A method and system for controlling the rotational speed of a fan of an air-cooled radiator

Abstract

The application discloses a kind of air-cooled radiator fan rotating speed control method and system, it is related to fan control technical field, its method includes: the vibration acceleration data on the controller circuit board of multiple fan array is collected;Fan system is currently based on the vibration acceleration data evaluation vibration state;Based on the vibration state adjustment multiple fan array Synchronization error allowable range;Real-time rotating speed information of each fan in the multiple fan array is obtained;Based on the synchronization error allowable range and the real-time rotating speed information of each fan Signal correction instruction is generated;Based on the signal correction instruction, the rotating speed of each fan in multiple fan array is adjusted, and the embodiment of the application has the advantages that the synchronization error allowable range is adjusted dynamically, effectively cope with the rotating speed synchronization control challenge of fan system under different vibration states, avoid the instability of control system and the disadvantage of hardware degradation caused by mechanical resonance.

Description

Technical Field

[0001] This application relates to the field of fan control technology, and more specifically, to a method and system for controlling the fan speed of an air-cooled radiator. Background Technology

[0002] In modern electronic devices, especially in environments with stringent heat dissipation requirements such as data centers and server racks, multi-fan arrays are widely used to provide efficient airflow cooling. To ensure uniform airflow output and reduce audible noise, a speed difference synchronization control method is typically employed to maintain a high degree of consistency in the speed of each fan in the array, thereby eliminating the beat frequency effect caused by speed differences.

[0003] Traditional speed difference synchronization control methods for multi-fan arrays aim to maintain highly consistent fan speeds to eliminate beat frequency effects caused by speed differences. However, in real-world operating environments, the interplay of factors such as equipment structure, environmental conditions, and maintenance operations can pose challenges. Under certain conditions, this method may even induce or exacerbate system instability, making it difficult to eliminate the beat frequency effect and causing continuous negative impacts on the hardware. Particularly when maintenance commands require maintaining extremely small speed differences, mechanical resonance may occur, leading to intermittent data read errors in the controller's microprocessor chip. This makes precise synchronization of fan speed differences difficult to achieve and can result in self-oscillation of the control system and continuous hardware degradation. Therefore, ensuring precise synchronization of fan speed differences to effectively eliminate the beat frequency effect and avoid the resulting self-oscillation of the control system and continuous hardware degradation is a pressing problem in the current technological field, particularly in situations where maintenance commands require maintaining extremely small speed differences in multi-fan arrays, leading to mechanical resonance and intermittent data read errors in the controller's microprocessor chip. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for controlling the fan speed of an air-cooled radiator, which has the advantages of effectively addressing the challenge of speed synchronization control of the fan system under different vibration states by dynamically adjusting the allowable range of synchronization error, and avoiding instability of the control system and hardware degradation caused by mechanical resonance.

[0005] This application provides a method for controlling the fan speed of an air-cooled radiator, including:

[0006] Collect vibration acceleration data from the controller circuit board of the multi-fan array;

[0007] Assess the current vibration state of the fan system based on vibration acceleration data;

[0008] Adjust the allowable range of synchronization error for the multi-fan array based on vibration status;

[0009] Obtain the real-time speed information of each fan in a multi-fan array;

[0010] Signal correction instructions are generated based on the allowable range of synchronization error and the real-time speed information of each fan;

[0011] Adjust the speed of each fan in the multi-fan array based on signal correction instructions.

[0012] The above approach enables the method to dynamically adjust the speed synchronization control strategy based on the vibration state of the fan system, effectively avoiding system instability and hardware damage caused by traditional fixed synchronization error control under complex vibration environments.

[0013] Furthermore, the vibration acceleration data collected from the controller circuit board of the multi-fan array includes:

[0014] The controller circuit board acquires vibration acceleration data in three-dimensional space in real time at a preset frequency based on an accelerometer.

[0015] The vibration acceleration data is digitally filtered.

[0016] The above scheme improves the accuracy and reliability of vibration state assessment by acquiring and filtering vibration acceleration data in real time.

[0017] Furthermore, assessing the current vibration state of the fan system based on vibration acceleration data includes:

[0018] Calculate the root mean square value of the vibration acceleration data within a preset time window;

[0019] Vibration state is assessed based on root mean square value, low vibration threshold, and high vibration threshold.

[0020] Vibration states include: stable vibration state, critical vibration state, and high vibration state, among which:

[0021] The stable vibration state is the vibration state when the root mean square value is lower than the low vibration threshold.

[0022] The critical vibration state is the vibration state when the root mean square value is between the low vibration threshold and the high vibration threshold.

[0023] The high vibration state is the vibration state when the root mean square value is higher than the high vibration threshold.

[0024] By introducing multi-level vibration state assessment, the system can more accurately identify the current vibration level, providing a more accurate basis for subsequent synchronization error adjustment.

[0025] Furthermore, the allowable range of synchronization error for adjusting the multi-fan array based on vibration status includes:

[0026] When the fan system is identified as being in a stable vibration state, a preset minimum value is set for the allowable range of synchronization error;

[0027] When the fan system is identified as being in a critical vibration state, the allowable range of synchronization error is set based on a piecewise linear interpolation method between the low vibration threshold and the high vibration threshold.

[0028] When the fan system is identified as being in a state of high vibration, a relatively large preset value is set for the allowable range of synchronization error.

[0029] The above scheme dynamically adjusts the allowable range of synchronization error according to different vibration states, making the control strategy more adaptable. It can maintain high precision when stable and avoid oversensitivity when under high vibration.

[0030] Furthermore, when the fan system is identified as being in a high-vibration state, a preset large value is set for the allowable range of synchronization error, including:

[0031] When the fan system is identified as being in a state of high vibration, a probing operational disturbance is applied to the multi-fan array;

[0032] The monitoring controller circuit board receives information on the physical vibration state under probing operational disturbances.

[0033] Determine whether there is a correlation between physical vibration state information and detective operational disturbances;

[0034] If a correlation is found, the high vibration state is determined to be internal resonance, and the allowable range of synchronization error is adjusted based on the internal resonance.

[0035] If no correlation is found, the high vibration state is determined to be external background vibration, and the synchronization error is maintained within the preset range based on the external background vibration.

[0036] By employing the above scheme, under high vibration conditions, by applying probing disturbances and analyzing the response, it is possible to distinguish between internal resonance and external background vibration, thereby adopting a more precise synchronization error adjustment strategy and avoiding misjudgment and excessive intervention.

[0037] Furthermore, determining whether there is a correlation between physical vibration state information and probing operational disturbances includes:

[0038] Determine the repetition period of the probe operational disturbance and the temporal characteristics of the probe operational disturbance within each period;

[0039] Synchronously acquire physical vibration status information of the controller circuit board;

[0040] The physical vibration state information is segmented according to the repetition period, and the vibration values ​​of each segment are periodically superimposed to obtain the superimposed average vibration waveform.

[0041] Determine whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probed operational disturbance;

[0042] If a vibration mode corresponding to the time-series characteristics exists, it is determined that the change in physical vibration state information is related to the probe-based operational disturbance; if no vibration mode corresponding to the time-series characteristics exists, it is determined that the change in physical vibration state information is not related to the probe-based operational disturbance.

[0043] The above scheme, through periodic superposition processing and pattern matching, can effectively identify the correlation between probing disturbances and system vibrations, thereby improving the accuracy of internal resonance judgment.

[0044] Furthermore, determining whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing operational disturbance includes:

[0045] Extract the characteristic parameters of the superimposed average vibration waveform;

[0046] Obtain the characteristic parameters of the expected response mode corresponding to the temporal characteristics of the exploratory operational disturbance;

[0047] Pattern matching is performed between the characteristic parameters of the superimposed average vibration waveform and the characteristic parameters of the expected response mode.

[0048] Based on the pattern matching results, determine whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing operational disturbance.

[0049] The above scheme, through feature parameter extraction and pattern matching, further quantifies the correspondence between vibration waveforms and disturbance timing features, thereby improving the objectivity and automation of the judgment.

[0050] Furthermore, pattern matching between the characteristic parameters of the superimposed average vibration waveform and the characteristic parameters of the expected response mode includes:

[0051] Calculate the correlation coefficient between the characteristic parameters of the superimposed average vibration waveform and the characteristic parameters of the expected response mode;

[0052] The correlation coefficient is compared with a preset correlation threshold;

[0053] Based on the comparison results of the correlation threshold, it is determined whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probed operational disturbance.

[0054] The above scheme provides a quantitative pattern matching method by calculating the correlation coefficient and comparing it with a threshold, making the correlation judgment more accurate and reliable.

[0055] Furthermore, the characteristic parameters extracted from the superimposed average vibration waveform include:

[0056] Frequency domain analysis was performed on the superimposed average vibration waveform;

[0057] Extract the characteristic parameters of the superimposed average vibration waveform after frequency domain analysis. The characteristic parameters include the dominant frequency and the energy or bandwidth within the preset frequency range.

