drive efficiency, or abnormal heat dissipation channels can be identified in advance before significant fan performance degradation occurs. Furthermore, by combining multi-dimensional evolution characteristics such as speed decay, temperature rise acceleration, and current drift, a fault development trend profile is constructed, enabling full-process characterization and hierarchical prompting of fan anomalies from the "incipient stage—evolution stage—failure stage." This significantly improves the sensitivity, accuracy, and interpretability of anomaly identification while ensuring the temporal consistency and physical correlation of monitoring results. It effectively solves the problem of low accuracy and slow response speed in timely identification of fan performance degradation caused by relying solely on a single temperature parameter and fault-triggered control.

[0052] Specifically, the theoretical heat dissipation demand intensity level is calculated based on a thermal power mapping model constructed from the overall load rate and the active power on the power input side to characterize the equivalent heat dissipation demand level of the equipment under different load conditions.

[0053] Specifically, the theoretical heat dissipation demand intensity level is divided into several discrete demand level intervals according to a preset grading rule. Each demand level interval corresponds to a set of target coordinated control intervals of fan speed and PWM drive duty cycle.

[0054] In S2 of this embodiment, the process of determining the theoretical heat dissipation demand level of the fan at the current moment based on the overall load rate and the active power on the power input side includes:

[0055] S21. The overall load rate is normalized and denoted as the load normalization coefficient Ln, where:

[0056] Ln = Lc / Lmax, where Lc is the overall load rate and Lmax is the preset rated maximum load rate;

[0057] Simultaneously, the active power on the power input side is normalized and denoted as the power consumption normalization coefficient Pn, where:

[0058] Pn=Pc / Pr, where Pc is the active power on the power input side of the whole machine, and Pr is the preset rated power;

[0059] S22. Based on the load normalization coefficient and the power consumption normalization coefficient, construct the current equivalent thermal power characterization value Qe of the device, and calculate it according to the following thermal power mapping model:

[0060] Qe = α × Ln + β × Pn, where α is the preset load weighting coefficient, β is the preset power consumption weighting coefficient, and α + β = 1;

[0061] S23. Based on the equivalent heat power characterization value, the theoretical heat dissipation demand intensity level is divided into several discrete demand level intervals according to the preset classification rules.

[0062] The preset grading rules specifically include:

[0063] When Qe < Q1, it corresponds to a low heat dissipation requirement level;

[0064] When Q1≤Qe<Q2, it corresponds to a medium to low heat dissipation requirement level;

[0065] When Q2≤Qe<Q3, it corresponds to a medium-to-high heat dissipation requirement level;

[0066] When Qe≥Q3, it corresponds to a high heat dissipation requirement level;

[0067] Wherein, Q1 is the preset first boundary threshold, Q2 is the preset second boundary threshold, and Q3 is the preset third boundary threshold;

[0068] After determining the theoretical heat dissipation demand intensity level, the corresponding fan speed target range and PWM drive duty cycle target range are called from the preset mapping table according to the heat dissipation demand intensity level, forming the target coordinated control range of the fan under the current operating conditions, which serves as the benchmark for subsequent energy efficiency response relationship set construction and matching analysis.

[0069] In this embodiment, the preset mapping table is divided into four discrete levels according to the theoretical heat dissipation demand intensity level. The target range of fan speed and the target range of PWM drive duty cycle corresponding to each level are shown in the table below:

[0070] Preset mapping table

[0071] ;

[0072] The preset rated maximum load rate refers to the maximum load ratio that the equipment is allowed to operate stably for a long time under safe and reliable conditions. It depends on the processor specifications, power supply design capabilities and overall heat dissipation design margin, and is usually set between 85% and 100%. In this embodiment, it is set to 100%, which can fully cover all operating conditions of the equipment from light load to full load, and is used for uniform calibration of load intensity.

[0073] The preset rated power refers to the maximum active power value that the device is allowed to continuously input under rated operating conditions. It depends on the rated output capacity of the power module, the capacity of the power supply line, and the upper limit of the overall power consumption design. It is usually set between 90% and 100% of the power supply design power. In this embodiment, it is set to the rated power value, which can serve as a stable benchmark for power consumption normalization calculation.

[0074] The preset load weighting coefficient is used to characterize the proportion of the overall load rate in the equivalent heat power calculation. It depends on the correlation between the device's computing load and heat generation, as well as the power consumption ratio of the main heat sources such as CPU and GPU. It is usually set between 0.4 and 0.7. In this embodiment, it is set to 0.6, which can highlight the dominant role of changes in computing load on heat dissipation requirements.

[0075] The preset power consumption weighting coefficient is used to characterize the proportion of the influence of the active power on the power input side in the calculation of the equivalent heat power. It depends on the efficiency characteristics of the conversion of electrical energy into heat and the contribution of peripheral power consumption to the overall heat load. It is usually set between 0.3 and 0.6. In this embodiment, it is set to 0.4, which can effectively compensate for the actual heat fluctuation that cannot be fully reflected by the load rate alone.

