Method and device for controlling heat dissipation noise of sand-dust-proof heat dissipation fan in high-temperature environment

Abstract

The invention relates to a cooling noise control method and device for a sand-proof cooling fan in a high-temperature environment. The method comprises the following steps: measuring and analyzing fan blade parameters, a dustproof net cover structure and temperature-dust environment data, and establishing a high-temperature dust coupling dynamic model considering dust adhesion and temperature deformation; performing cyclostationary characteristic decomposition on the fan operation acoustic signal to obtain an environment sensitive Gaussian mixture cyclostationary noise model; calculating a cyclic spectrum correlation function for the fan noise signal and extracting a feature vector to obtain a noise source identification result under a high-temperature sand and dust working condition; and constructing a heat dissipation efficiency function and a noise level function in a temperature-dust-rotating speed three-dimensional parameter space, executing weight adaptive adjustment calculation, and outputting an optimal rotating speed control strategy. Dynamic balance between heat dissipation requirements and noise control is achieved, and the noise level is effectively reduced on the premise that the heat dissipation efficiency is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of heat dissipation noise control, and particularly to a method and device for controlling the heat dissipation noise of a dust-proof and heat-dissipating fan for high-temperature environments. Background Art

[0002] In high-temperature and dusty environments such as deserts, mining areas, and outdoor communication base stations, heat-dissipating fans not only face material property changes and structural deformations caused by rising temperatures but also need to cope with problems such as changes in mass distribution, increased friction, and decreased heat dissipation efficiency caused by the attachment and accumulation of dust particles. Heat dissipation devices in such harsh environments must ensure sufficient heat dissipation capacity and control the noise level to meet the dual requirements of equipment operation and environmental protection.

[0003] Traditional heat-dissipating fan control methods are mainly designed based on the fan characteristic curve under a single working condition and cannot effectively cope with the complex coupling effects of high-temperature and dusty environments. Existing acoustic signal demodulation and blind deconvolution techniques usually only extract periodic features for a single noise source. When dealing with composite noise signals in high-temperature and dusty environments, nonlinear filtering often destroys multiple local periods of acoustic features and cannot provide quantitative confidence for noise control. At the same time, existing technologies lack a unified modeling method for the coupled system of temperature changes, dust accumulation, and fan vibration, resulting in difficulty in accurately balancing heat dissipation efficiency and noise control when environmental conditions change. Summary of the Invention

[0004] The main object of the present invention is to provide a method and device for controlling the heat dissipation noise of a dust-proof and heat-dissipating fan for high-temperature environments. The present invention realizes the dynamic balance between heat dissipation requirements and noise control, and effectively reduces the noise level on the premise of ensuring heat dissipation efficiency.

[0005] To achieve the above object, the present invention provides a method for controlling the heat dissipation noise of a dust-proof and heat-dissipating fan for high-temperature environments, including the following steps: Measure and analyze the fan blade parameters, dust-proof mesh structure, and temperature-dust environment data, and establish a high-temperature dust coupling dynamics model considering dust attachment and temperature deformation; According to the high-temperature dust coupling dynamics model, perform cyclic stationary feature decomposition on the fan operation acoustic signal to obtain an environment-sensitive mixed Gaussian cyclic stationary noise model; Use the environment-sensitive mixed Gaussian cyclic stationary noise model to calculate the cyclic spectral correlation function of the fan noise signal and extract the feature vector to obtain the noise source identification result under high-temperature and dust conditions; Based on the noise source identification result, construct a heat dissipation efficiency function and a noise level function in the three-dimensional parameter space of temperature-dust-rotation speed, perform weight adaptive adjustment calculation, and output the optimal rotation speed control strategy.

[0006] The present invention also provides a heat dissipation noise control device for a dust-proof and heat-dissipating fan in a high-temperature environment, comprising: a measurement and analysis module for measuring and analyzing fan blade parameters, dust-proof net cover structure and temperature-dust environment data, and establishing a high-temperature dust coupling dynamics model considering dust adhesion and temperature deformation; a feature decomposition module for performing cyclic stationary feature decomposition on the acoustic signal of the fan operation according to the high-temperature dust coupling dynamics model to obtain an environment-sensitive mixed Gaussian cyclic stationary noise model; a calculation module for calculating the cyclic spectral correlation function of the fan noise signal and extracting feature vectors by using the environment-sensitive mixed Gaussian cyclic stationary noise model to obtain a noise source identification result under high-temperature dust conditions; an output module for constructing a heat dissipation efficiency function and a noise level function in a three-dimensional parameter space of temperature-dust-rotation speed based on the noise source identification result, performing weight adaptive adjustment calculation, and outputting an optimal rotation speed control strategy.

[0007] In summary, the technical solution provided by the present invention systematically integrates the coupling effects of fan blade rotation, air flow vibration, dust load and temperature change into a unified framework, realizes the accurate description of the dynamic behavior of the fan in a high-temperature dust environment, adopts an environment-sensitive mixed Gaussian cyclic stationary noise model, realizes the effective modeling of various cyclic stationary noises generated by the fan operation under high-temperature and dusty conditions, overcomes the limitations of traditional methods in dealing with composite noises, and significantly improves the accuracy of noise feature extraction. Based on cyclic spectral correlation analysis and Mahalanobis distance classification technology, accurate identification and classification of different noise sources are realized, including bearing fault noise, blade aerodynamic noise, dust particle collision noise and material deformation noise caused by high temperature, etc. A heat dissipation efficiency function and a noise level function in a three-dimensional parameter space of temperature-dust-rotation speed are constructed, and through a weight adaptive adjustment mechanism, a dynamic balance between heat dissipation requirements and noise control is realized, and the noise level is effectively reduced on the premise of ensuring heat dissipation efficiency. The influence of key fan parameters on noise control and heat dissipation efficiency is quantitatively evaluated, providing a scientific basis for system parameter optimization, reducing the noise level of the system under standard test conditions, and at the same time improving the heat dissipation efficiency. Through an adaptive control and environment response conversion decision mechanism, the adaptability and stability of the system in a high-temperature and dusty environment are enhanced, and the service life of the fan system is extended. Description of the Drawings

[0008] Figure 1 is a schematic diagram of the steps of a heat dissipation noise control method for a dust-proof and heat-dissipating fan in a high-temperature environment according to an embodiment of the present invention; Figure 2It is a structural block diagram of a heat dissipation noise control device for a dust-proof and heat-dissipating fan used in a high-temperature environment in an embodiment of the present invention.

[0009] The realization of the object, functional characteristics and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments

[0010] In order to make the object, technical solution and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0011] Referring to Figure 1 , this embodiment provides a method for controlling the heat dissipation noise of a dust-proof and heat-dissipating fan used in a high-temperature environment, including the following steps: S1, measure and analyze the fan blade parameters, the structure of the dust-proof net cover, and the temperature-dust environment data, and establish a high-temperature dust coupling dynamics model considering dust adhesion and temperature deformation; Among them, structural measurements are carried out on the number, diameter, thickness, material properties of the fan blades and the parameters of the dust-proof net cover. These structural parameters directly affect the aerodynamic performance of the fan, so they accurately reflect the design and working characteristics of the fan during the modeling process. At the same time, the operating parameters of the fan are collected, including the rated speed, actual speed, bearing type, bearing clearance value, etc. These parameters determine the rotational performance and stability of the fan. In terms of environmental factors, environmental parameter collection is carried out on the environmental temperature range, dust particle concentration, and particle size distribution, because they affect the interaction between the fan and dust during operation and the influence of temperature on the fan material. Through this step, a fan system parameter set is obtained. Based on the fan system parameter set, mathematical models are established for the blade displacement angle, angular velocity change, temperature, and dust load mass. The mechanical response of the fan blades is described by a non-linear aerodynamic restoring force function, reflecting the dynamic behavior of the fan under different working conditions, especially in high-temperature and dusty environments, the effects of blade deformation, airflow vibration, and dust adhesion. The elastic modulus of the fan material changes with temperature, so the change in the elastic modulus of the blade material at different temperatures is considered when establishing the dynamic model. This change is calculated through temperature-related material deformation characteristic data to obtain the rigidity change characteristics of the blade material at different temperatures. According to the dust particle data and temperature-related material deformation characteristic data in the fan system parameter set, the mass distribution of the system after dust adhesion is calculated to obtain a system mass matrix considering dust adhesion. Based on the temperature-related material deformation characteristic data and the non-linear aerodynamic restoring force function, the stiffness of the system is calculated. Stiffness is an important factor affecting the operating stability and anti-deformation ability of the fan, so the influence of temperature on the system stiffness is considered to obtain a temperature-related stiffness matrix. At the same time, the damping of the fan system is also affected by the coupling of temperature and dust. The friction and adhesion of dust particles increase the damping of the system, while the change in temperature causes a change in the viscosity of the material, further affecting the damping characteristics. The damping of the dust and temperature coupling is analyzed to calculate the dust-temperature coupling damping matrix. The system mass matrix considering dust adhesion, the temperature-related stiffness matrix, the dust-temperature coupling damping matrix, and the non-linear aerodynamic restoring force function are numerically solved to obtain a high-temperature dust coupling dynamic model. Using finite element analysis or other numerical calculation methods, the comprehensive influence of temperature and dust load on the fan system is considered during the solution process to obtain the motion response of the fan in a high-temperature dust environment. A high-temperature dust coupling dynamic model is obtained.