[0058] By using the above scheme and extracting feature parameters through frequency domain analysis, the characteristics of vibration waveforms can be characterized more comprehensively and accurately, providing richer information for pattern matching.

[0059] Furthermore, this application also proposes a fan speed control system for an air-cooled radiator, comprising:

[0060] The information acquisition module is used to collect vibration acceleration data from the controller circuit board of the multi-fan array;

[0061] The condition assessment module is used to assess the current vibration state of the fan system based on vibration acceleration data;

[0062] The range adjustment module is used to adjust the allowable range of synchronization error of the multi-fan array based on the vibration status;

[0063] The speed acquisition module is used to obtain the real-time speed information of each fan in the multi-fan array;

[0064] The instruction generation module is used to generate signal correction instructions based on the allowable range of synchronization error and the real-time speed information of each fan;

[0065] The speed adjustment module is used to adjust the speed of each fan in the multi-fan array based on signal correction commands.

[0066] Through the above scheme, the system, through modular design, realizes real-time perception and evaluation of the vibration state of the fan array and dynamic adjustment of the speed synchronization error, effectively solving the problem of instability of traditional control systems in complex vibration environments.

[0067] As can be seen from the above, the air-cooled radiator fan speed control method and system provided in this application effectively address the challenge of fan speed synchronization control under different vibration states by dynamically adjusting the allowable range of synchronization error, avoiding instability of the control system and hardware degradation caused by mechanical resonance. It has the advantages of effectively addressing the challenge of fan speed synchronization control under different vibration states by dynamically adjusting the allowable range of synchronization error, and avoiding instability of the control system and hardware degradation caused by mechanical resonance. Attached Figure Description

[0068] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0069] Figure 1 This is a flowchart of the air-cooled radiator fan speed control method in an embodiment of the present invention;

[0070] Figure 2 This is a flowchart of a method for determining whether there is a correlation between physical vibration state information and probing operational disturbances in an embodiment of the present invention;

[0071] Figure 3 This is a schematic diagram of the fan speed control system for an air-cooled radiator in an embodiment of the present invention. Detailed Implementation

[0072] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0073] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0074] This invention proposes an adaptive control strategy that dynamically adjusts the allowable range for fan speed synchronization based on the current vibration state of the fan system. By acquiring vibration acceleration data from the controller circuit board in real time, the vibration state of the fan system can be accurately assessed. Based on this vibration state, the system can intelligently adjust the allowable range of synchronization error for the multi-fan array. For example, when the system is in a high-vibration state, the allowable range of synchronization error can be appropriately widened to avoid exacerbating resonance due to overly precise synchronization control; while in a stable vibration state, the allowable range of synchronization error can be tightened to achieve more precise synchronization. This dynamic adjustment mechanism, combined with real-time speed information, can generate more reasonable signal correction commands, thereby ensuring the elimination of beat frequency effects while avoiding inducing or exacerbating system instability and protecting the integrity of the hardware.

[0075] Specifically, Figure 1 A flowchart of a fan speed control method for an air-cooled radiator according to an embodiment of the present invention is shown, which specifically includes the following:

[0076] S101. Collect vibration acceleration data on the controller circuit board of the multi-fan array;

[0077] Vibration acceleration data refers to the quantitative representation of the rate of change of velocity of an object during vibration. It can be realized using accelerometers, piezoelectric sensors, or microelectromechanical systems (MEMS) sensors, such as triaxial accelerometers and single-axis accelerometers. The main purpose is to achieve real-time quantitative monitoring of the mechanical vibration generated during the operation of a fan system.

[0078] S102. Evaluate the current vibration state of the fan system based on vibration acceleration data;

[0079] Vibration status refers to the overall vibration level or mode of the fan system at present. It can be evaluated by methods such as root mean square analysis, spectrum analysis or time domain feature extraction. For example, it can be judged by comparing the root mean square value of vibration acceleration data with a preset threshold. Its main purpose is to judge the operational stability of the fan system and identify potential resonance or abnormal vibration.

[0080] S103. Adjust the allowable range of synchronization error of the multi-fan array based on vibration status;

[0081] The allowable range of synchronization error refers to the maximum possible difference in the speed of each fan in a multi-fan array. It can be set by means of a fixed value, a dynamic adjustment function, or a lookup table. For example, its value can be dynamically adjusted according to the current vibration state of the system. Its main purpose is to achieve flexible control over the synchronization accuracy of the fan array speed, so as to ensure heat dissipation performance while taking into account system stability.

[0082] S104. Obtain the real-time speed information of each fan in the multi-fan array;

[0083] S105. Generate signal correction instructions based on the allowable range of synchronization error and the real-time speed information of each fan;

[0084] Signal correction commands are control commands used to adjust the drive signals of fan motors. They can be generated in the form of pulse width modulation (PWM) signals, voltage control signals, or frequency control signals. For example, the fan speed can be adjusted by changing the duty cycle of the PWM signal. The main purpose is to adjust the real-time speed of each fan to meet the requirements of the synchronization error allowable range.

[0085] S106. Adjust the speed of each fan in the multi-fan array based on the signal correction command.

[0086] The method in this embodiment of the invention combines real-time acquisition of vibration acceleration data on the controller circuit board with dynamic evaluation of the vibration state of the fan system, and adaptively adjusts the allowable range of synchronization error of the multi-fan array based on the evaluation results. This solves the technical problem that traditional fixed synchronization control strategies are prone to inducing system self-excited oscillation and hardware degradation in complex vibration environments, and achieves the effect of improving system operation stability and hardware reliability while ensuring fan speed synchronization.

[0087] In practice, the system continuously collects vibration acceleration data from the controller circuit board of the multi-fan array. As the core of the system, the vibration status of the controller circuit board directly reflects the overall operational stability of the fan array and the presence of abnormalities such as resonance. By acquiring this vibration data, the system obtains real-time feedback on its own physical state. Based on the collected vibration acceleration data, the system then assesses the current vibration state of the fan system. This assessment process transforms the raw vibration data into understandable system state information, such as determining whether the system is in a stable, critical, or high-vibration state. This state recognition is the basis for subsequent adaptive adjustments, enabling the system to make decisions based on changes in the actual operating environment, rather than solely relying on preset parameters. Subsequently, the system dynamically adjusts the allowable range of synchronization error for the multi-fan array based on the assessed vibration state. When the system is in a high-vibration state, it may indicate a risk of resonance. In this case, widening the allowable range of synchronization error can prevent the exacerbation of resonance due to excessive pursuit of synchronization; conversely, in a stable vibration state, the allowable range can be tightened to achieve speed synchronization. This dynamic adjustment strategy ensures that the system can find an equilibrium point under different vibration environments, avoiding the self-excited oscillations that may result from traditional fixed synchronization control. While the allowable range of synchronization error is dynamically adjusted, the system continuously acquires the real-time speed information of each fan in the multi-fan array. This speed information is the direct basis for speed synchronization control. By combining the dynamically adjusted allowable range of synchronization error and the real-time speed information of each fan, the system can generate a signal correction command. The generation process of this command takes into account the current vibration status of the system and the actual requirements for synchronization accuracy, ensuring that the generated correction command meets the requirements. Finally, the system adjusts the speed of each fan in the multi-fan array based on the generated signal correction command. This adjustment process enables the speed of each fan to be maintained within the dynamically allowable synchronization error range, thereby eliminating the beat frequency effect caused by speed differences. The entire control loop forms a closed loop. By sensing its own vibration status in real time, the system adaptively adjusts the control strategy and controls the fan speed, thus maintaining operation in complex and changing environments and avoiding control failures and hardware degradation caused by environmental changes in traditional solutions.