[0076] The preset first dividing threshold is used to divide the critical equivalent thermal power point between low heat dissipation demand level and medium-low heat dissipation demand level. It depends on the stable heat exchange capability of the fan at low speed and the safe temperature rise range of the key heat-generating components. It is usually set between 0.2 and 0.35. In this embodiment, it is set to 0.25, which can ensure that the equipment can achieve stable heat dissipation with low energy consumption under light load.

[0077] The preset second dividing threshold is used to divide the critical equivalent thermal power point between low and medium heat dissipation demand levels and medium and high heat dissipation demand levels. It depends on the airflow output capacity of the fan under medium speed conditions and the heat intensity distribution under typical business load. It is usually set between 0.45 and 0.65. In this embodiment, it is set to 0.55, which can cover the main heat dissipation demand range of the equipment under normal operation.

[0078] The preset third dividing threshold is used to divide the critical equivalent thermal power point between medium-high heat dissipation demand level and high heat dissipation demand level. It depends on the maximum heat exchange capacity of the fan under high-speed operation and the maximum safe operating temperature allowed by the key heat-generating components. It is usually set between 0.8 and 0.95. In this embodiment, it is set to 0.85, which can trigger the forced enhanced heat dissipation control in advance when the equipment is under high load.

[0079] By normalizing the overall load rate and the active power on the power input side, and constructing an equivalent thermal power characterization value in a weighted manner, the computational load intensity and power consumption level of the equipment are uniformly mapped into a quantifiable heat dissipation demand intensity index. This allows for a true reflection of the actual pressure on the heat dissipation system caused by changes in the output heat of the heat source under different operating conditions. Simultaneously, by dividing the continuously changing equivalent thermal power into multiple discrete heat dissipation demand levels and configuring target collaborative control ranges for fan speed and PWM duty cycle for each level, a stable magnitude correspondence is established between fan output airflow and equipment heat intensity. This avoids excessive heat dissipation and ineffective energy consumption under low loads, while also preventing the risk of heat accumulation due to insufficient heat dissipation under high loads. Overall, this achieves a coordinated match between heat dissipation demand intensity, drive power allocation, and fan output capacity, improving the response consistency and energy efficiency of heat dissipation control.

[0080] Specifically, the energy efficiency response relationship set is composed of the temperature difference decrease corresponding to a unit increase in fan speed and the airflow gain corresponding to a unit increase in current under different theoretical heat dissipation demand intensity levels.

[0081] In S2 of this embodiment, the process of constructing the energy efficiency response relationship set of the fan based on the mapping relationship between the theoretical heat dissipation demand intensity level and the fan speed and the PWM drive duty cycle, combined with the fan's temperature difference response characteristics and airflow response characteristics under the corresponding operating conditions, includes:

[0082] S24. Within each theoretical heat dissipation demand intensity level range, obtain the inlet speed, exhaust speed and key heat-generating component temperature corresponding to different fan speeds and PWM drive duty cycles, and calculate the temperature difference response ΔT under the corresponding working conditions. ΔT=Td-Ta, where Td is the temperature of the key heat-generating component and Ta is the equivalent air heat exchange temperature on the exhaust side calculated based on the exhaust speed, preset ambient reference temperature and preset air duct heat transfer coefficient.

[0083] The preset ambient reference temperature is the initial ambient air temperature in the fan installation environment, serving as a reference for heat exchange calculation. The preset duct heat transfer coefficient is used to characterize the influence of duct structure, material, and internal flow state on the heat exchange capacity between air and heat-generating devices. Specifically, based on the air flow intensity characterized by the exhaust velocity, the ambient reference temperature is corrected by the heat transfer increment to obtain the equivalent air heat transfer temperature Ta, which reflects the actual heat removal capacity of the exhaust side per unit time. This allows the air to truly reflect the heat dissipation capacity level under the current airflow conditions. Specifically, this includes calculating the air mass flow rate: Air mass flow rate = air density × effective duct cross-sectional area × exhaust velocity. Air density is commonly taken as approximately 1.2 kg / m³. The convective heat transfer coefficient can be estimated using an empirical relationship: convective heat transfer coefficient = h0 × (exhaust velocity)^n, where h0 and n are factory-calibrated or preset parameters given in references (typically n ≈ 0.6–0.9). The heat power carried away per unit time (convective heat transfer approximation) is calculated as follows: convective heat power ≈ convective heat transfer coefficient × effective heat transfer area × (device surface temperature − ambient reference temperature). The equivalent air heat transfer temperature on the exhaust side is calculated based on energy conservation: equivalent air heat transfer temperature on the exhaust side = ambient reference temperature + convective heat power / (mass flow rate × air specific heat capacity), where the air specific heat capacity is taken as approximately 1005 J / (kg·K).