[0012] Perform piecewise linear fitting on the temperature-related material deformation characteristic data. Since the material of the fan blade exhibits different stiffness characteristics at different temperatures, the stiffness coefficients of the temperature-sensitive material are obtained through piecewise linear fitting. In this way, an accurate temperature-stiffness relationship is obtained, enabling subsequent stiffness calculations to accurately reflect the influence of temperature changes on the blade material stiffness. Based on the obtained stiffness coefficients of the temperature-sensitive material, finite element calculations are performed on the geometric deformation of the fan blade at different temperatures to obtain the temperature-related structural stiffness distribution, taking into account the geometric deformation of the blade under the influence of temperature changes. The finite element method can accurately simulate the response of the structure under different temperature conditions, calculate the stiffness changes at each node, and obtain the temperature-related structural stiffness distribution of the entire blade. At the same time, using the nonlinear aerodynamic restoring force function, linearize the interaction between the airflow and the blade to obtain the contribution matrix of aerodynamic stiffness. The aerodynamic restoring force function describes the interaction between the airflow and the blade. Especially in high-temperature and sandy dust environments, the properties of the airflow will change, which will affect the aerodynamic response of the fan blade. Therefore, by linearizing this function, simplify the complex aerodynamic action process, calculate the contribution matrix of aerodynamic stiffness, and combine it with the structural stiffness matrix for subsequent analysis. Couple the temperature-related structural stiffness distribution and the contribution matrix of aerodynamic stiffness through Gaussian integration. Gaussian integration is a numerical integration method that effectively couples different types of stiffness matrices to obtain a comprehensive temperature-related stiffness matrix. Based on the sandy dust particle data in the fan system parameters, simulate the distribution of sandy dust particles on the blade surface. After the sandy dust particles adhere to the blade surface, additional friction will be increased, which will in turn affect the damping characteristics of the fan system. By simulating the distribution of sandy dust particles, calculate its incremental contribution to the system damping to obtain the function of the influence of sandy dust load on damping. This process involves considering the particle size, concentration, and distribution law of the sandy dust particles. Through the comprehensive analysis of these factors, the influence of the sandy dust load on damping is obtained. Based on the sandy dust load damping influence function and the stiffness coefficients of the temperature-sensitive material, perform temperature correction calculations on the internal damping of the material and the contact damping of the sandy dust to obtain the sandy dust-temperature coupled damping matrix. In the actual working environment, temperature and sandy dust load are intertwined. The increase in temperature will change the internal damping characteristics of the material, and at the same time, the accumulation of sandy dust will also affect the damping. Correct the coupled effect of temperature and sandy dust, and calculate to obtain a coupled damping matrix that comprehensively considers the influence of both.

[0013] S2. According to the high-temperature sandy dust coupled dynamics model, perform cyclic stationary feature decomposition on the fan operation acoustic signal to obtain an environment-sensitive mixed Gaussian cyclic stationary noise model; Specifically, based on the high-temperature sand-dust coupling dynamics model, the acoustic signals during the operation of the fan are collected with high precision through a piezoelectric sensor array to obtain the original noise signals. These sensor arrays are distributed around the fan and can capture the complex acoustic signals generated when the fan operates in a high-temperature and sand-dust environment through a high sampling rate and a wide frequency response range. These signals contain information on various noise sources such as fan blades, bearings, airflows, and sand-dust particle collisions. The short-time Fourier transform is performed on the original noise signals to obtain the time-frequency analysis result matrix. The short-time Fourier transform is a method that converts a signal from the time domain to the frequency domain. By calculating the spectrum of the signal within different time windows, a time-frequency distribution map is generated, thereby revealing the changes in the frequency components of the signal at different time points. In a high-temperature and sand-dust environment, the noise signals of the fan are usually non-stationary. The short-time Fourier transform effectively extracts the time-frequency characteristics of these non-stationary signals, making the subsequent cyclic stationary component analysis more accurate. To improve the analysis accuracy, during the Fourier transform process, a window function (such as a Hamming window or a Gaussian window) is used to weight the signal to reduce the influence of spectral leakage. According to the time-frequency analysis result matrix, spectral peak identification and segmentation are performed on the signal energy distribution to obtain a multi-source cyclic stationary component set. By analyzing the energy distribution of different frequency components in the time-frequency matrix, the main spectral peak positions in the signal are identified. These spectral peaks correspond to different cyclic stationary noise sources. For example, bearing fault noise appears as a cyclic component at a lower frequency, while sand-dust particle collision noise generates obvious harmonic components in the higher frequency band. Through an adaptive threshold method or a clustering algorithm based on frequency distribution characteristics, the signal is decomposed into multiple cyclic stationary component sets, and each component corresponds to a specific noise source. For each component in the multi-source cyclic stationary component set, the cyclic autocorrelation function is calculated and Fourier-transformed. The cyclic autocorrelation function is used to quantify the similarity of the signal at different delays. By calculating the inner product of the noise signal and its delayed signal, the autocorrelation values at different time delays are obtained. These autocorrelation values are transformed into the frequency domain representation through Fourier transform to obtain the cyclic frequency distribution of each component. According to the temperature and sand-dust parameters in the high-temperature sand-dust coupling dynamics model, parametric modeling is performed on the amplitude and phase of each component to establish a dual modulation mapping relationship of the environmental parameters corresponding to the amplitude modulation function and the phase modulation function. The temperature, sand-dust concentration, and acoustic characteristics are correlated, enabling the noise model to dynamically respond to environmental changes. For example, through methods such as experimental data fitting or numerical simulation, the amplitude modulation function A(T,M) and the phase modulation function φ(T,M) are obtained, where A(T,M) describes the variation law of the signal amplitude under different temperature and sand-dust concentration conditions, and φ(T,M) represents the variation of the signal phase with environmental parameters. These modulation functions enable the model to automatically adjust the output of the noise model according to the real-time monitored environmental data in practical applications, realizing the dynamic prediction of noise characteristics.Based on the multi-source cyclostationary component set, the cyclic frequencies of each component, and the double modulation mapping relationship of environmental parameters, an environment-sensitive hybrid Gaussian cyclostationary noise model is created. In the model, each cyclostationary component is represented as a noise source controlled by a Gaussian distribution, and these noise sources are modulated by environmental parameters, and the overall performance is a hybrid Gaussian model. This model structure can accurately simulate various noise characteristics of the fan in a high-temperature and dust environment, and by adjusting the weights of the Gaussian components and the modulation parameters, the prediction and control of the noise characteristics under different environmental conditions can be achieved. For example, when the environmental temperature rises or the dust concentration increases, by increasing the weight of a specific Gaussian component, the model reflects the trend of the fan noise becoming larger, helping the control system to take countermeasures in advance.

[0014] Based on the high-temperature dust coupling dynamics model, simulation calculations are carried out under the combined conditions of different temperatures and dust loads to obtain the dynamic response data set of the system in the two-dimensional temperature-dust parameter space. Through multiple simulations in a wide temperature range and different dust concentration conditions, the dynamic responses of the fan system under various environmental working conditions are obtained. According to the dynamic response data set, the least squares fitting is performed on the amplitude changes of each component under different temperature and dust load conditions to obtain the non-linear response function of the amplitude to the environmental two-parameters. The least squares fitting method ensures that the response function accurately reflects the relationship between the amplitude of the noise signal and the environmental parameters by minimizing the error between the fitting curve and the actual data points. The change of the amplitude is not just linear, but a complex function form affected by multiple non-linear factors, so the non-linear fitting method is adopted. The non-linear response function obtained by fitting accurately describes the dynamic change characteristics of the noise signal amplitude under different combinations of temperature changes and dust loads. At the same time, when analyzing the dynamic response data set, the time-domain analysis of the phase information is carried out to obtain the discrete phase response matrix. The change of the phase information reflects the time-series characteristics of the noise signal under different working conditions. In the time-domain analysis process, the instantaneous phase of the noise signal is extracted through the Hilbert transform, and then the phase response at different times is sampled by the sliding window method to construct a discrete phase response matrix, showing how the phase changes with the temperature and dust load. Based on the discrete phase response matrix, the bivariate spline interpolation is performed on the variation law of the phase with temperature and dust load to obtain the phase modulation function. The spline interpolation constructs a piecewise polynomial, which ensures the fitting accuracy while maintaining the smoothness of the response curve. The bivariate spline interpolation method is suitable for dealing with the phase data change in the two-dimensional parameter space. By interpolating the discrete phase response matrix, a continuous phase modulation function is generated, so that the function can smoothly predict the phase response characteristics of the noise signal under different temperature and dust load conditions. Combining the non-linear response function of the amplitude to the environmental two-parameters with the phase modulation function, the synthesis simulation of each component is carried out to obtain the environmental parameter coupling optimization coefficient. The process of the synthesis simulation is equivalent to dynamically adjusting each noise component through the modulation function, so that these components show corresponding noise characteristics under specific environmental parameters. In this process, the calculation of the environmental parameter coupling optimization coefficient is based on the principle of minimizing the error between the fitting model and the actual noise data, and this goal is achieved through the genetic algorithm or the particle swarm optimization algorithm, so as to ensure the generalization ability and prediction accuracy of the model under different environmental conditions. After obtaining the environmental parameter coupling optimization coefficient, the three-dimensional response surfaces of temperature-dust-amplitude and temperature-dust-phase are constructed. These two three-dimensional response surfaces respectively describe the response characteristics of the noise signal amplitude and phase of the fan system under different temperature and dust conditions.Through these response surfaces, the influence of environmental changes on noise characteristics is observed. For example, when the temperature rises or the dust concentration increases, whether the noise components at specific frequencies will increase or decrease. On the basis of constructing the three-dimensional response surface, a sensitivity matrix of the mode to environmental parameters is established to obtain the double modulation mapping relationship of environmental parameters. The sensitivity matrix quantifies the influence degree of each environmental parameter on the modulation effect of the noise signal by calculating the change rate of the noise characteristics caused by different parameter changes on the response surface. This process uses the method of numerical differentiation to calculate the influence degree of each parameter on a specific position of the response surface, and organizes these sensitivity data into a matrix form. Through the analysis of the sensitivity matrix, the control strategy of the fan is optimized. For example, in a high-temperature and dusty environment, certain frequency components that contribute significantly to the noise are preferentially suppressed, or the operating parameters of the fan are adjusted under specific temperature conditions to maximize the heat dissipation efficiency and minimize the noise level.