[0088] In some preferred embodiments, this application is implemented as follows. To collect vibration acceleration data on the controller circuit board of a multi-fan array, a microelectromechanical system (MEMS) accelerometer, such as a triaxial accelerometer, can be integrated on the controller circuit board. This sensor acquires the vibration acceleration data of the circuit board in three-dimensional spatial directions in real time at a preset frequency, such as 1000 times per second. The acquired raw data is then processed by a digital filter, for example, using a low-pass filter to remove high-frequency noise, to obtain a vibration signal. Next, to evaluate the current vibration state of the fan system, the root mean square (RMS) value of the filtered vibration acceleration data can be calculated within a preset time window, such as every 5 seconds. Then, the calculated RMS value is compared with preset low vibration thresholds and high vibration thresholds. For example, when the RMS value is lower than the low vibration threshold, the system is determined to be in a stable vibration state; when the RMS value is between the low vibration threshold and the high vibration threshold, the system is determined to be in a critical vibration state; and when the RMS value is higher than the high vibration threshold, the system is determined to be in a high vibration state. Based on the assessed vibration state, the system adjusts the allowable synchronization error range of the multi-fan array accordingly. Specifically, when the system is in a stable vibration state, a preset minimum value can be set for the allowable synchronization error range to achieve speed synchronization. When the system is in a critical vibration state, piecewise linear interpolation can be used to dynamically set the allowable synchronization error range based on the position of the root mean square value between the low and high vibration thresholds, allowing it to widen appropriately as the vibration level increases. When the system is in a high vibration state, a preset larger value can be set for the allowable synchronization error range to avoid exacerbating resonance due to excessive synchronization control. Simultaneously, the system continuously acquires real-time speed information of each fan in the multi-fan array using Hall effect sensors or photoelectric encoders. This real-time speed information, along with the currently dynamically adjusted allowable synchronization error range, is input into the control logic. Based on these inputs, the control logic calculates the deviation between each fan speed and the target speed or between them, and generates corresponding signal correction commands, such as adjusting the duty cycle of the pulse width modulation (PWM) signal of each fan drive circuit. Finally, these signal correction commands are sent to the fan motor driver, thereby adjusting the speed of each fan in the multi-fan array to operate within the dynamically permissible synchronization error range.

[0089] The steps for acquiring vibration acceleration data on the controller circuit board of a multi-fan array in this embodiment of the invention include: acquiring vibration acceleration data of the controller circuit board in three-dimensional spatial directions in real time at a preset frequency based on an accelerometer; and performing digital filtering processing on the vibration acceleration data.

[0090] Among them, the preset frequency refers to the sampling rate determined before data acquisition. It can be set according to the frequency range of the vibration signal and the required analysis accuracy. Its purpose is to ensure the timeliness and completeness of vibration data acquisition. Real-time acquisition means that the data acquisition system can continuously receive and record vibration data in a near-synchronous manner. Its purpose is to capture the dynamics of the vibration state. Three-dimensional spatial direction refers to simultaneously measuring vibration acceleration in the three mutually perpendicular coordinate axes of X, Y, and Z. Its purpose is to capture the vibration trajectory and intensity of the object in space. Digital filtering processing refers to processing the acquired raw vibration data through digital signal processing algorithms to eliminate noise and interference. It can be implemented by low-pass filtering, high-pass filtering, band-pass filtering, or notch filtering, etc., with the aim of improving the signal-to-noise ratio and data quality of the vibration signal.

[0091] The proposed solution involves first acquiring vibration acceleration data of the controller circuit board of a multi-fan array in real time, using an accelerometer at a preset frequency, across three spatial directions. This method allows the system to capture comprehensive vibrations of the circuit board in all directions, avoiding potential information omissions from data from a single direction, thus providing a more complete and reliable data foundation for subsequent vibration state assessment. Simultaneously, real-time acquisition at a preset frequency ensures data freshness and timeliness, enabling the system to respond promptly to dynamic vibration conditions. Furthermore, the acquired vibration acceleration data undergoes digital filtering. This processing step removes potential noise and interference from the data, improving the purity and signal-to-noise ratio of the vibration signal. It is precisely because of the acquisition of comprehensive and purified vibration acceleration data that the accuracy of subsequent assessments of the current vibration state of the fan system based on this data is improved. This allows the entire control chain, including adjusting the synchronization error range of the multi-fan array, generating signal correction commands, and adjusting the speed of each fan based on the vibration state, to be built on a solid and accurate data foundation. This solves the control deviation and system instability problems caused by insufficient data quality and comprehensiveness, ultimately achieving synchronization of fan speed differences, eliminating beat frequency effects, and avoiding self-excited oscillation and hardware degradation of the control system.

[0092] In some embodiments, a triaxial MEMS accelerometer, such as the ADXL345, can be selected and fixed at a specific location on the controller circuit board to ensure it can sense the vibration of the circuit board. This sensor can be configured to sample data at a preset frequency of 1600Hz and transmit the vibration acceleration data in the X, Y, and Z directions to the microprocessor in real time via an SPI interface. Inside the microprocessor, a digital signal processing module can run, which performs digital filtering on the received vibration acceleration data. For example, a Butterworth low-pass filter with a cutoff frequency of 500Hz can be used to filter out high-frequency noise and interference, thereby obtaining a smooth and accurate vibration acceleration signal. The filtered three-dimensional vibration acceleration data can then be used for subsequent vibration state assessment.

[0093] The steps for evaluating the current vibration state of a fan system based on vibration acceleration data in this embodiment of the invention include: calculating the root mean square value of the vibration acceleration data within a preset time window; and evaluating the vibration state based on the root mean square value and a low vibration threshold and a high vibration threshold.

[0094] It should be noted that the vibration states here include: stable vibration state, critical vibration state, and high vibration state. Among them, the stable vibration state is the vibration state when the root mean square value is lower than the low vibration threshold; the critical vibration state is the vibration state when the root mean square value is between the low vibration threshold and the high vibration threshold; and the high vibration state is the vibration state when the root mean square value is higher than the high vibration threshold.

[0095] The preset time window refers to a fixed duration for collecting vibration acceleration data samples. This can be a fixed time length, such as several seconds or tens of seconds. Its purpose is to ensure that the analyzed vibration data is sufficiently representative, effectively smoothing out instantaneous noise and occasional impacts, thereby obtaining more stable vibration trend information. The root mean square (RMS) value is the square root of the average of the squares of the vibration acceleration data within the preset time window. Specifically, it can be obtained through mathematical calculations on the collected discrete vibration acceleration data points. Its purpose is to quantify the effective intensity or energy level of the vibration signal, providing a statistically significant measure of vibration amplitude. Compared to peak or average values, the RMS value has better robustness to random noise in the signal. The low vibration threshold and high vibration threshold are two preset numerical limits used to classify the vibration level of the fan system. Specifically, they can be obtained through statistical analysis of historical vibration data of the fan system under normal operating conditions, or according to the equipment manufacturer's recommended standards. The system is designed based on industry standards and the needs of practical applications. Its purpose is to provide clear criteria for classifying vibration states, enabling the system to distinguish between three different vibration levels: normal, warning, and abnormal. A stable vibration state refers to a fan system operating within a safe and acceptable range, specifically characterized by a root mean square (RMS) value below the low vibration threshold. This indicates smooth system operation without intervention. A critical vibration state refers to a fan system vibration level exceeding the normal range but not yet reaching a serious malfunction level, specifically characterized by an RMS value between the low and high vibration thresholds. This indicates potential system risks requiring attention and potentially preventative measures. A high vibration state refers to a fan system vibration level reaching a dangerous level, possibly indicating an impending or existing malfunction, specifically characterized by an RMS value above the high vibration threshold. This aims to trigger emergency response mechanisms, such as immediate corrective action or shutdown for inspection.

[0096] This application's solution effectively addresses the problem that raw vibration data is susceptible to noise interference and cannot accurately reflect the true vibration state of a fan system by introducing the calculation of the root mean square (RMS) value of vibration acceleration data and vibration state assessment based on multi-level thresholds. Specifically, after acquiring vibration acceleration data from the controller circuit board of a multi-fan array, the system first calculates the RMS value of this data within a preset time window. The RMS value is used because it effectively reflects the energy level of the vibration signal and has a good suppression effect on instantaneous noise and occasional impacts, thus providing a more stable and reliable vibration intensity index. By introducing the preset time window, short-term fluctuations can be smoothed, allowing the obtained RMS value to more accurately represent the overall vibration trend of the fan system over a period of time, avoiding misjudgments caused by instantaneous high or low values. Based on this, the system compares the calculated RMS value with preset low and high vibration thresholds. This hierarchical assessment mechanism clearly divides the vibration state of the fan system into three levels: stable vibration state, critical vibration state, and high vibration state. When the root mean square (RMS) value is below the low vibration threshold, the system is considered to be in a stable vibration state, indicating that the fan system is operating smoothly and the vibration level is within the normal range. When the RMS value is between the low and high vibration thresholds, the system is considered to be in a critical vibration state, suggesting that the fan system may have potential vibration risks that require attention. When the RMS value is above the high vibration threshold, the system is considered to be in a high vibration state, indicating that the fan system is vibrating violently and has a high risk of failure, requiring immediate intervention. This evaluation method based on RMS value and multi-level thresholds means that the judgment of the fan system vibration state no longer relies solely on raw, easily disturbed instantaneous data, but on a processed and more representative statistical quantity. Therefore, this solution can provide a more accurate, stable, and early warning-capable assessment result of the fan system vibration state. This accurate vibration state assessment result serves as a reliable basis for subsequent adjustments to the allowable range of synchronization errors in the multi-fan array, enabling the entire air-cooled radiator fan speed control method to respond more intelligently and robustly to the actual operating conditions of the fan system. For example, when the system detects that the vibration state changes from stable to critical or high vibration, it can adjust the allowable range of synchronization error in a timely manner. This allows the synchronization requirements to be relaxed when the vibration intensifies to avoid resonance, or the synchronization requirements to be tightened when the vibration is stable to eliminate beat frequency. This effectively avoids self-excited oscillation of the control system and continuous hardware degradation caused by vibration misjudgment, and improves the operational stability and reliability of the fan array.