[0084] In this embodiment, with an air density of 1.2 kg / m³, an exhaust velocity of 5 m / s, and an effective duct cross-sectional area of ​​0.002 m², the calculated air mass flow rate is approximately 0.012 kg / s. With h0 = 10 and n = 0.8, the calculated convective heat transfer coefficient is approximately 10 × 5^0.8 ≈ 36 W / (m²·K). With an effective heat transfer area of ​​0.01 m², a device temperature of 60°C, and an ambient reference temperature of 25°C, the calculated convective heat power is approximately 36 × 0.01 × 35 ≈ 12.7 W. Therefore, the equivalent air heat transfer temperature on the exhaust side is approximately 25 + 12.7 / (0.012 × 1005) ≈ 26.1°C.

[0085] S25. Calculate the absolute value of the difference between the intake velocity and the exhaust velocity to obtain the airflow gain of the fan under the current operating conditions.

[0086] S26. Under the same theoretical heat dissipation demand intensity level, calculate the ratio of the temperature difference response to the change in speed within different speed change ranges to obtain the temperature difference response coefficient KT corresponding to the unit speed change, which is used to characterize the temperature difference response characteristics.

[0087] S27. Under the same theoretical heat dissipation demand intensity level, calculate the ratio of airflow gain to current change in different motor drive current change ranges to obtain the airflow efficiency coefficient KV corresponding to unit current change, which is used to characterize airflow response characteristics.

[0088] S28. Based on the temperature difference response coefficient and airflow efficiency coefficient under the same theoretical heat dissipation demand intensity level, construct the fan energy efficiency response feature vector E: E={KT, KV} corresponding to the theoretical heat dissipation demand intensity level, and form the energy efficiency response relationship set by the set of energy efficiency response feature vectors corresponding to all theoretical heat dissipation demand intensity levels.

[0089] By simultaneously introducing temperature difference response and airflow gain under different theoretical heat dissipation demand intensity levels, and further constructing the temperature difference response coefficient corresponding to unit speed change and the airflow efficiency coefficient corresponding to unit current change, this embodiment unifies fan speed regulation, motor current input, airflow change, and temperature change of key heat-generating components into the same energy efficiency evaluation framework. This transforms the response relationship between heat dissipation performance and power input from discrete empirical judgment into a quantifiable proportional relationship and efficiency feature vector expression. Through this energy efficiency response relationship set composed of temperature difference response coefficient and airflow efficiency coefficient, it is possible not only to distinguish the marginal contribution difference of fan speed increase to actual cooling effect under different load conditions, but also to effectively identify whether current increase is truly converted into effective airflow output. This provides stable, consistent, and condition-discriminating basic characteristic parameters for subsequent heat dissipation demand matching analysis, abnormal candidate state judgment, and fault dominant mechanism identification, significantly improving the accuracy, interpretability, and engineering applicability of fan condition evaluation.

[0090] Specifically, the coupling analysis of the matching relationship is determined by comparing the trend consistency and phase synchronization of the changing trends of heat dissipation demand, airflow response, and electrical input.

[0091] Specifically, when the airflow response change trend and the heat dissipation demand change trend show a directional inconsistency or amplitude imbalance within a preset duration, the fan is determined to enter the abnormal working candidate state.

[0092] Please see Figure 2 As shown, this is a flowchart of S3 in this embodiment. In S3 of this embodiment, the process of performing coupled analysis on the matching relationship between heat dissipation demand, airflow response, and electrical energy input, which are characterized by the energy efficiency response relationship set, and determining the fan to enter an abnormal working candidate state based on the analysis result that there is a continuous mismatch between airflow response and heat dissipation demand, includes:

[0093] S31. Within the preset sliding analysis time window, characterize the equivalent thermal power corresponding to the theoretical heat dissipation demand intensity level at the current moment and for N consecutive historical moments. airflow response characteristics and motor drive current Time series were constructed separately to obtain the equivalent heat power series. airflow response sequence and drive current sequence ;

[0094] The value of N is between 5 and 30, and is determined by the sampling period and the abnormal response time constant. In this embodiment, N=10.

[0095] S32, respectively, for the equivalent heat power sequence airflow response sequence and drive current sequence Trend extraction processing is performed to obtain the corresponding trend sequence, thereby revealing the changing trend of heat dissipation demand. Trends in airflow response and the trend of changes in electrical energy input ;

[0096] S33. Calculate the phase shift angle between the airflow response change trend and the heat dissipation demand change trend;

[0097] S34. Within a preset duration window, assess the consistency of direction, amplitude ratio matching, and correlation between the changing trend of heat dissipation demand and the changing trend of airflow response, as well as the correlation between the changing trend of electrical energy input and the airflow response sequence. If any of the following conditions are continuously met, it is marked as a state of trend consistency failure:

[0098] The trends in heat dissipation demand and airflow response continue to move in opposite directions.

[0099] The amplitude ratio between them deviates from the preset matching ratio range;

[0100] The phase offset angle remains greater than the preset phase synchronization threshold;

[0101] For positive and Continuously negative, or The correlation coefficient is less than the preset current-airflow correlation threshold;

[0102] S35. When a trend consistency disruption occurs and a preset duration threshold is met, the fan is determined to enter an abnormal working candidate state.