[0015] S3. Using the environment-sensitive hybrid Gaussian cyclostationary noise model, calculate the cyclic spectral correlation function of the fan noise signal and extract the eigenvectors to obtain the noise source identification result under the high-temperature and dusty working conditions; It should be noted that based on the environment-sensitive mixed Gaussian cyclostationary noise model, the cyclic autocorrelation function of the collected fan noise signal is calculated to obtain the cyclic autocorrelation function matrix. The cyclic autocorrelation function effectively extracts the cyclostationary characteristics in the noise signal by calculating the correlation between the signal and its delayed version. The autocorrelation values at different time delays are calculated by the time-averaging method and arranged in matrix form, which can comprehensively reflect the cyclic characteristics of the signal in the time domain. The Fourier transform is performed on the cyclic autocorrelation function matrix to obtain the cyclic spectral correlation function. In the cyclic spectral correlation function, different frequency components correspond to different cyclic frequencies. For example, bearing faults are manifested as cyclic characteristics at specific low frequencies, while the collision noise of sand and dust particles appears in a higher frequency range. By performing the Fourier transform on the cyclic autocorrelation function matrix, the cyclic spectral correlation function in the frequency domain representation is obtained. Based on the cyclic spectral correlation function, a statistical test is performed on the cyclostationary characteristics of the signal to determine whether there are significant cyclic frequency components in the signal. In the statistical test, a method of comparing the test statistic with a preset threshold is used. When the test statistic exceeds the threshold, it can be determined that there are cyclic frequency components. For example, by calculating the energy density or spectral peak value of the cyclic spectral correlation function and comparing it with the noise background level, a set of cyclic frequencies is obtained, and each cyclic frequency in this set corresponds to a potential noise source. In this way, the complex acoustic signal is decomposed into multiple cyclostationary components, and each component represents a specific type of noise source. According to the obtained set of cyclic frequencies, a peak search is performed on the cyclic spectral correlation function to extract the multi-dimensional feature vector group of the signal. During the peak search process, by analyzing the peak position, peak size, and frequency bandwidth of each frequency component in the cyclic spectral correlation function, a feature vector is constructed. Each feature vector contains important characteristic information of the noise source, such as cyclic frequency, corresponding spectral frequency, spectral peak intensity, and bandwidth, etc. Using the multi-dimensional feature vector group, the Mahalanobis distance of the detected noise characteristics is calculated, and the noise characteristics are classified into categories such as bearing fault noise, blade aerodynamic noise, sand and dust particle collision noise, and material deformation noise caused by high temperature, to obtain the classification result of the noise source. The Mahalanobis distance calculation is a classification method based on statistical distribution. By calculating the distance between the sample feature vector and the mean feature vectors of each noise category and combining the covariance matrix of the feature vectors, the possibility of the sample belonging to each noise category is accurately quantified. The Mahalanobis distance can effectively eliminate the influence between different feature dimensions and provide higher accuracy for the classification of noise characteristics. In the actual calculation process, statistical models of each noise category are constructed, including the mean vector and covariance matrix, and then the Mahalanobis distance from each feature vector to the models of each category is calculated, and the sample is classified into the category with the smallest distance to achieve the accurate identification of the noise source.Based on the classification results of noise sources, calculate the feature significance index and severity evaluation value for each type of noise feature, so as to evaluate the contribution degree of each noise source to the overall noise level of the fan system. The feature significance index is calculated by the ratio of the intensity of the noise feature to the standard deviation of the background noise, while the severity evaluation value is obtained by weighted calculation of multiple factors such as feature significance, frequency bandwidth, and the occurrence frequency of the noise feature to get a comprehensive evaluation value. After completing the evaluation of feature significance and severity, establish a rotational speed-noise feature mapping table, a temperature-noise feature mapping table, and a dust load-noise feature mapping table to finally obtain the noise source identification results under high-temperature and dusty conditions. These mapping tables record the variation rules of noise features under different working conditions, enabling the system to dynamically adjust the operating parameters of the fan according to the actual operating state. For example, when the temperature rises and a specific noise source is significantly enhanced, the noise can be effectively suppressed by reducing the fan speed or adjusting the fan angle. In the case of a high dust concentration, the anti-sand mode of the fan is automatically adjusted to reduce the impact noise of dust particles on the blades, thereby achieving all-round intelligent control of the fan noise in a complex environment.

[0016] S4. Based on the noise source identification results, construct a heat dissipation efficiency function and a noise level function in the three-dimensional parameter space of temperature-dust-rotational speed, perform weight adaptive adjustment calculation, and output the optimal rotational speed control strategy.

[0017] Specifically, based on the noise source identification results and the operating parameters of the fan, a three-dimensional parameter model of the heat dissipation efficiency is established to obtain the heat dissipation efficiency function. The heat dissipation efficiency function takes into account the effects of temperature, dust concentration, and fan speed on the heat dissipation performance. By comprehensively analyzing these parameters, the heat dissipation capacity of the fan under different environmental conditions is simulated. According to the severity evaluation value and the feature mapping table in the noise source identification results, a three-dimensional parameter space model of the noise level is established to obtain the noise level function. The noise level function describes the noise level of the fan under different operating conditions by combining the fan speed, ambient temperature, and dust load. According to the severity evaluation value of the noise source, weights are assigned to each noise source to reflect the contribution of different noise sources to the overall noise level. For example, certain types of noise sources, such as the collision noise of dust particles, will have a greater impact under high dust concentration, so a higher weight will be given in the noise level function, while for noise sources with less impact, such as bearing fault noise, a lower weight will be given. Through these evaluations and mappings, the noise level function is accurately constructed. Based on the heat dissipation efficiency function and the noise level function, a multi-objective optimization model of the heat dissipation requirement and noise limit is established to obtain the control optimization objective function. The goal of the multi-objective optimization model is to adjust the fan speed so that the system can not only meet the sufficient heat dissipation requirement but also keep the noise level within the set limit range. The control optimization objective function combines the heat dissipation requirement and the noise limit in a weighted sum manner, where the weight coefficients of the heat dissipation requirement and the noise limit are dynamically adjusted according to the real-time monitored environmental changes. By optimizing the objective function, an optimal fan speed control strategy is obtained to ensure that while meeting the heat dissipation requirements, the noise level is controlled within an acceptable range. According to the real-time monitored ambient temperature and the feature significance index in the noise source identification results, the weight parameters are adaptively adjusted to obtain the adaptive weight coefficients that respond to the environment. The adaptive weight adjustment dynamically adjusts the priorities of heat dissipation efficiency and noise control according to the current environmental conditions and the actual impact of the noise source. This process enables the system to maintain the best operating state under different environmental conditions, thus achieving the balance between heat dissipation and noise control. Based on the adaptive weight coefficients and the control optimization objective function, the fan speed is optimized to obtain the optimal control speed. The solution of the optimal speed uses numerical optimization algorithms, such as the gradient descent method, genetic algorithm, or particle swarm optimization, etc. These algorithms can quickly find the optimal fan speed under the given environmental conditions. During the calculation process of the optimal control speed, the system continuously evaluates the effects of different speeds on the heat dissipation efficiency and the noise level, and finally selects a speed value that minimizes the objective function. This optimal control speed can maximize the heat dissipation effect while minimizing the noise, thus ensuring the efficient operation of the fan in the high-temperature and dusty environment. Using the optimal control speed, the fan speed is smoothly controlled to output the optimal speed control strategy. The smooth transition control is used to avoid the negative impact on the system caused by the speed change.Smoothing transition control uses a time constant to control the rate of change of the rotational speed, ensuring that the change in the fan's rotational speed is gradual and does not cause unnecessary vibration or noise. In the design of the control strategy, a maximum rotational speed change rate limit is set to prevent the system from becoming unstable due to too rapid adjustment of the rotational speed. Through these smoothing transition control strategies, based on real-time monitoring, the operating state of the fan is flexibly adjusted, thereby optimizing the heat dissipation and noise levels and ensuring the long-term stable operation of the fan in a high-temperature and dusty environment.