[0097] In some preferred embodiments, this application is implemented as follows: Assume the fan system collects vibration acceleration data from the controller circuit board in real time via an accelerometer, with a sampling frequency of 1000Hz. To evaluate the vibration state of the fan system, a preset time window of 5 seconds can be set. The system continuously collects vibration acceleration data points within these 5 seconds, for example, 1000 data points per second, totaling 5000 data points. Then, the root mean square (RMS) value is calculated for these 5000 data points. Specifically, each data point can be squared, all squared values ​​can be added together, the average value can be calculated, and finally the square root can be taken to obtain a RMS value representing the vibration intensity within these 5 seconds. Further, a low vibration threshold of 0.5g (gravitational acceleration) and a high vibration threshold of 1.5g can be set. The system compares the calculated RMS value with these two thresholds. For example, if the calculated RMS value is 0.3g, since it is lower than 0.5g, the system determines that the fan system is in a stable vibration state. If the root mean square (RMS) value is 0.8g, the system determines that the fan system is in a critical vibration state because it falls between 0.5g and 1.5g. If the RMS value is 1.8g, the system determines that the fan system is in a high vibration state because it is higher than 1.5g. In this way, the system can obtain the vibration state information of the fan system in real time and accurately, providing precise input for subsequent speed control strategies.

[0098] The step of adjusting the allowable range of synchronization error of a multi-fan array based on vibration state in this embodiment of the invention includes: when the fan system is identified as being in a stable vibration state, a preset minimum value is set for the allowable range of synchronization error; when the fan system is identified as being in a critical vibration state, a piecewise linear interpolation method is used to set the allowable range of synchronization error between a low vibration threshold and a high vibration threshold; when the fan system is identified as being in a high vibration state, a preset large value is set for the allowable range of synchronization error.

[0099] Piecewise linear interpolation refers to a method of interpolating between given data points using a linear function. Specifically, it can be between two or more preset vibration thresholds, where the root mean square value of the current vibration acceleration data is used to calculate the corresponding allowable range of synchronization error according to a linear proportional relationship. The purpose is to achieve smooth and dynamic adjustment of the allowable range of synchronization error under critical vibration conditions, avoid sudden changes in the allowable range of synchronization error, and thus improve the precision of control and the stability of the system.

[0100] This solution identifies the current vibration state of the fan system and dynamically adjusts the allowable range of synchronization error for the multi-fan array based on the identification results, thereby achieving fine-grained control of fan speed synchronization. Specifically, when the fan system is in a stable vibration state, the system identifies that the root mean square (RMS) value is below the low vibration threshold. At this time, the fans run smoothly with small speed differences between fans. Therefore, the allowable range of synchronization error is set to a preset minimum value to ensure accurate speed synchronization, reduce noise, and improve heat dissipation efficiency. When the fan system enters a critical vibration state, i.e., the RMS value is between the low and high vibration thresholds, it indicates that there may be potential instability factors in the system. At this time, the system no longer uses a fixed allowable range of synchronization error. Instead, it sets the allowable range of synchronization error based on piecewise linear interpolation between the low and high vibration thresholds. This method allows the allowable range of synchronization error to be smoothly adjusted according to subtle changes in vibration level, maintaining a certain level of synchronization accuracy while avoiding system instability caused by overly sensitive adjustments. Furthermore, when the fan system is in a high-vibration state, i.e., the root mean square value is higher than the high-vibration threshold, the system may be significantly affected by external disturbances or internal resonance, and the speed differences between fans may be large. In this situation, if a strict synchronization error allowable range is still maintained, the control system may frequently and drastically adjust the fan speed, which will exacerbate vibration and noise. Therefore, this solution sets the synchronization error allowable range to a relatively large preset value to appropriately relax the control precision, thereby improving the robustness and stability of the system and preventing the control system from falling into self-excited oscillation. This solution is closely integrated with the step of evaluating the current vibration state of the fan system. This evaluation step provides an accurate classification of the fan system's vibration level, providing a basis for setting different synchronization error allowable ranges in this solution. It is precisely based on the distinction between stable, critical, and high-vibration states that this solution can take targeted different synchronization error allowable range adjustment strategies, enabling the fan speed control system to adaptively adjust according to the complexity of the actual operating environment. This significantly improves the overall stability and reliability of the system while ensuring heat dissipation performance, effectively avoiding control failure and hardware degradation caused by a single or coarse synchronization error allowable range setting.

[0101] In some preferred embodiments, this solution is implemented as follows: Assume the fan system continuously collects vibration acceleration data via an accelerometer and calculates its root mean square (RMS) value within a preset time window. The system presets a low vibration threshold, such as 0.5g, and a high vibration threshold, such as 2.0g. Simultaneously, it presets a very small allowable synchronization error range, such as 0.5 RPM, and a larger allowable synchronization error range, such as 5.0 RPM. Specifically, when the calculated RMS value is below 0.5g, the system identifies it as a stable vibration state, and the allowable synchronization error range is set to 0.5 RPM. This ensures that the fan speed remains highly consistent when the system is running smoothly, thereby optimizing heat dissipation and reducing noise. When the RMS value is between 0.5g and 2.0g, the system identifies it as a critical vibration state. In this case, the allowable synchronization error range is set using piecewise linear interpolation. For example, if the root mean square (RMS) value is 1.0g (exactly in the middle), the allowable synchronization error range can be calculated as (0.5 RPM + 5.0 RPM) / 2 = 2.75 RPM. More generally, the relative position of the RMS value within the [0.5g, 2.0g] interval can be linearly mapped to the corresponding value within the [0.5 RPM, 5.0 RPM] interval. For example, if the RMS value is RMS, then the allowable synchronization error range = 0.5 + (RMS - 0.5) / (2.0 - 0.5) * (5.0 - 0.5). This dynamic adjustment allows the allowable synchronization error range to widen smoothly as the vibration level gradually increases, avoiding oversensitivity in the control. When the RMS value is higher than 2.0g, the system is identified as being in a high vibration state. At this point, the allowable synchronization error range is set to 5.0 RPM. This allows for greater fan speed fluctuation range under severe vibration environments, thus preventing the control system from exacerbating instability by frequently attempting to correct the speed, and improving system robustness. Through this segmented and dynamic adjustment strategy, the fan speed control system can better adapt to the operating requirements under different vibration environments.

[0102] The step of setting a preset large value for the allowable range of synchronization error when the fan system is identified as being in a high vibration state in this embodiment of the invention includes: applying a probing operation disturbance to the multi-fan array when the fan system is identified as being in a high vibration state; monitoring the physical vibration state information of the controller circuit board under the probing operation disturbance; determining whether there is a correlation between the physical vibration state information and the probing operation disturbance; if a correlation is determined, determining that the high vibration state is internal resonance, and adjusting the allowable range of synchronization error based on the internal resonance; if no correlation is determined, determining that the high vibration state is external background vibration, and maintaining the allowable range of synchronization error at the preset value based on the external background vibration.

[0103] Among them, the detective operational disturbance refers to a purposeful and controllable change in the operating state of a multi-fan array. This can be achieved by, for example, briefly changing the speed of one or more fans, applying a pulse signal of a specific frequency, or adjusting the fan power supply voltage for a short period. The purpose is to detect the source of vibration by observing the fan system's response to this disturbance. The physical vibration state information refers to the vibration characteristic data exhibited by the controller circuit board when affected by the detective operational disturbance. This can be represented by, for example, vibration acceleration data, vibration frequency spectrum, and vibration amplitude collected in real time by an accelerometer. Its purpose is to provide objective evidence for subsequent determination of the vibration source. Determining whether there is a correlation between the physical vibration state information and the detective operational disturbance refers to analyzing whether there is a causal or synchronous relationship between the detective operational disturbance and the physical vibration state information of the controller circuit board. This can be achieved by... This can be achieved through methods such as comparing the consistency between the vibration frequency and the disturbance frequency, analyzing the trend of vibration amplitude changes with the disturbance intensity, or performing time-domain or frequency-domain correlation analysis. The purpose is to distinguish whether the high vibration state is caused by internal system factors or external environmental factors. Among them, internal resonance refers to the resonance phenomenon of the mechanical structure or component inside the fan system at a specific frequency, which amplifies the vibration energy. It can manifest as severe vibration of the fan blades, bearings, brackets, or the entire chassis structure at a specific speed or frequency. The purpose is to clarify that the vibration originates from the system's own operating characteristics. External background vibration refers to the vibration caused by external environmental factors of the fan system, such as vibration from equipment cabinets, floors, other adjacent equipment, or environmental noise. It can manifest as a continuous, random, or periodic vibration that is not directly related to the fan's own operating state. The purpose is to clarify that the vibration originates from the system's external environment.