[0103] The preset sliding analysis time window is a time span parameter used for time-series analysis of heat dissipation demand, airflow response and power input change trends. It depends on the fan response inertia, thermal inertia and BMC sampling period, and is usually set between 5 and 60 seconds. In this embodiment, it is set to 20 seconds, which can avoid interference from short-term disturbances on the judgment results while taking into account trend stability and abnormal response sensitivity.

[0104] The preset phase synchronization threshold is a phase angle threshold parameter used to determine the degree of time synchronization between the changing trend of heat dissipation demand and the changing trend of airflow response. It depends on the fan airflow establishment time constant and the heat conduction delay characteristics of the equipment. It is usually set between 10 degrees and 45 degrees. In this embodiment, it is set to 25 degrees, which can effectively distinguish between normal response delay and abnormal response lag.

[0105] The preset matching ratio range is a range parameter used to constrain the proportional relationship between the change amplitude of airflow response and the change amplitude of heat dissipation demand. It depends on the fan structure size, airflow resistance characteristics and heat exchange capacity of the radiator, and is usually set between [0.6, 1.4]. In this embodiment, it is set to [0.8, 1.2], which can accurately reflect the reasonable matching range between the change of heat dissipation demand and the change of airflow response.

[0106] The preset current-airflow correlation threshold is a correlation criterion parameter used to determine the degree of coupling between changes in motor drive current and changes in airflow response. It depends on motor efficiency, fan impeller load characteristics, and duct impedance characteristics, and is usually set between 0.5 and 0.85. In this embodiment, it is set to 0.7, which can effectively identify abnormal decoupling states between motor output and actual airflow response.

[0107] The preset duration threshold is a time scale parameter used to determine whether a trend consistency disruption state constitutes a persistent mismatch. It depends on the equipment's thermal safety margin and fault tolerance response time, and is usually set between 10 seconds and 120 seconds. In this embodiment, it is set to 40 seconds, which can promptly lock in the real abnormal operating conditions while avoiding misjudgment of instantaneous fluctuations.

[0108] By introducing time-synchronous modeling and trend coupling analysis of equivalent thermal power characterization, airflow response characteristics, and motor drive current, this embodiment integrates the changes in fan cooling demand, airflow output capacity, and electrical input under load changes into the same dynamic analysis framework for comprehensive judgment. This not only characterizes the consistency of dynamic response among the three from multiple dimensions such as direction of change, amplitude of change, and phase synchronization relationship, but also identifies potential performance mismatch states in advance when the airflow response to cooling demand is lagging, insufficient in amplitude, or out of coordination with energy input. This effectively distinguishes between "normal power increase due to load increase" and abnormal operating conditions such as "fan performance degradation, decreased drive efficiency, and abnormal airflow supply," avoiding misjudgments or omissions caused by relying solely on a single parameter threshold. This ensures that the judgment of abnormal operating candidate states simultaneously possesses the rationality of thermal load, the verifiability of airflow response, and the traceability of electrical input, significantly improving the accuracy, stability, and engineering applicability of fan anomaly identification.

[0109] Specifically, the abnormal dominant mechanism is located based on the joint constraint judgment of the characteristics of rotational speed change, the characteristics of motor drive current change, and the airflow thermal response coupling relationship between air intake speed, exhaust speed and the temperature of key heat-generating devices.

[0110] Please see Figure 3 As shown, this is a flowchart of S4 in this embodiment. In S4 of this embodiment, the process of performing cross-constraint analysis on the rotational speed, the motor drive current, the air intake speed, the air exhaust speed, and the temperature of the key heat-generating device based on the abnormal operating candidate state to determine the abnormal dominant mechanism of the abnormal operating candidate state includes:

[0111] S41. During multiple consecutive sampling moments within a preset duration window after the candidate state of abnormal operation is determined to be established, the trends of rotation speed change, motor drive current change, air intake speed change, air exhaust speed change, and temperature change of key heat-generating components are extracted respectively. Correlation calculation is performed on the consistency between the rotation speed change trend and the air intake speed change trend and the air exhaust speed change trend. When at least one of the rotation speed change direction and the air intake speed change direction or the air exhaust speed change direction is continuously inconsistent within the preset duration window, or when its correlation coefficient is continuously lower than the preset airflow consistency threshold, it is determined that there is a drive mismatch abnormality between rotation drive and airflow response.

[0112] S42. Perform constraint analysis on the correspondence between the trend of motor drive current change and the trend of speed change. When the motor drive current continues to increase within the preset duration window and the speed change amplitude is lower than the minimum response increase threshold under the corresponding load conditions, it is determined that there is abnormal motor-side drive efficiency or abnormal mechanical load.

[0113] S43. Consistency judgment is made on the airflow heat transfer coupling relationship between the inlet air velocity, the exhaust air velocity and the temperature of the key heat-generating device. When the inlet air velocity and the exhaust air velocity are both within the allowable fluctuation range within the preset duration window, but the temperature of the key heat-generating device still exceeds the upper limit of the allowable temperature change range, it is determined that there is an abnormal airflow heat transfer caused by heat dissipation channel blockage, air duct leakage or local heat accumulation. When the inlet air velocity or the exhaust air velocity is continuously lower than the lower limit of the corresponding allowable fluctuation range, and the temperature of the key heat-generating device continues to rise, it is determined that the main abnormal mechanism is insufficient airflow supply capacity.