[0018] In one example, the fan blade parameters, dust-proof net cover structure, and temperature-dust environment data are measured and analyzed to establish a high-temperature dust-coupled dynamic model considering dust adhesion and temperature deformation, including: Structural measurements are carried out on the number, diameter, thickness, material properties of the fan blades and the parameters of the dust-proof net cover. Operating parameter measurements are made on the rated rotational speed, actual rotational speed, bearing type, and bearing clearance value of the fan. Environmental parameter collection is performed on the environmental temperature range, dust particle concentration, and particle size distribution to obtain the fan system parameter set; Based on the fan system parameter set, mathematical models are established for the blade displacement angle, angular velocity change, temperature, and dust load mass to obtain a non-linear aerodynamic restoring force function; Using the material property data in the fan system parameter set, the change in the elastic modulus of the blade material at different temperatures is calculated to obtain temperature-related material deformation characteristic data; According to the dust particle data and temperature-related material deformation characteristic data in the fan system parameter set, the mass distribution of the system after dust adhesion is calculated to obtain the system mass matrix considering dust adhesion; Based on the temperature-related material deformation characteristic data and the non-linear aerodynamic restoring force function, the temperature influence on the system stiffness is calculated to obtain the temperature-related stiffness matrix, and the dust and temperature coupling analysis of the system damping is carried out to obtain the dust-temperature coupling damping matrix; The system mass matrix considering dust adhesion, the temperature-related stiffness matrix, the dust-temperature coupling damping matrix, and the non-linear aerodynamic restoring force function are numerically solved to obtain the high-temperature dust-coupled dynamic model.

[0019] In this example, parameter measurements of the fan are carried out, including the number of fan blades, diameter, thickness, material properties, and structural measurements of the dust-proof mesh cover parameters. At the same time, the operating parameters of the fan are collected, including rated speed, actual speed, bearing type, bearing clearance value, etc. Environmental parameters are collected, including environmental temperature range, dust particle concentration, particle size distribution, etc. These measurement data constitute the parameter set of the fan system. Based on the fan system parameter set, mathematical models are established for blade displacement angle, angular velocity change, temperature, and dust load mass to obtain a non-linear aerodynamic restoring force function. The non-linear aerodynamic restoring force function is an important equation describing the mechanical response of the fan blades under the influence of air flow, dust, and temperature changes. The restoring force function By considering the blade displacement angle , angular velocity change , temperature , and dust load mass , the restoring force characteristics of the system are described. This function can reflect the deformation and recovery ability of the fan blades under different working conditions, and thus help to design more efficient heat dissipation and noise control strategies. Using the material property data in the fan system parameter set, the change in the elastic modulus of the blade material at different temperatures is calculated to obtain temperature-related material deformation characteristic data. The elastic modulus of the blade material changes with the increase in temperature, so the influence of temperature on the material deformation characteristics is considered. Assume that the relationship between the elastic modulus of the blade and temperature change is described by the following formula:

[0020] where is the elastic modulus of the material at the reference temperature , is the temperature sensitivity coefficient, is the actual temperature, is the reference temperature. Through this formula, the stiffness and damping analysis of the system are adjusted according to the material properties at different working temperatures, so that the fan can still maintain a stable operating state in a high-temperature environment. After obtaining the temperature-related material deformation characteristic data, according to the dust particle data in the fan system parameter set and these temperature-related material deformation characteristic data, the system mass distribution after dust attachment is calculated. This process considers the influence of dust particles on the mass distribution on the blade surface, because the accumulation of dust particles will change the mass distribution of the system, thereby affecting the dynamic performance of the fan. Assume that the influence of dust load mass on the mass distribution is described by the following formula:

[0021] where is the original mass of the system, is the The mass density of dust particles is It is The relative speed of dust particles, is the area of ​​contact between the dust particles and the blades, is the number of dust particle types considered. This formula is used to simulate the incremental contribution of dust attachment to the system mass. Based on the temperature-dependent material deformation characteristic data and the nonlinear aerodynamic restoring force function, the temperature effect on the stiffness of the system is calculated to obtain the temperature-dependent stiffness matrix. The stiffness matrix describes the degree of deformation of the system under the action of external forces and is directly related to the temperature. In order to calculate the effect of temperature on the stiffness of the system, it is assumed that the stiffness matrix of the system The relationship with temperature change is:

[0022] in, is the system at the reference temperature The stiffness under is the temperature sensitivity coefficient. Through this formula, the stiffness matrix of the fan system under different temperature conditions is obtained, so as to accurately predict the deformation behavior of the fan blades in a high temperature environment. In addition to the stiffness calculation, the damping of the system also needs to consider the coupling effect of dust and temperature. The adhesion of dust particles and the increase in temperature will affect the damping characteristics, so the coupling of dust and temperature is analyzed. Assume that the change of system damping is described by the following formula:

[0023] in, is the original damping of the system, is the temperature sensitivity coefficient, is the contribution coefficient of dust particles to damping, is the function of the influence of temperature on damping. Through this formula, a coupled damping matrix is ​​obtained, which takes into account the joint influence of temperature and dust load on the system damping. The system mass matrix, temperature-related stiffness matrix, dust-temperature coupled damping matrix and nonlinear aerodynamic restoring force function considering dust attachment are numerically solved to obtain a high-temperature dust coupled dynamic model. The finite element analysis method is used to simulate the dynamic behavior of the system by solving the joint equations of the stiffness matrix, mass matrix and damping matrix. For example, the system's equation of motion is written as follows:

[0024] in, is the mass matrix, is the damping matrix, is the stiffness matrix, is the external excitation force. By numerically solving this equation, the dynamic response of the fan system in a high temperature and dust environment is obtained.

[0025] In one example, based on temperature-related material deformation characteristic data and a nonlinear aerodynamic restoring force function, the temperature influence on the system stiffness is calculated to obtain a temperature-related stiffness matrix, and the dust and temperature coupling analysis of the system damping is carried out to obtain a dust-temperature coupling damping matrix, including: Perform piecewise linear fitting on the temperature-related material deformation characteristic data to obtain the stiffness coefficient of the temperature-sensitive material; Based on the stiffness coefficient of the temperature-sensitive material, perform finite element calculation on the geometric deformation of the fan blade at different temperatures to obtain the temperature-related structural stiffness distribution; Use the nonlinear aerodynamic restoring force function to linearize the interaction between the airflow and the blade to obtain the aerodynamic stiffness contribution matrix; Perform Gaussian integral coupling on the temperature-related structural stiffness distribution and the aerodynamic stiffness contribution matrix to obtain the temperature-related stiffness matrix; According to the dust particle data in the fan system parameters, simulate the distribution of dust particles on the blade surface, calculate the incremental contribution of dust particles to the system damping, and obtain the dust load damping influence function; Based on the dust load damping influence function and the stiffness coefficient of the temperature-sensitive material, perform temperature correction calculation on the internal damping of the material and the contact damping of the dust to obtain the dust-temperature coupling damping matrix.

[0026] In this example, the elastic modulus change data of the material at different temperatures are collected, and these data are obtained through experimental methods. For example, the fan blade material is placed in different temperature environments, and its stress-strain curve is measured to calculate the elastic modulus. The piecewise linear fitting method is used to process the discrete material characteristic data to ensure that the change characteristics of the material stiffness in different temperature intervals can be accurately described. The temperature interval is divided into several segments, and a linear model is constructed in each segment. The coefficients of the linear equation are determined by the least squares method so that the fitting curve is as close as possible to the experimental data. For example, for the material stiffness in the temperature interval to inside, it is expressed as:

[0027] where is the material stiffness at temperature , is the stiffness coefficient of the temperature-sensitive material, representing the linear slope of the material stiffness changing with temperature. By fitting different temperature intervals, a set of stiffness coefficients of the temperature-sensitive material set. Based on the stiffness coefficient of the temperature-sensitive material, finite element calculations are performed on the geometric deformation of the fan blade at different temperatures to obtain the temperature-dependent structural stiffness distribution. In finite element analysis, the fan blade is discretized into grids, and the stress, strain, and stiffness characteristics of the material are calculated at each grid node. By setting different temperature boundary conditions, the geometric deformation characteristics of the blade under effects such as thermal expansion and material softening are analyzed. Assume the geometric stiffness matrix of the blade is expressed as:

[0028] where, is the elastic modulus of the material, which is a function of temperature ; is the geometric deformation matrix, which reflects the deformation characteristics of the blade under the action of temperature; is the volume integral region of the blade. By the finite element method, the volume is discretized into node integrals, and the geometric stiffness of each node is solved, and then the structural stiffness distribution of the entire blade at different temperatures is obtained. This structural stiffness distribution can accurately describe the mechanical behavior of the blade in a complex thermodynamic environment, especially how the anti-deformation ability of the blade will change under high temperature and dust attachment. After obtaining the temperature-dependent structural stiffness distribution, the nonlinear aerodynamic restoring force function is used to linearize the interaction between the airflow and the blade to obtain the aerodynamic stiffness contribution matrix. The nonlinear aerodynamic restoring force function usually has a complex form, for example:

[0029] where, is the blade displacement angle, is the angular velocity change, is the temperature, is the dust load mass, , and are coefficients related to temperature and dust load. When performing linearization, a first-order linear approximation of the restoring force function at the equilibrium point is performed by the Taylor expansion method to obtain the aerodynamic stiffness contribution matrix in the form of:

[0030] This matrix describes the linear relationship between the airflow and the blade under small perturbations, and can be coupled with the structural stiffness distribution to comprehensively reflect the system stiffness characteristics under the coupling of aerodynamic and structural effects. By performing Gaussian integral coupling on the temperature-dependent structural stiffness distribution and the aerodynamic stiffness contribution matrix, the final temperature-dependent stiffness matrix is obtained. The spatially distributed stiffness characteristics are numerically solved by the Gaussian integral method to calculate the comprehensive stiffness matrix of the entire blade system. The comprehensive stiffness matrix Expressed as:

[0031] Wherein, is the integral region of the blade surface, is the weight function, representing the relative contributions of aerodynamic and structural stiffness at different positions. Through this integral formula, the combined effects of temperature and dust load on stiffness are effectively integrated, providing an accurate mathematical model for the stability analysis of the fan system. To improve the dynamic characteristics of the system, based on the dust particle data in the fan system parameters, the distribution of dust particles on the blade surface is simulated, the incremental contribution of dust particles to the system damping is calculated, and the dust load damping influence function is obtained. The distribution of dust particles on the blade surface is usually realized by the Monte Carlo simulation method, and the movement trajectories of dust particles with different sizes, masses and velocities under different wind speeds and wind directions are simulated by random sampling. The incremental contribution to damping is expressed as:

[0032] Wherein, is the mass density of the th type of dust particle, is the relative velocity of the particle, is the contact area between the particle and the blade, is the number of particle types. Combining this incremental contribution function, the influence of dust load on damping is accurately reflected in the model. Based on the dust load damping influence function and the stiffness coefficient of the temperature-sensitive material, the temperature correction calculation of the internal damping of the material and the contact damping with dust is carried out to obtain the dust-temperature coupled damping matrix. The coupled damping matrix The calculation formula is:

[0033] Wherein, is the initial damping, is the temperature influence coefficient, is the dust load damping increment, is the temperature correction function. This formula comprehensively considers the internal damping characteristics of the material and the external damping changes caused by dust contact, so that the finally obtained dust-temperature coupled damping matrix is applicable to high-temperature environments and can effectively cope with the complex effects of dust particles on the dynamic behavior of the fan system.

[0034] In an example, according to the high-temperature dust coupling dynamic model, the cyclic stationary feature decomposition of the fan operation acoustic signal is carried out to obtain the environment-sensitive mixed Gaussian cyclic stationary noise model, including: Based on the high-temperature dust coupling dynamic model, the acoustic signal is collected through a piezoelectric sensor array during the fan operation process to obtain the original noise signal; Perform short-time Fourier transform on the original noise signal to obtain the time-frequency analysis result matrix; According to the time-frequency analysis result matrix, perform spectral peak identification and segmentation on the signal energy distribution to obtain a multi-source cyclostationary component set; For each component in the multi-source cyclostationary component set, calculate the cyclic autocorrelation function and perform Fourier transform to obtain the cyclic frequency of each component; According to the temperature and dust parameters of the high-temperature dust coupling dynamics model, perform parametric modeling on the amplitude and phase of each component, and establish a dual modulation mapping relationship of environmental parameters corresponding to the amplitude modulation function and the phase modulation function; Based on the multi-source cyclostationary component set, the cyclic frequency of each component, and the dual modulation mapping relationship of environmental parameters, create an environment-sensitive hybrid Gaussian cyclostationary noise model.

[0035] In this example, during the operation of the fan, a piezoelectric sensor array is used to collect acoustic signals, obtaining the original noise signals. These piezoelectric sensor arrays are distributed at key positions of the fan, and high-precision signal acquisition is used to record the noise data of the fan in a high-temperature and dust environment. The piezoelectric sensor can convert mechanical vibration signals into electrical signals, capture the vibration characteristics of various noise sources such as fan blades, bearings, airflows, and dust particles, and obtain detailed acoustic signals at a relatively high sampling frequency. These collected original noise signals contain information on multiple complex noise sources. The short-time Fourier transform is performed on the original noise signals to obtain the time-frequency analysis result matrix. The short-time Fourier transform is a tool for analyzing the variation of the frequency components of a signal over time by segmenting the signal and applying the Fourier transform to each segment. It performs local weighting on the signal using a sliding window function and then conducts the Fourier transform to obtain the spectral information of the signal within different time windows. By performing the short-time Fourier transform on the original noise signals, the time-frequency analysis result matrix is obtained, where each column corresponds to the spectral information at a certain moment, and each row corresponds to the time variation of a frequency component. These time-frequency analysis results can reveal the noise characteristics of the fan in different time periods and frequency ranges. Especially in a high-temperature and dust environment, this method can help identify the noise components related to environmental changes. Based on the time-frequency analysis result matrix, the spectral peak identification and segmentation of the signal energy distribution are carried out to obtain a set of multi-source cyclostationary components. The signal energy distribution can help identify the main components in the signal, that is, the frequency components generated by different noise sources. Since during the operation of the fan, the signals of multiple noise sources often have obvious different characteristics in the frequency domain, these noise sources are separated through spectral peak identification. For example, the bearing fault noise shows periodic components at specific low frequencies, while the dust particle collision noise generates multiple high-frequency harmonic components. By applying the peak detection algorithm, each frequency peak is identified from the time-frequency matrix, and the signal is segmented according to this frequency information to obtain a set of multi-source cyclostationary components. Each component represents a periodic noise source. For each component in the obtained set of multi-source cyclostationary components, its cyclic autocorrelation function is calculated and Fourier transformed to obtain the cyclic frequency of each component. The cyclic autocorrelation function is an important tool for reflecting the periodic characteristics of a signal. By calculating the correlation between the signal and its delayed version, the periodic components in the signal can be identified. The autocorrelation function is defined as:

[0036] where is the time-domain representation of the signal, is the time delay, is the total duration of the signal. By calculating the cyclic autocorrelation function for each stationary component and Fourier transforming the result, the cyclic frequency of each component is extracted , i.e., the frequencies of the periodic components in the signal. The spectrum obtained through Fourier transform shows the energy distribution of each frequency component, thus helping to determine the specific frequency characteristics of each noise source. According to the temperature and dust parameters in the high-temperature dust coupling dynamics model, parametric modeling is carried out on the amplitude and phase of each component, and a double modulation mapping relationship of environmental parameters corresponding to the amplitude modulation function and the phase modulation function is established. In this model, the amplitude and phase are not only affected by the characteristics of the noise source itself, but also closely related to the environmental temperature and dust concentration. The change in temperature causes a change in the elastic modulus of the fan material, thereby changing the vibration characteristics of the blades; the change in dust concentration causes particles to collide or adhere to the blade surface, thereby affecting the interaction between the airflow and the blades. In order to link these environmental factors with the changes in the amplitude and phase of the noise signal, an amplitude modulation function and a phase modulation function are established, and parametric modeling is carried out in the following way:

[0037]

[0038] where and are the initial amplitude and phase, and are the adjustment coefficients of temperature and dust on the amplitude, and are the adjustment coefficients of temperature and dust on the phase. Through these functions, the noise characteristics of the fan are dynamically adjusted according to the actual environmental conditions. Based on the multi-source cyclic stationary component set, the cyclic frequencies of each component, and the double modulation mapping relationship of environmental parameters, an environment-sensitive hybrid Gaussian cyclic stationary noise model is created. This model uses a Gaussian distribution to simulate the frequency components of each noise source, taking into account the changes in the amplitude and phase of each noise source with environmental parameters. The contribution of each noise source is represented by a Gaussian pulse function with an environmental modulation function, as follows:

[0039] where is the amplitude of the th noise source, is the phase of the th noise source, is the width of this noise source. By superimposing the Gaussian pulse functions of multiple noise sources, an overall noise signal is generated, which comprehensively considers the effects of factors such as temperature, dust load, and fan speed, providing an accurate model for the real-time monitoring and control of noise sources.

[0040] In one example, according to the temperature and dust parameters of the high-temperature dust coupling dynamics model, parametric modeling is carried out on the amplitudes and phases of each component, and a dual-modulation mapping relationship of environmental parameters corresponding to the amplitude modulation function and the phase modulation function is established, including: Perform simulation calculations on the high-temperature dust coupling dynamics model under different combinations of temperature and dust load to obtain a dynamic response data set of the system in the two-dimensional parameter space of temperature-dust; According to the dynamic response data set, perform least squares fitting on the amplitude changes of each component under different temperature and dust load conditions to obtain a non-linear response function of the amplitude to the dual environmental parameters; Perform time-domain analysis on the phase information in the dynamic response data set to obtain a discrete phase response matrix; Based on the discrete phase response matrix, perform bivariate spline interpolation on the variation law of the phase with temperature and dust load to obtain a phase modulation function; Use the non-linear response function of the amplitude to the dual environmental parameters and the phase modulation function to synthesize and simulate each component to obtain an environmental parameter coupling optimization coefficient; According to the environmental parameter coupling optimization coefficient, construct three-dimensional response surfaces of temperature-dust-amplitude and temperature-dust-phase, establish a sensitivity matrix of the mode to environmental parameters, and obtain a dual-modulation mapping relationship of environmental parameters.