[0104] This application's solution addresses the blindness of synchronization error tolerance adjustment strategies under high vibration conditions by introducing a refined vibration source identification mechanism. When the fan system is assessed as being in a high vibration state, instead of simply setting the synchronization error tolerance to a preset large value, a probing operational disturbance is first applied to the multi-fan array. This disturbance is carefully designed to "stimulate" the fan system in a controllable way, thereby triggering different responses to internal resonance or external vibration. Subsequently, the system monitors the physical vibration state information of the controller circuit board under the probing operational disturbance in real time, including the precise capture of key parameters such as vibration frequency and amplitude. The key is to determine whether there is a correlation between the monitored physical vibration state information and the applied probing operational disturbance. If the changes in the vibration state information show a clear synchronicity or causal relationship with the probing operational disturbance—for example, the vibration frequency is highly consistent with the disturbance frequency, or the vibration amplitude changes significantly with the disturbance intensity—then it can be accurately determined that the current high vibration state is caused by resonance within the fan system. In this scenario, the system adjusts the allowable range of synchronization error based on the characteristics of internal resonance. For example, it can further tighten the allowable range to encourage fan speeds to avoid the resonant frequency, or it can activate an active vibration suppression algorithm to counteract resonance by subtly modulating the speed of a specific fan at a high frequency, thus fundamentally solving the vibration problem. Conversely, if there is no clear correlation between changes in physical vibration state information and probing disturbances, it indicates that the current high vibration state is caused by external background vibration. In this case, since the vibration does not originate from the fan system itself, the system maintains the allowable range of synchronization error at a preset value, avoiding unnecessary reduction in fan speed synchronization and ensuring that the cooling effect is not affected. In this way, this solution complements and optimizes the previous solution that directly adjusts the allowable range of synchronization error based on vibration state. When a high vibration state is identified, instead of simply expanding the allowable range, the system intelligently identifies the source of vibration and takes targeted adjustment strategies. This differentiated processing allows the system to more accurately address different types of vibration problems, avoiding control failures or performance degradation caused by misjudging the source of vibration. For example, when high vibration is caused by internal resonance, this solution can take more aggressive suppression measures instead of simply expanding the allowable range, thereby effectively eliminating resonance; when high vibration is caused by external background vibration, it avoids unnecessary sacrifice of synchronization. This refined strategy enables the fan system to maintain higher stability and better heat dissipation performance in complex and variable operating environments, while effectively extending hardware lifespan.

[0105] In some preferred embodiments, this application is implemented as follows: When the fan system is assessed as being in a high vibration state by the root mean square value, the control system can apply a probing operational disturbance to the multi-fan array. For example, the speed of all fans in the multi-fan array can be briefly increased or decreased simultaneously by a preset small amount, such as 50 RPM, and the disturbance can be maintained for a short period of time, such as 200 milliseconds, before returning to the original speed. While applying the probing operational disturbance, the system continuously monitors the physical vibration state information on the controller circuit board. This can be achieved by acquiring vibration acceleration data in real time using a high-precision accelerometer mounted on the controller circuit board, for example, continuously acquiring triaxial acceleration data at a sampling frequency of 1000 Hz. Subsequently, the system determines whether the acquired physical vibration state information is correlated with the applied probing operational disturbance. Specifically, it can analyze whether the vibration frequency spectrum of the controller circuit board shows a peak corresponding to the disturbance frequency during the probing operational disturbance, or whether the vibration amplitude increases or decreases synchronously with the disturbance intensity. For example, if the probing disturbance is a periodic change in rotational speed, the system can perform a Fourier transform on the collected vibration data to observe whether there is a significant concentration of vibration energy near the disturbance frequency. If a correlation is found, for example, if the vibration frequency closely matches the frequency of the probing disturbance, or if the vibration amplitude increases significantly with the disturbance intensity, then the current high vibration state can be determined to be internal resonance. In this case, the system can adjust the allowable range of synchronization error based on the characteristics of internal resonance. For example, it can fine-tune the overall target rotational speed of the fan array based on the identified resonance frequency to avoid the resonance frequency point, or it can activate an active vibration suppression algorithm to cancel the resonance by slightly modulating the rotational speed of a specific fan at a high frequency, thereby effectively suppressing internal resonance without significantly expanding the allowable range of synchronization error. If no correlation is found, for example, if the vibration state of the controller circuit board does not change significantly before and after the probing disturbance, or if its vibration frequency does not match the disturbance frequency, then the current high vibration state can be determined to be external background vibration. In this situation, since the vibration does not originate from the fan system itself, the system can maintain the synchronization error within the preset range, that is, keep it at the previously set larger value, to avoid unnecessarily reducing the fan speed synchronization, thereby ensuring that the heat dissipation effect is not affected, and at the same time avoid unnecessary intervention in the normal operation of the fan.

[0106] Specifically, Figure 2 The flowchart illustrates a method for determining whether there is a correlation between physical vibration state information and probing operational disturbances according to an embodiment of the present invention, specifically including:

[0107] S201. Determine the repetition period of the probed operational disturbance and the temporal characteristics of the probed operational disturbance in each period;

[0108] S202. Synchronously acquire physical vibration status information of the controller circuit board;

[0109] S203. The physical vibration state information is segmented according to the repetition period, and the vibration values ​​of each segment of physical vibration state information are periodically superimposed to obtain the superimposed average vibration waveform.

[0110] S204. Determine whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probed operational disturbance;

[0111] S205. If a vibration mode corresponding to the time sequence characteristics exists, it is determined that the change in physical vibration state information is related to the probed operational disturbance.

[0112] S206. If there is no vibration mode corresponding to the time sequence characteristics, it is determined that the change in physical vibration state information is not related to the probed operational disturbance.

[0113] The repetition period of the probing disturbance refers to the minimum time interval in which the probing disturbance applied to the multi-fan array repeatedly occurs. It can be a preset fixed time interval or dynamically calculated using a specific algorithm, and its purpose is to provide a time reference for subsequent vibration signal analysis. The temporal characteristics of the probing disturbance within each period refer to the specific patterns of its intensity, frequency, or waveform changes over time within a repetition period. These characteristics can be obtained using a predefined waveform template or by real-time monitoring of the disturbance signal's characteristics, and their purpose is to serve as a reference standard for determining whether the physical vibration state information is affected by the disturbance. The physical vibration state information refers to the physical vibration data generated by the controller circuit board when affected by the probing disturbance. This data can be acceleration data, vibration velocity data, or vibration displacement data collected by an accelerometer, and its purpose is to reflect the actual vibration response of the circuit board. Periodic superposition processing refers to a signal processing method that aligns the physical vibration state information within multiple repetitive periods in time, accumulates the vibration values ​​at corresponding time points, and calculates the average value. It can be implemented using a digital signal processor or dedicated hardware circuitry. Its purpose is to enhance the periodic signal components related to the detected operational disturbance while suppressing random noise and aperiodic interference, thereby improving the signal-to-noise ratio. The superimposed average vibration waveform refers to the average vibration signal waveform obtained after periodic superposition processing, reflecting the controller circuit board's periodic response to the detected operational disturbance. It can be presented as a time-domain waveform or a frequency-domain spectrogram, aiming to clearly show the vibration modes related to the detected operational disturbance. Vibration modes refer to the specific shape, frequency components, amplitude variation patterns, or phase relationships exhibited by the superimposed average vibration waveform in the time or frequency domain. They can be identified using waveform matching, spectrum analysis, or feature parameter extraction, aiming to characterize the inherent regularity of the vibration signal.