[0114] Based on the combined analysis results of the rotational speed variation characteristics, motor drive current variation characteristics, and airflow heat transfer coupling constraint between the inlet and outlet air speeds and the temperature of key heat-generating components, the dominant abnormality type of the candidate abnormal working state is identified, and the corresponding abnormality mechanism judgment result is output.

[0115] The preset airflow consistency threshold is a threshold used to characterize the degree of correlation between changes in rotational speed and changes in inlet or outlet air velocity. It depends on the fan type, duct structure, and level of environmental disturbance, and is usually set between 0.6 and 0.9. In this embodiment, it is set to 0.75, which can effectively distinguish between normal linkage airflow response and abnormal decoupling conditions.

[0116] The minimum response increase threshold is used to limit the minimum effective response amplitude threshold of the motor drive current change under the corresponding load conditions. It depends on the rated power of the motor, mechanical load characteristics and transmission structure efficiency, and is usually set between 5% and 15%. In this embodiment, it is set to 8% to avoid small random fluctuations from interfering with the abnormal determination of drive efficiency.

[0117] The allowable fluctuation range is used to limit the physical fluctuation range of air intake speed, air exhaust speed and temperature of key heat-generating components under normal operating conditions. It depends on the equipment structure size, heat dissipation capacity and ambient temperature stability, and is usually set between ±10% and ±25% of the rated reference value. In this embodiment, it is set to ±15%, which can effectively distinguish between normal operating condition fluctuations and abnormal deviations.

[0118] By synchronously introducing multi-parameter coupling constraints between fan speed, motor drive current, and the inlet and outlet air speeds and temperatures of key heat-generating components, a complete closed-loop correlation is formed between the fan's "drive behavior—airflow output—actual heat exchange result." On one hand, the consistency between fan speed and airflow speed reflects whether the fan's mechanical output is truly converted into effective air delivery; on the other hand, the correspondence between drive current and speed directly characterizes the matching state between the motor's electromagnetic drive force and the mechanical load; simultaneously, the linkage between inlet and outlet airflow and the temperature of heat-generating components directly reflects whether the airflow truly participates in the effective heat exchange process. Through continuous constraints and cross-verification of the above multi-channel parameters on the same time scale, it is possible not only to distinguish different failure sources such as motor drive anomalies, mechanical load anomalies, and airflow supply anomalies, but also to avoid misjudgment based on a single parameter, thereby significantly improving the accuracy, stability, and engineering applicability of anomaly localization.

[0119] Specifically, the fault development trend profile is composed of the joint trajectory of the speed decay trend curve, the temperature rise acceleration curve of key heat-generating components, and the motor drive current drift trend curve on the same evolution time axis.

[0120] In S5 of this embodiment, based on the anomaly-dominant mechanism, the process of jointly analyzing the attenuation rate of rotational speed, the temperature rise acceleration of key heat-generating components, and the drift trend of motor drive current within a preset evolution monitoring window to form a fault development trend profile corresponding to the anomaly mechanism includes:

[0121] S51. Within the preset evolution monitoring window, the fan speed is continuously sampled according to a unified time sampling benchmark, and the speed decay rate sequence is calculated based on the ratio of the speed difference at each sampling time to the corresponding time interval. The speed decay rate sequence is then continuously fitted over time to form a speed decay trend curve.

[0122] S52. Within the same preset evolution monitoring window, the temperature of key heating devices is continuously sampled, and the acceleration sequence of temperature change is further calculated based on the rate of temperature change within the time window. The temperature rise acceleration sequence is then processed by time expansion to form the acceleration trend curve of the temperature rise of key heating devices. Within the same preset evolution monitoring window, the motor drive current is continuously sampled, and the current sampling sequence is processed by trend fitting to extract the motor drive current drift trend curve that characterizes the direction and amplitude of current change with time.

[0123] S53. Map the speed decay trend curve, the temperature rise acceleration trend curve of key heating device, and the motor drive current drift trend curve to the same evolution time axis according to the same starting time and the same evolution duration. Align the change direction, change amplitude and change rate of each curve at the corresponding time position to obtain the multi-parameter joint change trajectory characterizing the evolution of the abnormal dominant mechanism.

[0124] S54. Based on the overall morphological characteristics of the multi-parameter joint change trajectory, the distribution of change slope, and the synergistic change relationship between curves, a fault development trend profile corresponding to the dominant abnormal mechanism is formed.