[0041] In this example, the fan system is modeled through numerical simulation. The high-temperature dust coupling dynamics model considers multiple factors, such as temperature, dust load, geometric characteristics of the fan blades, material characteristics, and the interaction between the airflow and the blades, etc. During the simulation process, different temperature and dust load conditions are set, and the system is calculated step by step to simulate the dynamic response of the fan under various working conditions. These working conditions are achieved by varying the temperature and dust concentration

[0042]

[0043] wherein, represents the amplitude under different temperatures and dust concentrations , is the initial amplitude, and are coefficients related to temperature and dust concentration respectively, and are the reference temperature and dust concentration, is an adjustment factor related to environmental conditions, and are the fitting points during the fitting process, and are the standard deviations of temperature and dust concentration. Through the least squares fitting method, optimize the parameters , , etc., to obtain the amplitude response function that best fits the simulation data. After obtaining the nonlinear response function of the amplitude, perform time-domain analysis on the phase information in the dynamic response dataset to obtain the discrete phase response matrix. The change of phase information reflects the periodic characteristics of the noise source over time. By analyzing the phase of each response component in the dataset, extract the time-varying characteristics of the signal phase under different temperature and dust load conditions. The phase response matrix is expressed as: ; where, represents the change of phase with time , is the initial phase, is the adjustment coefficient related to environmental conditions, is the frequency, is the phase shift, and are the standard deviations of temperature and dust concentration, and are the fitting points of temperature and dust concentration. This formula reveals the influence of environmental temperature and dust concentration on the phase of the noise signal. Based on the discrete phase response matrix, perform bivariate spline interpolation to obtain the variation law of phase with temperature and dust load. The spline interpolation method can perform smooth fitting on discrete data points and can better capture complex nonlinear relationships. By performing bivariate spline interpolation on the phase response matrix, construct the phase modulation function , and its form is:

[0044] where, is the interpolation coefficient, is the spline basis function, through temperature and dust load Interpolation on the phase response matrix to generate a smooth phase modulation curve. Through this interpolation method, the phase response under any temperature and dust load is obtained. Using the obtained amplitude and the non-linear response functions of the environmental dual parameters and the phase modulation function, the components are synthesized and simulated to obtain the environmental parameter coupling optimization coefficient. The optimization coefficient is calculated based on the amplitude and phase response relationship, aiming to maximize the influence of the changes in environmental temperature and dust concentration on the amplitude and phase changes of the noise signal. During the calculation process, the amplitude and phase are comprehensively calculated with the corresponding environmental parameters to obtain the optimization response coefficient for each noise source , which is expressed in the form of:

[0045] where, represents the environmental parameter coupling optimization coefficient, is the amplitude modulation function, is the phase modulation function. By combining the influence of environmental factors on the amplitude and phase, the optimal response of each noise source is calculated. After obtaining the environmental parameter coupling optimization coefficient, three-dimensional response surfaces of temperature-dust-amplitude and temperature-dust-phase are constructed. These two response surfaces can intuitively show the variation laws of the amplitude and phase of the noise signal under different temperature and dust load conditions. Through these response surfaces, the influence of environmental changes on the fan noise can be better understood. The amplitude response surface and the phase response surface are established using interpolation algorithms or fitting methods to connect the amplitude and phase data points under different temperature and dust conditions into continuous three-dimensional surfaces, which can effectively show the variation trends of the noise signal under various working conditions. By constructing these three-dimensional response surfaces, a sensitivity matrix of the mode to environmental parameters is established to obtain the dual modulation mapping relationship of environmental parameters. The sensitivity matrix reveals the variation laws of each noise source under different environmental conditions by calculating the influence degree of the change of each environmental parameter (temperature, dust load) on the amplitude and phase changes

[0046] In an example, using the environment-sensitive hybrid Gaussian cyclostationary noise model, the cyclic spectral correlation function is calculated for the fan noise signal and the eigenvectors are extracted to obtain the noise source identification results under high-temperature and dust conditions, including: Based on the environment-sensitive hybrid Gaussian cyclostationary noise model, the cyclic autocorrelation function is calculated for the collected fan noise signal to obtain the cyclic autocorrelation function matrix; The Fourier transform is performed on the cyclic autocorrelation function matrix to obtain the cyclic spectral correlation function; Based on the cyclic spectral correlation function, a statistical test is performed on the cyclic stationary characteristics of the signal to obtain the set of cyclic frequencies; According to the set of cyclic frequencies, peak search is performed on the cyclic spectral correlation function to obtain a multi-dimensional eigenvector group; Using a multi-dimensional feature vector group, calculate the Mahalanobis distance for the detected noise features, classify the noise features into categories such as bearing fault noise, blade aerodynamic noise, dust particle collision noise, and material deformation noise caused by high temperature, and obtain the noise source classification result; Based on the noise source classification result, calculate the feature significance index and severity evaluation value for each type of noise feature, and establish a rotational speed-noise feature mapping table, a temperature-noise feature mapping table, and a dust load-noise feature mapping table to obtain the noise source identification result under high-temperature and dusty conditions.

[0047] In this example, the fan noise signal includes noises from multiple sources, and these noises often exhibit obvious cyclostationary characteristics. By using a piezoelectric sensor array, collect the noise signal of the fan during operation and convert it into an electrical signal. Apply the cyclic autocorrelation function to the collected noise signal to calculate a matrix containing the cyclic features of the signal. The cyclic autocorrelation function is the correlation function between the signal and its delayed version, which can reveal the periodicity and phase characteristics of the signal. Assume the noise signal is , and its cyclic autocorrelation function is expressed as:

[0048] where is the total duration of the signal, and is the time delay. By calculating this autocorrelation function, the similarity of the signal at different time delays is obtained, thereby determining its periodic characteristics. This matrix reflects the cyclostationarity of the fan noise signal. Based on the cyclic autocorrelation function matrix, perform a Fourier transform to obtain the cyclic spectral correlation function. The Fourier transform can convert the time-domain signal into a frequency-domain representation, thereby revealing the various frequency components in the signal and their variations at different times. By performing a Fourier transform on the autocorrelation matrix, the frequency distribution of the signal is obtained, and further analyze the fluctuations of each frequency component over time. For the signal , its cyclic autocorrelation function is expressed as

[0049] where is the total duration of the signal, and is the time delay. By calculating the autocorrelation function, the similarity of the signal at different time delays is obtained, thereby determining its periodic characteristics. This matrix reflects the cyclostationarity of the fan noise signal. Based on the cyclic autocorrelation function matrix, a Fourier transform is performed to obtain the cyclic spectral correlation function. The Fourier transform can convert the time-domain signal into a frequency-domain representation, thereby revealing the various frequency components in the signal and their variations at different times. By performing a Fourier transform on the autocorrelation matrix, the frequency distribution of the signal is obtained, and further analysis is carried out on the fluctuations of each frequency component over time. For the signal , its Fourier transform is expressed as:

[0050] Through the Fourier transform, the time-frequency characteristics of the signal are revealed, and the obtained cyclic spectral correlation function can reflect the intensity distribution of the frequency components. Based on the cyclic spectral correlation function, a statistical test is performed on the cyclostationary characteristics of the signal to obtain the set of cyclic frequencies. The statistical test helps to judge whether the signal has obvious periodic characteristics and to determine which frequency components are significant according to the frequency distribution. Common methods include significance tests based on spectral peak analysis or energy analysis of periodic components. By analyzing the cyclic spectral correlation function, a set of significant cyclic frequency sets are identified, and these frequency components represent the main noise sources in the signal. In this way, the multi-source noise signal is decomposed into a series of cyclostationary components of specific frequencies. Based on the obtained set of cyclic frequencies, a peak search is performed on the cyclic spectral correlation function to obtain a multi-dimensional feature vector group. The purpose of the peak search is to identify the peak positions corresponding to different noise sources from the cyclic spectral correlation function. By finding the local maxima in the spectrum, the main frequency components in the signal are determined. Assume that there are several local peaks in the cyclic spectral correlation function , then the frequency components corresponding to these peaks are expressed as , and each peak represents an independent noise source. By analyzing these peaks, a multi-dimensional feature vector group is obtained, and each feature vector contains information such as the frequency, amplitude, and bandwidth of the peak, which helps to classify and identify the noise sources. Using the multi-dimensional feature vector group, the Mahalanobis distance is calculated for the detected noise features, and the noise features are classified into bearing fault noise, blade aerodynamic noise, dust particle collision noise, and material deformation noise caused by high temperature. The Mahalanobis distance is a metric method based on a statistical model that takes into account the correlation between various features, so it has high accuracy in the noise classification task. Assume that the noise feature vector belongs to the category , then the Mahalanobis distance is calculated by the following formula:

[0051] where is the mean vector of the category , and is the covariance matrix of the category . is the feature vector to be classified. By calculating the Mahalanobis distance from each noise feature vector to the means of different categories, the noise features are classified into the closest noise categories, thus obtaining the noise source classification result. For example, the noise source of bearing faults has specific frequency characteristics and a small bandwidth, while the sand particle collision noise shows a wider frequency spectrum range. After completing the noise source classification, calculate the feature significance index and severity evaluation value for each type of noise feature according to the classification result. It is comprehensively evaluated from multiple aspects such as the intensity, frequency bandwidth, and occurrence frequency of the noise features. For example, the feature significance is represented by the ratio of the amplitude of the noise feature to the background noise, while the severity evaluation value is comprehensively scored by combining the intensity and influence range of the noise. The feature significance index is expressed by the following formula:

[0052] where is the noise amplitude of the category , and is the standard deviation of the background noise. The severity evaluation value is calculated by combining multiple factors such as frequency bandwidth and occurrence frequency:

[0053] where is the frequency bandwidth of the category , is the frequency of occurrence, is the weight coefficient. By calculating these indicators for each noise source, evaluate the degree of influence of the noise source on the fan performance. Based on the noise source classification result, feature significance index, and severity evaluation value, establish a rotational speed-noise feature mapping table, temperature-noise feature mapping table, and dust load-noise feature mapping table to obtain the noise source identification result under high-temperature and dusty conditions. These mapping tables relate the relationship between environmental conditions (such as temperature and dust concentration) and noise features, and can adjust the fan's operating parameters in real time according to environmental changes to reduce noise and maintain the best heat dissipation effect. For example, when the temperature is too high or the dust load is large, automatically adjust the rotational speed of the fan, giving priority to suppressing the noise sources with greater influence on performance to ensure the stable operation of the fan in a high-temperature and dusty environment.

[0054] In an example, based on the noise source identification result, construct a heat dissipation efficiency function and a noise level function in the three-dimensional parameter space of temperature-dust-rotational speed, perform weight adaptive adjustment calculation, and output the optimal rotational speed control strategy, including: Based on the noise source identification result and the fan operating parameters, perform three-dimensional parameter modeling on the heat dissipation efficiency to obtain the heat dissipation efficiency function; Based on the severity evaluation value in the noise source identification result and the feature mapping table, a three-dimensional parameter space model of the noise level is established to obtain the noise level function; Based on the heat dissipation efficiency function and the noise level function, a multi-objective optimization model of the heat dissipation requirement and the noise limit is established to obtain the control optimization objective function; According to the ambient temperature monitored in real time and the feature significance index in the noise source identification result, the weight parameter is adaptively adjusted to obtain the adaptive weight coefficient of the environmental response; Based on the adaptive weight coefficient of the environmental response and the control optimization objective function, the optimal solution of the fan speed is obtained to get the optimal control speed; Using the optimal control speed, a smooth transition control of the fan speed is carried out to output the optimal speed control strategy.

[0055] In this example, the noise source identification results of the fan under different working conditions are collected, and these results include the types, intensities, and frequency characteristics of different noise sources. On this basis, combined with the working parameters of the fan, such as speed, ambient temperature, dust concentration, etc., a model of the heat dissipation efficiency is established. The heat dissipation efficiency is not only related to the fan speed but also affected by the ambient temperature and dust concentration. Therefore, the heat dissipation efficiency is expressed as a function of temperature, dust concentration, and speed, that is:

[0056] Among them, is the heat dissipation efficiency function, is the initial heat dissipation efficiency under the reference conditions, are the influence coefficients of temperature, dust concentration, and speed on the heat dissipation efficiency respectively, are the ambient temperature, dust concentration, and speed respectively, are the reference temperature, dust concentration, and speed. Through the least squares method or other regression analysis methods, these coefficients are fitted based on experimental data or simulation results, and then the heat dissipation efficiency function is obtained. Based on the severity evaluation value in the noise source identification result and the feature mapping table, a three-dimensional parameter space model of the noise level is established to obtain the noise level function. The noise level depends not only on the fan speed and ambient temperature but also on the dust concentration and the working state of the fan. Therefore, the noise level function is expressed as a multivariate function of temperature, dust concentration, and speed, in the form of:

[0057] Among them, is the noise level function, is the initial noise level under the reference conditions, They are the influence coefficients of temperature, dust concentration, and rotational speed on the noise level, respectively. By conducting experiments or simulation calculations on the fan under different temperature, dust concentration, and rotational speed conditions, these coefficients are obtained, and then the noise level function is obtained. The noise level is relatively independent of the heat dissipation efficiency, but they jointly affect the overall performance of the fan. Therefore, it is necessary to combine the two for optimization. Based on the obtained heat dissipation efficiency function and noise level function, a multi-objective optimization model is established to obtain the control optimization objective function. The objective function is expressed as:

[0058] Where, is the control optimization objective function, is the heat dissipation capacity, is the target heat dissipation capacity, is the noise level, is the maximum allowable noise level, and are the weight coefficients of heat dissipation requirements and noise limits. The optimization objective is to make the heat dissipation capacity as close as possible to the target value while keeping the noise level not exceeding the maximum limit. The weight coefficients and are adjusted according to the specific application scenario and automatically regulated through experimental data or optimization algorithms. According to the ambient temperature monitored in real time and the feature significance index in the noise source identification result, the weight parameters are adaptively adjusted to obtain the adaptive weight coefficients for environmental response. Since the change of environmental conditions will affect the relationship between heat dissipation and noise, the weight coefficients in the optimization objective are adjusted according to the real-time data. Assume that the feature significance index is calculated in real time according to the current ambient temperature and dust concentration, and the adaptive weight coefficients and are adjusted by the following formula:

[0059]

[0060] Where, and are the initial weight coefficients, and are the adjustment coefficients, is the feature significance index, which reflects the influence degree of the noise source on the fan performance. By calculating these coefficients in real time and dynamically adjusting the optimization objective function according to the change of the environment, the fan system can balance heat dissipation and noise under different environmental conditions. Based on the adaptive weight coefficients and the control optimization objective function, the rotational speed of the fan is optimized to obtain the optimal control rotational speed. The optimization is solved through numerical optimization methods such as the gradient descent method and the particle swarm optimization method. Assume that the objective of the optimization problem is to make the control optimization objective function If it is minimized, the optimal control speed is solved by the following method :

[0061] This optimization process will minimize the noise level while the fan speed meets the heat dissipation requirements. By adjusting the speed , find the optimal balance point under different environmental conditions, thereby improving the overall performance of the fan system. Using the obtained optimal control speed , perform smooth transition control on the fan speed to avoid excessive vibration or noise during the adjustment process. Smooth transition control is achieved by introducing a time constant or buffer, that is, gradually adjusting the fan speed over a period of time until the optimal control speed is reached. This smooth transition can reduce sudden noise and ensure the stability of the system, avoiding mechanical stress or system instability caused by rapid changes in speed. The mathematical form of smooth transition control is expressed as:

[0062] where is the time constant of smooth transition, is the change of fan speed with time. In this way, smooth transition of the fan speed is achieved, ensuring stable operation of the system when the optimal control speed is reached.

[0063] Referring to Figure 2 , this embodiment provides a heat dissipation noise control device for a dust-proof and heat-dissipating fan in a high-temperature environment, including: Measurement and analysis module 1, used to measure and analyze fan blade parameters, dust-proof net cover structure and temperature-dust environment data, and establish a high-temperature dust coupling dynamics model considering dust adhesion and temperature deformation; Feature decomposition module 2, used to perform cyclic stationary feature decomposition on the fan operation acoustic signal according to the high-temperature dust coupling dynamics model to obtain an environment-sensitive mixed Gaussian cyclic stationary noise model; Calculation module 3, used to calculate the cyclic spectral correlation function of the fan noise signal and extract the eigenvector using the environment-sensitive mixed Gaussian cyclic stationary noise model to obtain the noise source identification result under high-temperature dust conditions; Output module 4, used to construct a heat dissipation efficiency function and a noise level function in the three-dimensional parameter space of temperature-dust-speed based on the noise source identification result, perform weighted adaptive adjustment calculation, and output the optimal speed control strategy.

[0064] In this embodiment, for the specific implementation of each unit in the above device embodiment, please refer to that described in the above method embodiment, and will not be elaborated here.

[0065] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, device, article or method. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, device, article or method comprising such an element.

[0066] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.

Claims

1. A method for controlling heat dissipation noise of a dust-proof heat dissipation fan in a high temperature environment, characterized in that: The following steps are involved: The fan blade parameters, dust cover structure and temperature-dust environment data were measured and analyzed, and a high-temperature dust coupling dynamic model considering dust adhesion and temperature deformation was established; According to the high-temperature dust coupling dynamics model, the fan operation acoustic signal is subjected to cyclostationary feature decomposition to obtain an environment-sensitive mixed Gaussian cyclostationary noise model; Using the environment-sensitive mixed Gaussian cyclostationary noise model, the cyclic spectrum correlation function of the fan noise signal is calculated and the characteristic vector is extracted to obtain the noise source identification result under high temperature and dust conditions; Based on the noise source identification result, a heat dissipation efficiency function and a noise level function in the temperature-dust-speed three-dimensional parameter space are constructed, and a weight adaptive adjustment calculation is performed to output an optimal speed control strategy.