[0114] This application's solution achieves accurate determination of the correlation between the physical vibration state information of the controller circuit board and the probed operational disturbance through a series of coordinated steps. First, the system determines the repetition period of the probed operational disturbance and its temporal characteristics within each period. This predetermined information serves as a benchmark for subsequent analysis, ensuring accurate reference for the vibration response. Then, while applying the probed operational disturbance, the physical vibration state information of the controller circuit board is simultaneously acquired. This synchronous acquisition ensures precise temporal alignment between the vibration data and the disturbance signal, providing a reliable data foundation for subsequent correlation analysis. Based on this, to effectively filter out environmental noise and non-periodic interference and improve the signal-to-noise ratio, the acquired physical vibration state information is segmented according to a predetermined repetition period. Next, the vibration values ​​of these segmented physical vibration state information are periodically superimposed to obtain a superimposed average vibration waveform. This superimposed averaging process significantly enhances the periodic vibration component associated with the probed operational disturbance, highlighting potential correlation signals. Ultimately, by determining whether the superimposed average vibration waveform exhibits a vibration pattern corresponding to the timing characteristics of the probed operational disturbance, the system can accurately identify the correlation between the two. If the superimposed average vibration waveform exhibits a vibration pattern consistent with the timing characteristics of the disturbance, it indicates that the vibration change of the controller circuit board is caused by the probed operational disturbance, thus determining a correlation; conversely, if there is no corresponding pattern, it is determined that there is no correlation. The coordinated operation of this series of steps enables precise identification of the vibration source when the fan system is in a high-vibration state. When a high-vibration state occurs, by applying a probed operational disturbance and using the aforementioned precise correlation determination method, the system can reliably distinguish whether the high vibration originates from resonance within the fan array or is caused by external environmental background vibration. This accurate differentiation capability allows for more precise and targeted adjustments to the allowable range of synchronization errors of the multi-fan array. For example, if it is determined to be internal resonance, more aggressive measures can be taken to suppress the resonance; if it is determined to be external background vibration, unnecessary internal system adjustments can be avoided. It is precisely because of this accurate judgment that the fan speed control system can avoid self-excited oscillation and continuous hardware degradation caused by misjudgment, thereby improving the stability and reliability of the system.

[0115] In some preferred embodiments, when the fan system detects a high vibration state, the control system can apply a probing operational disturbance to the multi-fan array. For example, this disturbance could be a periodic speed pulse with a repetition period set to 500 milliseconds. Within each cycle, the fan speed can be briefly increased by 10% from the current speed and maintained for 100 milliseconds before returning to the current speed. The rise and fall slopes and duration of this speed pulse constitute the timing characteristics of the probing operational disturbance within each cycle. Simultaneously with the application of this probing operational disturbance, a triaxial accelerometer mounted on the controller circuit board can synchronously acquire vibration acceleration data of the circuit board in the X, Y, and Z directions at a sampling frequency of 1000 Hz. This data constitutes the physical vibration state information. Subsequently, a digital signal processor (DSP) can receive this synchronously acquired vibration acceleration data. The DSP segments the continuous vibration acceleration data stream according to the 500-millisecond repetition period of the probing operational disturbance. For example, every 500 milliseconds of data is considered a segment. Then, the DSP periodically superimposes these segmented vibration acceleration data. Specifically, vibration acceleration data from multiple consecutive cycles (e.g., 100 cycles) can be time-aligned, and the acceleration values ​​at corresponding time points can be accumulated and averaged to obtain a superimposed average vibration waveform. This superimposed average vibration waveform significantly highlights the vibration response associated with a 500-millisecond periodic rotational speed pulse, while suppressing random noise and non-periodic environmental vibrations. The DSP can then analyze this superimposed average vibration waveform to determine whether it exhibits a vibration mode corresponding to the temporal characteristics of the probed operational disturbance. For example, if the probed operational disturbance is a rotational speed pulse, the expected superimposed average vibration waveform should also show a significant vibration peak or a specific frequency response near the time point of disturbance application. The DSP can extract characteristic parameters of the superimposed average vibration waveform, such as its dominant frequency, energy distribution within a specific frequency range, or waveform shape. Simultaneously, the system can pre-store characteristic parameters of the expected response mode corresponding to the temporal characteristics of the probed operational disturbance. By comparing the characteristic parameters of the superimposed average vibration waveform with those of the expected response mode, for example, calculating their correlation coefficient, and comparing it with a preset correlation threshold, the system can further refine its analysis. If the correlation coefficient is higher than the threshold, it indicates that the superimposed average vibration waveform and the temporal characteristics of the probed operational disturbance are highly matched. In this case, it is determined that the change in physical vibration state information is related to the probed operational disturbance, and the high vibration state can be further determined to be internal resonance. Conversely, if the correlation coefficient is lower than the threshold, it indicates that there is no obvious correlation between the two. In this case, it is determined that the change in physical vibration state information is not related to the probed operational disturbance, and the high vibration state can be determined to be external background vibration.

[0116] The step of determining whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probed operational disturbance in this embodiment of the invention includes: extracting feature parameters of the superimposed average vibration waveform; obtaining feature parameters of the expected response mode corresponding to the temporal characteristics of the probed operational disturbance; performing mode matching between the feature parameters of the superimposed average vibration waveform and the feature parameters of the expected response mode; and determining whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probed operational disturbance based on the mode matching result.

[0117] The process involves extracting characteristic parameters from the superimposed average vibration waveform. These parameters are quantitative information representing the waveform's characteristics, extracted from the vibration waveform. They can be achieved using time-domain features (such as peak value, root mean square value, and waveform factor) or frequency-domain features (such as spectral peak value, harmonic content, and power spectral density). The aim is to simplify the waveform into a set of easily comparable and analyzable values ​​for subsequent pattern recognition. The process also involves acquiring characteristic parameters of the expected response mode corresponding to the temporal characteristics of the probing disturbance. This expected response mode refers to the standard vibration mode that the system should theoretically or experimentally exhibit when subjected to a specific probing disturbance. The characteristic parameters are a quantitative description of this standard mode and can be obtained using a pre-established database, simulation model output, or historical experimental data. The purpose is to establish a benchmark to determine whether the actual vibration waveform is related to the probing disturbance. Among them, pattern matching is performed on the feature parameters of the superimposed average vibration waveform and the feature parameters of the expected response mode. This pattern matching refers to comparing the similarity of two sets of feature parameters through an algorithm to determine whether they belong to the same or similar modes. It can be implemented using Euclidean distance, dynamic time warping (DTW) algorithm or machine learning classifiers (such as support vector machines and neural networks). Its purpose is to quantify the degree of correlation between the actual vibration waveform and the expected response mode.

[0118] The proposed solution first extracts the feature parameters of the superimposed average vibration waveform, transforming it into a quantifiable data representation, which enables subsequent comparison and analysis. Then, it obtains the feature parameters of the expected response mode corresponding to the temporal characteristics of the probing disturbance, establishing a benchmark based on system response characteristics to measure whether the actual vibration waveform conforms to expectations. Next, it performs pattern matching between the feature parameters of the superimposed average vibration waveform and the feature parameters of the expected response mode, comparing their similarity through an algorithm to avoid errors in subjective judgment and noise interference. Finally, based on the pattern matching results, it determines whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing disturbance. This judgment mechanism based on feature parameter extraction and pattern matching ensures reliable judgment of the correlation between physical vibration state information and the probing disturbance. This judgment capability enables the differentiation between internal resonance and external background vibration when identifying a fan system in a high-vibration state. When internal resonance is identified, the allowable range of synchronization error can be adjusted accordingly to suppress resonance and prevent system self-excitement and hardware degradation. When external background vibration is identified, the allowable range of synchronization error can be maintained at a preset value, avoiding control adjustments. This vibration state assessment and corresponding adjustment of the allowable range of synchronization error ensure the accuracy and stability of fan speed control, solving the challenges of control failure and exacerbated beat frequency problems caused by misjudgment.

[0119] In some preferred embodiments, determining whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing disturbance can be implemented as follows: First, when extracting the characteristic parameters of the superimposed average vibration waveform, a Fast Fourier Transform (FFT) can be performed on the waveform for frequency domain analysis to extract its dominant frequency, energy distribution within a preset frequency range, and bandwidth as characteristic parameters. For example, if the probing disturbance is a periodic square wave signal, its expected response mode may exhibit energy concentration at a specific frequency point. Second, when obtaining the characteristic parameters of the expected response mode corresponding to the temporal characteristics of the probing disturbance, the vibration response of the system under ideal conditions can be recorded and analyzed in advance through experiments or simulations under the same probing disturbance, and its frequency domain characteristics, such as dominant frequency, energy, and bandwidth, can be extracted as characteristic parameters of the expected response mode. Then, when performing mode matching between the characteristic parameters of the superimposed average vibration waveform and the characteristic parameters of the expected response mode, the correlation coefficient between the characteristic parameters can be calculated, for example, the Pearson correlation coefficient between the two frequency spectra. Finally, based on the result of the mode matching, it is determined whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing disturbance. Specifically, if the calculated correlation coefficient is higher than a preset correlation threshold (e.g., 0.8), it can be determined that the superimposed average vibration waveform presents a vibration mode corresponding to the temporal characteristics of the probed operational disturbance; conversely, if the correlation coefficient is lower than the threshold, it is determined that no corresponding vibration mode is presented.

[0120] The step of pattern matching between the feature parameters of the superimposed average vibration waveform and the feature parameters of the expected response mode in this embodiment of the invention includes: calculating the correlation coefficient between the feature parameters of the superimposed average vibration waveform and the feature parameters of the expected response mode; comparing the correlation coefficient with a preset correlation threshold; and determining whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probed operational disturbance based on the comparison result of the correlation threshold.