[0125] By jointly modeling and analyzing the rotational speed decay rate, the temperature rise acceleration of key heat-generating components, and the drift trend of motor drive current on the same evolution time axis, a stable correspondence is formed between changes in mechanical output capability, changes in electromagnetic drive state, and the heat load accumulation process in the time dimension. This allows for the simultaneous characterization of the causal transmission chain between "drive-rotation-heat exchange-temperature rise," which not only avoids the risk of misjudgment caused by fluctuations in a single parameter, but also identifies the key inflection point in the evolution of anomalies from mild degradation to rapid failure in advance by using the linkage characteristics between the change rates and directions of different parameters. This improves the interpretability, traceability, and early warning capability for complex composite faults.

[0126] Specifically, the fan operation anomaly monitoring prompts are output in a graded manner based on the evolution slope and duration of the fault development trend profile, corresponding to three prompt levels: mild anomaly prompts, moderate risk warnings, and severe failure alarms.

[0127] In S6 of this embodiment, the process of outputting a fan operation anomaly monitoring prompt based on the fault development trend profile includes:

[0128] S61. Within the preset evolution monitoring window, extract the overall slope of the fault development trend profile composed of the speed decay trend curve, the temperature rise acceleration trend curve of key heat-generating components, and the motor drive current drift trend curve. Based on the change amplitude and change rate of the joint change trajectory within the preset evolution monitoring window, calculate the comprehensive evolution slope characterization value.

[0129] Simultaneously, the duration for which the comprehensive evolution slope characterization value continuously exceeds the preset stability threshold within the preset evolution monitoring window is statistically analyzed to obtain the corresponding abnormal evolution duration.

[0130] S62. When the comprehensive evolution slope characterization value is within the first preset slope range and the duration of abnormal evolution is within the first preset duration range, output a corresponding mild abnormality prompt.

[0131] When the comprehensive evolution slope characterization value is within the second preset slope range or the duration of abnormal evolution is within the second preset duration range, a corresponding moderate risk warning will be output.

[0132] When the comprehensive evolution slope characterization value exceeds the third preset slope threshold and the duration of abnormal evolution exceeds the third preset duration threshold, a corresponding severe failure alarm is output.

[0133] Among them, the mild abnormality prompt is used to indicate that the fan performance has initially deviated but has not yet affected the normal heat dissipation safety of the equipment. The prompt content includes at least one or more of the following: indicating that the fan speed response is slow, the airflow efficiency is slightly reduced or the current is slightly drifted, and providing maintenance prompt information such as "It is recommended to observe the operating status" or "It is recommended to check in the subsequent maintenance cycle".

[0134] The moderate risk warning is used to characterize that the performance degradation of the fan has had a substantial impact on the heat dissipation stability of the equipment. Its prompts include at least one or more of the following: the actual airflow of the fan continues to be mismatched with the theoretical heat dissipation demand, the temperature rise of key heat-generating components accelerates, or the motor drive current fluctuates abnormally. It also provides risk handling prompts such as "recommend replacing the fan within a preset time limit" or "recommend reducing the equipment load operation".

[0135] Severe failure alarms are used to indicate that the fan has entered a high failure risk or is about to fail. The alarm message includes at least one or more of the following: the fan's airflow output capacity is seriously insufficient, the temperature of key heat-generating components is rapidly approaching or exceeding the safety threshold, the motor drive current is abnormally sudden or continuously unstable, and a mandatory alarm command of "immediate shutdown", "forced switch to backup heat dissipation unit" or "emergency fan replacement" is output.

[0136] The preset stability threshold is a comprehensive slope judgment boundary used to distinguish whether the fault development trend profile is in a stable fluctuation state or an abnormal evolution state. It depends on the normal fluctuation upper limit of the combined changes of fan speed, temperature and current under rated operating conditions. It is usually set between 1 and 1.5 times the average comprehensive evolution slope of normal operation. In this embodiment, it is set to 1.2 times the rated stability slope, which can be used to filter out false anomalies caused by short-term disturbances.

[0137] The first preset slope range is a comprehensive evolution slope value range used to characterize the intensity of mild abnormal evolution. It depends on the combined change level of the rotational speed decay rate, temperature rise acceleration and current drift amplitude in the early performance decay stage. It is usually set between the preset stable threshold and the lower limit of the moderate risk slope. In this embodiment, it is set to 1.2 times to 1.8 times the stable threshold, which can characterize the evolution state of initial degradation but before obvious risk has formed.

[0138] The first preset duration interval is a time criterion used to limit the duration of mild anomalies. It depends on the device's tolerable runtime for short-term degradation and the operation and maintenance response cycle. It is usually set between 10 minutes and 2 hours. In this embodiment, it is set between 30 minutes and 60 minutes, which can be used to distinguish between instantaneous offsets and persistent initial anomalies.

[0139] The second preset slope range is a range of values ​​for the comprehensive evolution slope used to characterize the moderate risk level. It depends on the combined change magnitude when the attenuation of driving capability and the decrease in heat dissipation capability have a significant amplification effect on the temperature rise of the core device. It is usually set between the upper limit of the first preset slope range and the lower limit of the severe failure slope. In this embodiment, it is set to 1.8 to 3 times the stable threshold, which can characterize the degree of degradation that has had a substantial impact on the heat dissipation safety of the system.