2. The method for controlling heat dissipation noise of a dust-proof heat dissipation fan for high temperature environment according to claim 1, characterized in that: The fan blade parameters, dust cover structure and temperature-dust environment data are measured and analyzed to establish a high-temperature dust coupling dynamic model that takes into account dust adhesion and temperature deformation, including: The fan system parameter set was obtained by measuring the number, diameter, thickness, material properties and dust cover parameters of the fan blades, measuring the rated speed, actual speed, bearing type and bearing clearance of the fan, and collecting environmental parameters such as the ambient temperature range, dust particle concentration and particle size distribution. Based on the fan system parameter set, mathematical modeling of blade displacement angle, angular velocity change, temperature and dust load mass is performed to obtain a nonlinear aerodynamic restoring force function; Using the material property data in the fan system parameter set, the elastic modulus change of the blade material at different temperatures is calculated to obtain temperature-related material deformation property data; Calculating the mass distribution of the system after dust attachment based on the dust particle data in the fan system parameter set and the temperature-related material deformation characteristic data to obtain a system mass matrix considering dust attachment; Based on the temperature-related material deformation characteristic data and the nonlinear aerodynamic restoring force function, the temperature influence calculation is performed on the system stiffness to obtain the temperature-related stiffness matrix, and the dust and temperature coupling analysis is performed on the system damping to obtain the dust-temperature coupling damping matrix; The system mass matrix considering dust attachment, the temperature-related stiffness matrix, the dust-temperature coupling damping matrix and the nonlinear aerodynamic restoring force function are numerically solved to obtain a high-temperature dust coupling dynamics model.

3. The method for controlling heat dissipation noise of a dust-proof heat dissipation fan for high temperature environment according to claim 2, characterized in that: The temperature-related material deformation characteristic data and the nonlinear aerodynamic restoring force function are used to calculate the temperature influence on the system stiffness to obtain a temperature-related stiffness matrix, and the dust and temperature coupling analysis is performed on the system damping to obtain a dust-temperature coupling damping matrix, including: Performing piecewise linear fitting on the temperature-related material deformation characteristic data to obtain a stiffness coefficient of the temperature-sensitive material; Based on the stiffness coefficient of the temperature-sensitive material, finite element calculation is performed on the geometric deformation of the fan blade at different temperatures to obtain the temperature-dependent structural stiffness distribution; Using the nonlinear aerodynamic restoring force function, linearizing the interaction between the airflow and the blade to obtain an aerodynamic stiffness contribution matrix; Performing Gaussian integral coupling on the temperature-dependent structural stiffness distribution and the aerodynamic stiffness contribution matrix to obtain a temperature-dependent stiffness matrix; According to the dust particle data in the fan system parameter set, the distribution of dust particles on the blade surface is simulated, the incremental contribution of dust particles to the system damping is calculated, and the dust load damping influence function is obtained; Based on the dust load damping influence function and the temperature sensitive material stiffness coefficient, temperature correction calculation is performed on the material internal damping and the dust contact damping to obtain a dust-temperature coupled damping matrix.

4. The method for controlling heat dissipation noise of a dust-proof heat dissipation fan for high temperature environment according to claim 1, characterized in that: According to the high-temperature dust coupling dynamics model, the fan operation acoustic signal is subjected to cyclostationary feature decomposition to obtain an environment-sensitive mixed Gaussian cyclostationary noise model, including: Based on the high-temperature dust coupling dynamics model, acoustic signals are collected during the operation of the fan through a piezoelectric sensor array to obtain an original noise signal; Performing short-time Fourier transform on the original noise signal to obtain a time-frequency analysis result matrix; According to the time-frequency analysis result matrix, the signal energy distribution is subjected to spectral peak identification and segmentation to obtain a set of multi-source cyclostationary components; For each component in the multi-source cyclostationary component set, calculating and Fourier transforming the cyclic autocorrelation function to obtain the cyclic frequency of each component; According to the temperature and dust parameters of the high-temperature dust coupling dynamic model, the amplitude and phase of each component are parameterized and modeled, and a dual modulation mapping relationship of environmental parameters corresponding to the amplitude modulation function and the phase modulation function is established; An environment-sensitive mixed Gaussian cyclostationary noise model is created based on the multi-source cyclostationary component set, the cyclic frequency of each component and the dual modulation mapping relationship of the environmental parameters.

5. The method for controlling heat dissipation noise of a dust-proof heat dissipation fan for high temperature environment according to claim 4, characterized in that: The parameterized modeling of the amplitude and phase of each component is performed according to the temperature and dust parameters of the high-temperature dust coupling dynamic model, and the dual modulation mapping relationship of the environmental parameters corresponding to the amplitude modulation function and the phase modulation function is established, including: The high temperature dust coupling dynamics model is simulated and calculated under different temperature and dust load combination conditions to obtain a dynamic response data set of the system in the temperature-dust two-dimensional parameter space; According to the dynamic response data set, the amplitude changes of each component under different temperature and dust load conditions are fitted by least squares to obtain a nonlinear response function of the amplitude to the environmental dual parameters; Performing time domain analysis on the phase information in the dynamic response data set to obtain a discrete phase response matrix; Based on the discrete phase response matrix, bivariate spline interpolation is performed on the variation of phase with temperature and dust load to obtain a phase modulation function; Using the nonlinear response function of the amplitude to the dual environmental parameters and the phase modulation function, each component is synthetically simulated to obtain the environmental parameter coupling optimization coefficient; According to the environmental parameter coupling optimization coefficient, three-dimensional response surfaces of temperature-dust-amplitude and temperature-dust-phase are constructed, a sensitivity matrix of the modal to the environmental parameters is established, and a dual modulation mapping relationship of the environmental parameters is obtained.

6. The method for controlling heat dissipation noise of a dust-proof heat dissipation fan for high temperature environment according to claim 1, characterized in that: The method uses the environment-sensitive mixed Gaussian cyclostationary noise model to calculate the cyclic spectrum correlation function of the fan noise signal and extract the characteristic vector to obtain the noise source identification result under the high temperature and dust working condition, including: Based on the environment-sensitive mixed Gaussian cyclostationary noise model, a cyclic autocorrelation function is calculated on the collected fan noise signal to obtain a cyclic autocorrelation function matrix; Performing Fourier transform on the cyclic autocorrelation function matrix to obtain a cyclic spectral correlation function; Based on the cyclic spectrum correlation function, a statistical test is performed on the cyclic stationary characteristics of the signal to obtain a set of cyclic frequencies; According to the cyclic frequency set, a peak search is performed on the cyclic spectrum correlation function to obtain a multi-dimensional feature vector group; Using the multidimensional feature vector group, the detected noise features are calculated by Mahalanobis distance, and the noise features are classified into the categories of bearing fault noise, blade aerodynamic noise, sand and dust particle collision noise, and material deformation noise caused by high temperature, so as to obtain the noise source classification result; Based on the noise source classification results, the characteristic significance index and severity evaluation value are calculated for each type of noise feature, and a speed-noise feature mapping table, a temperature-noise feature mapping table and a dust load-noise feature mapping table are established to obtain the noise source identification results under high temperature and dust conditions.

7. The method for controlling heat dissipation noise of a dust-proof heat dissipation fan for high temperature environment according to claim 6, characterized in that: Based on the noise source identification result, a heat dissipation efficiency function and a noise level function in the temperature-dust-speed three-dimensional parameter space are constructed, a weight adaptive adjustment calculation is performed, and an optimal speed control strategy is output, including: Based on the noise source identification result and the fan operating parameters, three-dimensional parameter modeling is performed on the heat dissipation efficiency to obtain a heat dissipation efficiency function; According to the severity evaluation value and the feature mapping table in the noise source identification result, a three-dimensional parameter space model is performed on the noise level to obtain a noise level function; Based on the heat dissipation efficiency function and the noise level function, multi-objective optimization modeling is performed on heat dissipation requirements and noise restrictions to obtain a control optimization objective function; Adaptively adjusting the weight parameter according to the real-time monitored ambient temperature and the characteristic significance index in the noise source identification result to obtain an adaptive weight coefficient of the environmental response; Based on the adaptive weight coefficient of the environmental response and the control optimization objective function, the fan speed is optimized to obtain the optimal control speed; The optimal control speed is used to smoothly control the fan speed and output an optimal speed control strategy.

8. A device for controlling heat dissipation noise of a dust-proof heat dissipation fan used in a high temperature environment, characterized in that: The method according to any one of claims 1 to 7 is used to implement the steps of the method, wherein the anti-sand and dust cooling fan heat dissipation noise control device for high temperature environment comprises: The measurement and analysis module is used to measure and analyze the fan blade parameters, dust cover structure and temperature-dust environment data, and establish a high-temperature dust coupling dynamic model that takes into account dust adhesion and temperature deformation; A feature decomposition module is used to perform cyclostationary feature decomposition on the fan operation acoustic signal according to the high-temperature dust coupling dynamics model to obtain an environment-sensitive mixed Gaussian cyclostationary noise model; A calculation module, used to calculate the cyclic spectrum correlation function of the fan noise signal and extract the characteristic vector by using the environment-sensitive mixed Gaussian cyclostationary noise model, so as to obtain the noise source identification result under the high temperature and dust working condition; The output module is used to construct a heat dissipation efficiency function and a noise level function in the temperature-dust-speed three-dimensional parameter space based on the noise source identification result, perform weight adaptive adjustment calculation, and output an optimal speed control strategy.

Images