[0121] The correlation coefficient is a statistical indicator used to quantify the strength and direction of the linear relationship between two variables. It can be calculated using methods such as Pearson correlation coefficient, Spearman rank correlation coefficient, or Kendall rank correlation coefficient. Its purpose is to provide a quantification of the similarity between the vibration waveform and the expected pattern. The preset correlation threshold is a numerical standard set by the system before pattern matching judgment. It can be determined based on historical data analysis, experimental testing, or expert experience. Its purpose is to serve as a critical value for judging whether the superimposed average vibration waveform presents a vibration pattern corresponding to the temporal characteristics of the probing disturbance, so as to filter out insignificant matches and improve the reliability of the judgment.

[0122] This application's solution introduces a correlation coefficient as a quantitative indicator for pattern matching, combined with a preset correlation threshold, to assess the similarity between the expected response patterns corresponding to the temporal characteristics of the superimposed average vibration waveform and the probed operational disturbance. Specifically, the correlation coefficient between the characteristic parameters of the superimposed average vibration waveform and the characteristic parameters of the expected response pattern is first calculated. This correlation coefficient captures the similarity between the two in terms of morphology, trend, or frequency components, providing a relatively stable quantifiable value of similarity even in the presence of noise or signal distortion. Subsequently, the calculated correlation coefficient is compared with a preset correlation threshold. This comparison provides an objective judgment standard, avoiding subjective bias and distinguishing between true matching and random fluctuations. When the correlation coefficient reaches or exceeds the threshold, it indicates a high degree of similarity between the superimposed average vibration waveform and the expected response pattern, thus determining that the superimposed average vibration waveform exhibits a vibration pattern corresponding to the temporal characteristics of the probed operational disturbance. This pattern matching method further improves the logic for determining whether there is a correlation between physical vibration state information and the probed operational disturbance. When the fan system is under high vibration, this solution, by applying probing operational disturbances and monitoring the physical vibration status of the controller circuit board, can distinguish whether the high vibration originates from internal resonance or external background vibration based on more reliable pattern matching results. This distinction allows for more targeted adjustments to the allowable range of synchronization errors in the multi-fan array. For example, a more aggressive speed synchronization strategy can be adopted to eliminate the resonance source for internal resonance, while the existing synchronization strategy can be maintained for external background vibration, avoiding unnecessary system intervention. Therefore, this solution can prevent self-excited oscillations in the control system and continuous hardware degradation caused by judgment errors, ensuring synchronization of fan speed differences, thereby eliminating beat frequency effects and improving system stability and reliability.

[0123] In some preferred embodiments, when pattern matching is required between the characteristic parameters of the superimposed average vibration waveform and the characteristic parameters of the expected response mode, the following specific implementation can be performed: First, the Pearson correlation coefficient can be used to calculate the linear correlation between the sequence of characteristic parameters of the superimposed average vibration waveform and the sequence of characteristic parameters of the expected response mode. For example, if the characteristic parameters are time-varying amplitude sequences or frequency domain energy distribution sequences, the Pearson correlation coefficient formula can be directly applied for calculation. After calculation, a correlation coefficient value between -1 and 1 will be obtained. Next, this calculated correlation coefficient value is compared with a preset correlation threshold, which can be trained and optimized based on historical data of the system under no disturbance and known disturbance, for example, set to 0.8. If the calculated correlation coefficient is greater than or equal to 0.8, it can be determined that the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probe disturbance, indicating a strong correlation between the system vibration and the probe disturbance. Conversely, if the correlation coefficient is less than 0.8, it is considered that there is no such corresponding vibration mode, indicating a weak or no correlation between the system vibration and the probe disturbance. In this way, the degree of matching of vibration modes can be quantitatively and objectively assessed, thereby improving the accuracy of the judgment.

[0124] The step of extracting the feature parameters of the superimposed average vibration waveform in this embodiment of the invention includes: performing frequency domain analysis on the superimposed average vibration waveform; and extracting the feature parameters of the superimposed average vibration waveform after frequency domain analysis, wherein the feature parameters include the dominant frequency and the energy or bandwidth within a preset frequency range.

[0125] Frequency domain analysis refers to the method of converting time-domain signals into frequency-domain representations. Specifically, it can be achieved using signal processing techniques such as Fast Fourier Transform (FFT) or Short-Time Fourier Transform (STFT). Its purpose is to reveal the energy distribution of different frequency components in the signal, thereby effectively separating noise and useful signals. The dominant frequency refers to the frequency component with the highest energy or amplitude in the frequency domain analysis results, and its purpose is to characterize the main periodic features in the vibration signal. The energy within the preset frequency range refers to the total energy contained in the vibration signal within a specific frequency interval, and its purpose is to focus on the vibration intensity related to specific physical phenomena (such as fan speed or structural resonance). The bandwidth refers to the width of the frequency range where the vibration signal energy is concentrated in the frequency domain analysis results, and its purpose is to reflect the breadth or complexity of the distribution of vibration frequency components.

[0126] The proposed solution transforms the time-domain signal into the frequency domain by performing frequency domain analysis on the superimposed average vibration waveform, thereby revealing the energy distribution of different frequency components in the signal. This transformation effectively filters out noise interference, highlights the main frequency components of the signal, and improves the accuracy of feature parameter extraction. Specifically, by extracting feature parameters from the superimposed average vibration waveform after frequency domain analysis, including the dominant frequency and energy or bandwidth within a preset frequency range, the characteristics of the superimposed average vibration waveform can be described more accurately. This more accurate feature parameter extraction allows for higher accuracy in subsequent pattern matching with the feature parameters of the expected response mode. This improved feature parameter extraction method is closely integrated with the previous step of determining whether there is a correlation between physical vibration state information and probing disturbances. When the fan system is in a high-vibration state, a probing disturbance is applied, and the physical vibration state information of the controller circuit board is monitored. This information is then periodically superimposed to obtain the superimposed average vibration waveform. At this point, the feature parameters extracted using frequency domain analysis can more reliably determine whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing disturbance. If this pattern exists, it can be accurately determined that the high vibration state is caused by internal resonance, thus allowing for adjustment of the synchronization error tolerance range based on the internal resonance. If it does not exist, it is determined to be external background vibration, and the synchronization error tolerance range is maintained at the preset value. Therefore, by introducing frequency domain analysis to optimize the extraction of characteristic parameters, the judgment of the source of high vibration state becomes more accurate, enabling the adoption of more effective synchronization error tolerance range adjustment strategies. Ultimately, this ensures precise synchronization of fan speed differences, effectively eliminates beat frequency effects, and avoids the resulting self-excited oscillation of the control system and continuous hardware degradation.

[0127] In some preferred embodiments, a Fast Fourier Transform (FFT) algorithm can be used to perform frequency domain analysis on the superimposed average vibration waveform. For example, the sampling frequency can be set to 1000 Hz, the acquisition time to 1 second, resulting in 1000 sampling points. Then, FFT processing is performed on these 1000 sampling points to generate a spectrum of the vibration waveform. When extracting the characteristic parameters of the superimposed average vibration waveform after frequency domain analysis, the frequency point with the highest energy in the spectrum can be identified as the dominant frequency. For example, if a significant energy peak appears at 50 Hz, then 50 Hz is the dominant frequency. Simultaneously, a frequency range can be preset, such as 20 Hz to 80 Hz, and the sum of the energy of all frequency components within this range can be calculated as the energy within the preset frequency range. Furthermore, the width of the energy concentration region in the spectrum can be calculated; for example, the frequency range where the energy decays to half of the peak energy can be defined as the bandwidth, or other statistical methods can be used to quantify the breadth of the frequency distribution. Through these specific frequency domain characteristic parameters, the characteristics of the vibration waveform can be characterized more precisely, providing an accurate data foundation for subsequent pattern matching.

[0128] Specifically, Figure 3 A schematic diagram of a fan speed control system for an air-cooled radiator according to an embodiment of the present invention is shown. The system includes:

[0129] The information acquisition module is used to collect vibration acceleration data from the controller circuit board of the multi-fan array;

[0130] The condition assessment module is used to assess the current vibration state of the fan system based on vibration acceleration data;

[0131] The range adjustment module is used to adjust the allowable range of synchronization error of the multi-fan array based on the vibration status;

[0132] The speed acquisition module is used to obtain the real-time speed information of each fan in the multi-fan array;

[0133] The instruction generation module is used to generate signal correction instructions based on the allowable range of synchronization error and the real-time speed information of each fan;

[0134] The speed adjustment module is used to adjust the speed of each fan in the multi-fan array based on signal correction commands.