[0140] The second preset duration interval is a time criterion used to characterize the range of abnormal duration under moderate risk conditions. It depends on the thermal margin of the key heating device under risk conditions and the allowable continuous heating time. It is usually set between 1h and 12h. In this embodiment, it is set between 2h and 6h, which can be used to identify continuous risk conditions that have a cumulative impact on equipment reliability.

[0141] The third preset slope threshold is the upper limit threshold of the comprehensive evolution slope used to determine the severe failure state. It depends on the extreme combined change level when the rotation speed rapidly decreases, the temperature rise acceleration increases sharply, and the current drifts violently at the same time. It is usually set above the upper limit of the second preset slope range. In this embodiment, it is set to more than 3 times the stable threshold, which can be used to identify the high-risk evolution state that is close to the failure critical point.

[0142] The third preset duration threshold is the minimum duration boundary used to determine the triggering condition of a severe failure alarm. It depends on the safe tolerance time of the key heat-generating device under extreme heat dissipation failure state. It is usually set between 5 min and 30 min. In this embodiment, it is set to 10 min, which can be used to prevent severe abnormalities from continuing to expand in a short period of time and causing irreversible thermal damage.

[0143] By further compressing the combined change trajectory consisting of the speed decay trend, the acceleration of temperature rise of key heat-generating components, and the drift trend of motor drive current into two core characterization quantities—the comprehensive evolution slope and the duration of abnormality—the three types of change processes—drive capability decay, heat dissipation capacity decline, and heat load accumulation—are quantified and corresponded under a unified scale. This not only distinguishes between transient anomalies caused by short-term fluctuations and failure evolution caused by continuous degradation, but also enables graded warning control from mild to severe based on the inherent progressive relationship between change "speed—intensity—duration." This transforms maintenance decisions from experience-based judgment to dynamic judgment based on evolutionary characteristics, significantly improving the time accuracy of anomaly warnings, the rationality of risk stratification, and the foresight and safety boundary assurance capabilities of operation and maintenance.

[0144] Please see Figure 4 As shown, this is a schematic diagram of the fan status monitoring system based on BMC in this embodiment. Furthermore, this embodiment also provides a fan status monitoring system based on BMC, including:

[0145] The parameter acquisition module is used by the BMC to acquire in real time the fan speed, PWM drive duty cycle, motor drive current, intake speed, exhaust speed and temperature of key heat-generating components, as well as the current overall load rate and active power on the power input side of the device where the fan is located.

[0146] The relationship set construction module, which is connected to the parameter acquisition module, is used to determine the theoretical heat dissipation demand intensity level of the fan at the current moment based on the overall load rate and the active power on the power input side, and construct the energy efficiency response relationship set of the fan based on the mapping relationship between the theoretical heat dissipation demand intensity level and the speed and the PWM drive duty cycle, combined with the temperature difference response characteristics and airflow response characteristics of the fan under the corresponding operating conditions.

[0147] The state determination module, which is connected to the relationship set construction module, is used to perform coupled analysis on the matching relationship between heat dissipation demand, airflow response and power input represented by the energy efficiency response relationship set, and determine the fan to enter an abnormal working candidate state based on the analysis result that there is a continuous mismatch between airflow response and heat dissipation demand.

[0148] A dominant determination module, which is connected to the state determination module and the parameter acquisition module respectively, is used to perform cross-constraint analysis on the rotational speed, the motor drive current, the air intake speed, the air exhaust speed and the temperature of the key heating device based on the abnormal working candidate state, so as to determine the abnormal dominant mechanism of the abnormal working candidate state;

[0149] The profile determination module is connected to the dominant determination module and the parameter acquisition module respectively. It is used to perform joint analysis on the attenuation rate of the rotation speed, the temperature rise acceleration of the key heat-generating device, and the drift trend of the motor drive current within a preset evolution monitoring window based on the abnormal dominant mechanism, so as to form a fault development trend profile corresponding to the abnormal mechanism.

[0150] An anomaly alert module, which is connected to the profiling module, is used to output an abnormal fan operation monitoring alert based on the fault development trend profile.

[0151] Using BMC as a unified data acquisition and scheduling entry point, multi-source parameters such as load, power input, airflow response, and thermal response are correlated and modeled. Through coupled analysis of "demand-response-input", continuous judgment of fan anomalies from early mismatch to failure risk is achieved. Furthermore, a fault development trend profile is formed through multi-parameter joint evolution, which improves anomaly identification from single-point threshold judgment to process evolution judgment. This not only improves the sensitivity and accuracy of anomaly identification, but also enables early warning of fault type and risk level, thereby effectively ensuring the long-term stable operation and safe heat dissipation of equipment.

[0152] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A BMC-based fan status monitoring method, characterized in that, The application comprises: real-time acquisition of the fan speed, PWM drive duty cycle, motor drive current, air inlet speed, air outlet speed, and key heating device temperature by the BMC, as well as the current overall load rate of the device where the fan is located and the active power on the input side of the power supply; determination of the theoretical heat dissipation demand intensity level corresponding to the fan at the current time based on the overall load rate and the active power on the input side of the power supply, and construction of the energy efficiency response relationship set of the fan based on the mapping relationship between the theoretical heat dissipation demand intensity level and the fan speed and the PWM drive duty cycle, in combination with the temperature difference response characteristic and the airflow response characteristic of the fan under the corresponding working condition; coupling analysis of the matching relationship between the heat dissipation demand, airflow response, and electric energy input represented in the energy efficiency response relationship set, and determination of the abnormal working candidate state of the fan based on the analysis result of the continuous mismatch relationship between the airflow response and the heat dissipation demand; cross-constraint analysis of the fan speed, motor drive current, air inlet speed, air outlet speed, and key heating device temperature based on the abnormal working candidate state, to determine the abnormal dominant mechanism of the abnormal working candidate state; joint development analysis of the decay rate of the fan speed, the temperature rise acceleration of the key heating device temperature, and the drift trend of the motor drive current within a preset evolution monitoring window based on the abnormal dominant mechanism, to form a fault development trend portrait corresponding to the abnormal mechanism; output of a fan operation abnormality monitoring prompt based on the fault development trend portrait.

2. The BMC-based fan status monitoring method according to claim 1, wherein, The theoretical heat dissipation demand intensity level is calculated by a heat power mapping model constructed based on the overall load rate and the active power on the input side of the power supply, to represent the equivalent heat dissipation demand level of the device under different load working conditions.

3. The BMC-based fan status monitoring method of claim 2, wherein, The theoretical heat dissipation demand intensity level is divided into several discrete demand level intervals according to a preset grading rule, and any demand level interval corresponds to a target cooperative control interval of fan speed and PWM drive duty cycle.

4. The BMC-based fan status monitoring method of claim 3, wherein, The energy efficiency response relationship set is composed of the temperature difference drop amount corresponding to the unit speed increase of the fan and the airflow gain amount corresponding to the unit current increase under different theoretical heat dissipation demand intensity levels.

5. The BMC-based fan status monitoring method of claim 4, wherein, The coupling analysis of the matching relationship is determined by comparing the trend consistency and phase synchronization of the heat dissipation demand change trend, the airflow response change trend, and the electric energy input change trend.

6. The BMC-based fan status monitoring method of claim 5, wherein, When the airflow response change trend and the heat dissipation demand change trend present a direction inconsistency or amplitude imbalance relationship within a preset duration, it is determined that the fan enters the abnormal working candidate state.

7. The BMC-based fan status monitoring method of claim 6, wherein, The abnormal dominant mechanism positioning is based on the joint constraint determination of the speed change characteristic, the motor drive current change characteristic, and the airflow-thermal response coupling relationship between the air inlet speed, the air outlet speed, and the key heating device temperature.

8. The BMC-based fan status monitoring method of claim 7, wherein, The fault development trend portrait is composed of the joint trajectories of the speed decay trend curve, the key heating device temperature rise acceleration curve, and the motor drive current drift trend curve on the same evolution time axis.

9. The BMC-based fan status monitoring method of claim 8, wherein, The fan operation abnormality monitoring prompt is graded and output based on the evolution slope and duration of the fault development trend image, to correspond to three prompt levels of mild abnormality prompt, moderate risk early warning and severe failure warning.

10. A BMC-based fan status monitoring system constructed based on the BMC-based fan status monitoring method according to any one of claims 1-9, characterized by, Comprise: A parameter acquisition module is configured to acquire the fan speed, PWM drive duty ratio, motor drive current, air inlet speed, air outlet speed and key heating device temperature in real time by the BMC, as well as the current overall machine load rate and active power on the power input side of the device where the fan is located; A relationship set construction module is configured to determine the theoretical heat dissipation demand intensity level corresponding to the fan at the current time based on the overall machine load rate and the active power on the power input side, and to construct the energy efficiency response relationship set of the fan based on the mapping relationship between the theoretical heat dissipation demand intensity level and the fan speed and the PWM drive duty ratio, combined with the temperature difference response characteristic and the airflow response characteristic of the fan under the corresponding working condition; A state determination module is configured to analyze the matching relationship between the heat dissipation demand, airflow response and electrical energy input represented in the energy efficiency response relationship set, and to determine that the fan enters an abnormal working candidate state based on the analysis result that there is a persistent mismatch relationship between the airflow response and the heat dissipation demand; A dominant determination module is configured to cross-constrain analyze the fan speed, motor drive current, air inlet speed, air outlet speed and key heating device temperature based on the abnormal working candidate state, to determine the abnormal dominant mechanism of the abnormal working candidate state; An image determination module is configured to jointly develop analyze the attenuation rate of the fan speed, the temperature rise acceleration of the key heating device temperature and the drift trend of the motor drive current within a preset evolution monitoring window based on the abnormal dominant mechanism, to form a fault development trend image corresponding to the abnormal mechanism; An abnormality prompt module is configured to output a fan operation abnormality monitoring prompt based on the fault development trend image.

Images