[0135] This application's solution modularizes each step of the fan speed control method for air-cooled radiators at the hardware level, thereby constructing a system that effectively solves the problem of mechanical structure resonance caused by the maintenance command requirement to maintain a very small speed difference in multi-fan arrays, leading to intermittent data read errors in the controller microprocessor chip. Specifically, the information acquisition module first acquires vibration acceleration data from the multi-fan array controller circuit board in real time, which is the basis for the system to perceive its own vibration state. Subsequently, the state assessment module receives and analyzes this vibration acceleration data to determine the current vibration state of the fan system, such as whether it is stable, critical, or in a high-vibration state. Based on the judgment result of the state assessment module, the range adjustment module dynamically adjusts the allowable range of synchronization error of the multi-fan array. This means that the system can flexibly relax or tighten the requirements for fan speed synchronization accuracy according to the actual vibration situation. At the same time, the speed acquisition module continuously acquires the real-time speed information of each fan in the multi-fan array, providing accurate feedback for subsequent speed synchronization control. The command generation module comprehensively utilizes the synchronization error allowable range set by the range adjustment module and the real-time speed information of each fan provided by the speed acquisition module to calculate and generate signal correction commands. These instructions aim to guide the speeds of each fan towards synchronization and maintain them within an acceptable error range. Ultimately, the speed adjustment module receives and executes these signal correction instructions, precisely adjusting the speed of each fan in the multi-fan array. Through this tightly collaborative modular design, the system can monitor and respond to the vibration state of the fan array in real time, adaptively adjusting the speed synchronization strategy. This effectively avoids mechanical resonance induced by excessively pursuing minimal speed differences, thereby preventing data read errors from the microprocessor chip, ensuring stable operation and heat dissipation efficiency of the fan array, and eliminating beat frequency effects. This hardware-based system design significantly improves the real-time performance, stability, and reliability of control, overcoming the limitations of purely software implementations, and providing a solid guarantee for the stable operation of the fan array in complex operating environments.

[0136] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for controlling the fan speed of an air-cooled radiator, characterized in that, include: Collect vibration acceleration data from the controller circuit board of the multi-fan array; The vibration state of the fan system is assessed based on the vibration acceleration data. The synchronization error allowable range of the multi-fan array is adjusted based on the vibration state; Obtain the real-time speed information of each fan in the multi-fan array; Based on the allowable range of synchronization error and the real-time speed information of each fan, a signal correction command is generated; The speed of each fan in the multi-fan array is adjusted based on the signal correction command; The vibration acceleration data collected from the controller circuit board of the multi-fan array includes: The controller circuit board acquires vibration acceleration data in three-dimensional space in real time at a preset frequency based on an accelerometer. The vibration acceleration data is digitally filtered. The assessment of the current vibration state of the fan system based on the vibration acceleration data includes: Calculate the root mean square value of the vibration acceleration data within a preset time window; The vibration state is evaluated based on the root mean square value and the low and high vibration thresholds. The vibration states include: stable vibration state, critical vibration state, and high vibration state, wherein: The stable vibration state is the vibration state when the root mean square value is lower than the low vibration threshold. The critical vibration state is the vibration state when the root mean square value is between the low vibration threshold and the high vibration threshold. The high vibration state is the vibration state when the root mean square value is higher than the high vibration threshold. The allowable range of synchronization error for adjusting the multi-fan array based on the vibration state includes: When the fan system is identified as being in a stable vibration state, a preset minimum value is set for the allowable range of the synchronization error; When the fan system is identified as being in a critical vibration state, the allowable range of synchronization error is set based on the low vibration threshold and the high vibration threshold using piecewise linear interpolation. When the fan system is identified as being in a state of high vibration, a preset large value is set for the allowable range of the synchronization error; When the fan system is identified as being in a high vibration state, setting a preset large value for the allowable range of synchronization error includes: When the fan system is identified as being in a state of high vibration, a probing operational disturbance is applied to the multi-fan array; Monitor the physical vibration status information of the controller circuit board in response to the probed operational disturbance; Determine whether there is a correlation between the physical vibration state information and the probed operational disturbance; If a correlation is determined, the high vibration state is determined to be internal resonance, and the allowable range of the synchronization error is adjusted based on the internal resonance. If it is determined that there is no correlation, then the high vibration state is determined to be external background vibration, and the synchronization error is maintained within a preset range based on the external background vibration.

2. The fan speed control method for an air-cooled radiator according to claim 1, characterized in that, The determination of whether there is a correlation between the physical vibration state information and the probed operational disturbance includes: Determine the repetition period of the probe operational disturbance and the temporal characteristics of the probe operational disturbance within each period; The physical vibration status information of the controller circuit board is collected synchronously; The physical vibration state information is segmented according to the repetition period, and the vibration values ​​of each segment of the physical vibration state information are periodically superimposed to obtain the superimposed average vibration waveform. Determine whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing operational disturbance; If a vibration mode corresponding to the time-series characteristics exists, it is determined that the change in the physical vibration state information is related to the probe-based operational disturbance; if no vibration mode corresponding to the time-series characteristics exists, it is determined that the change in the physical vibration state information is not related to the probe-based operational disturbance.

3. The fan speed control method for an air-cooled radiator according to claim 2, characterized in that, The determination of whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing operational disturbance includes: Extract the characteristic parameters of the superimposed average vibration waveform; Obtain the feature parameters of the expected response mode corresponding to the temporal characteristics of the exploratory operational disturbance; Pattern matching is performed between the characteristic parameters of the superimposed average vibration waveform and the characteristic parameters of the expected response mode; Based on the pattern matching results, it is determined whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing operational disturbance.

4. The fan speed control method for an air-cooled radiator according to claim 3, characterized in that, The pattern matching of the feature parameters of the superimposed average vibration waveform with the feature parameters of the expected response mode includes: Calculate the correlation coefficient between the characteristic parameters of the superimposed average vibration waveform and the characteristic parameters of the expected response mode; The correlation coefficient is compared with a preset correlation threshold; Based on the comparison results of the correlation threshold, it is determined whether the superimposed average vibration waveform exhibits a vibration mode corresponding to the temporal characteristics of the probing operational disturbance.

5. The fan speed control method for an air-cooled radiator according to claim 3, characterized in that, The feature parameters extracted from the superimposed average vibration waveform include: Frequency domain analysis was performed on the superimposed average vibration waveform; The characteristic parameters of the superimposed average vibration waveform after frequency domain analysis are extracted, including the dominant frequency and the energy or bandwidth within a preset frequency range.

6. A fan speed control system for an air-cooled radiator, characterized in that, The system includes: The information acquisition module is used to collect vibration acceleration data from the controller circuit board of the multi-fan array; The status assessment module is used to assess the current vibration state of the fan system based on the vibration acceleration data; A range adjustment module is used to adjust the allowable range of synchronization error of the multi-fan array based on the vibration state; The rotation speed acquisition module is used to acquire the real-time rotation speed information of each fan in the multi-fan array; The instruction generation module is used to generate signal correction instructions based on the allowable range of the synchronization error and the real-time speed information of each fan; A speed adjustment module is used to adjust the speed of each fan in the multi-fan array based on the signal correction command; The vibration acceleration data collected from the controller circuit board of the multi-fan array includes: The controller circuit board acquires vibration acceleration data in three-dimensional space in real time at a preset frequency based on an accelerometer. The vibration acceleration data is digitally filtered. The assessment of the current vibration state of the fan system based on the vibration acceleration data includes: Calculate the root mean square value of the vibration acceleration data within a preset time window; The vibration state is evaluated based on the root mean square value and the low and high vibration thresholds. The vibration states include: stable vibration state, critical vibration state, and high vibration state, wherein: The stable vibration state is the vibration state when the root mean square value is lower than the low vibration threshold. The critical vibration state is the vibration state when the root mean square value is between the low vibration threshold and the high vibration threshold. The high vibration state is the vibration state when the root mean square value is higher than the high vibration threshold. The allowable range of synchronization error for adjusting the multi-fan array based on the vibration state includes: When the fan system is identified as being in a stable vibration state, a preset minimum value is set for the allowable range of the synchronization error; When the fan system is identified as being in a critical vibration state, the allowable range of synchronization error is set based on the low vibration threshold and the high vibration threshold using piecewise linear interpolation. When the fan system is identified as being in a state of high vibration, a preset large value is set for the allowable range of the synchronization error; When the fan system is identified as being in a high vibration state, setting a preset large value for the allowable range of synchronization error includes: When the fan system is identified as being in a state of high vibration, a probing operational disturbance is applied to the multi-fan array; Monitor the physical vibration status information of the controller circuit board in response to the probed operational disturbance; Determine whether there is a correlation between the physical vibration state information and the probed operational disturbance; If a correlation is determined, the high vibration state is determined to be internal resonance, and the allowable range of the synchronization error is adjusted based on the internal resonance. If it is determined that there is no correlation, then the high vibration state is determined to be external background vibration, and the synchronization error is maintained within a preset range based on the external background vibration.

Images