A compressor modulation characteristic quantitative analysis method, device, equipment and medium

Abstract

This application relates to the field of fault diagnosis technology, and discloses a method, apparatus, device, and medium for quantitative analysis of compressor modulation characteristics. The method includes: acquiring a signal to be processed from a compressor operating under steady-state conditions; performing structured frequency band decomposition on the signal to obtain several sub-band signals; performing Hilbert envelope analysis on each of the sub-band signals to obtain an envelope signal corresponding to each sub-band signal; performing spectral analysis on the envelope signals to obtain modulation characteristic parameters of each sub-band signal; determining multi-dimensional modulation indices based on the modulation characteristic parameters; comparing the modulation indices with a preset dynamic threshold; and comprehensively judging and classifying the fault state of the compressor based on the comparison results. The technical solution provided by this application can achieve the technical effects of accurate extraction, stable quantification, and unified judgment of modulation characteristics and fault state.

Description

Technical Field

[0001] This application relates to the field of fault diagnosis technology, and in particular to a method, apparatus, equipment and medium for quantitative analysis of compressor modulation characteristics. Background Technology

[0002] The off-line inspection of electric compressors requires matching the severity of faults through amplitude modulation / frequency modulation characteristics. The core is to separate the independent carrier modulation characteristics of different fault sources from the mixed signals of multi-source vibrations. However, the cross-modulation of broadband carriers from multiple vibration sources such as bearings and valves inside the compressor causes the signal to exhibit a multi-carrier nested coupling form, breaking the single-carrier applicability premise of traditional modulation analysis. Under the real-time constraint of ≤10 seconds / unit on the production line, traditional methods are prone to problems such as modulation sideband aliasing, poor noise resistance, and insufficient accuracy. Even if the sideband amplitude is extracted, it is impossible to form a unified quality grading engineering criterion because the energy background of different carriers is greatly different, the amplitude is affected by carrier fluctuations / window functions / spectral leakage interference, and the dimensions are inconsistent.

[0003] Therefore, overcoming the multi-carrier frequency coupling and strong background noise interference of compressors, and achieving accurate extraction, stable quantization, and unified judgment of modulation characteristics are technical problems that urgently need to be solved. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, and medium for quantitative analysis of compressor modulation characteristics, which achieves the technical effects of accurate extraction, stable quantification, and unified judgment of fault status of modulation characteristics.

[0005] To achieve the above objectives, the main technical solutions adopted in this application include: In a first aspect, embodiments of this application provide a method for quantitative analysis of compressor modulation characteristics, the method comprising: The signal to be processed under the steady-state operation of the compressor is acquired, and the signal to be processed is decomposed into a structured frequency band to obtain several sub-band signals. Perform Hilbert envelope analysis on each of the sub-band signals to obtain the envelope signal corresponding to each sub-band signal; Spectral analysis is performed on the envelope signal to obtain the modulation characteristic parameters of each sub-band signal; Based on the modulation feature parameters, multi-dimensional modulation indices are determined, and the modulation indices are compared with preset dynamic thresholds to comprehensively determine and classify the fault status of the compressor based on the comparison results.

[0006] In one embodiment, the step of acquiring the signal to be processed under steady-state operation of the compressor, and performing structured frequency band decomposition on the signal to be processed to obtain several sub-band signals, including: Steady-state monitoring is performed on the raw vibration signal collected during compressor operation. When the steady-state determination condition is met, the raw vibration signal is truncated to obtain the signal to be processed. The signal to be processed is subjected to structured frequency band decomposition to obtain multiple initial sub-band signals; For each of the initial sub-band signals, a single-component determination is performed. If the determination is that the approximate single-component condition is not met, the current initial sub-band signal is split into two new sub-band signals, and the single-component determination is recursively performed on the split sub-band signals until all sub-band signals meet the approximate single-component condition. All sub-band signals that satisfy the approximate single-component condition are summarized to obtain the plurality of sub-band signals.

[0007] In one embodiment, performing Hilbert envelope analysis on each of the plurality of sub-band signals to obtain an envelope signal corresponding to each sub-band signal includes: The beginning and end of each sub-band signal are mirrored and extended to generate extended sub-band signals. The extended subband signal is subjected to Hilbert transform to construct an analytic signal, and the magnitude of the analytic signal is determined to obtain the initial envelope signal; The original envelope signal is obtained by removing the extended segment corresponding to the mirror extension process from the initial envelope signal; The original envelope signal is filtered to obtain the envelope signal corresponding to each of the sub-band signals.

[0008] In one embodiment, performing spectral analysis on the envelope signal to obtain modulation characteristic parameters of each sub-band signal includes: Determine the envelope amplitude spectrum corresponding to each envelope signal; The DC component amplitude at zero frequency is extracted from the envelope amplitude spectrum, and the DC component amplitude is determined as the carrier amplitude. The envelope amplitude spectrum is searched based on the preset target fault characteristic frequency peak value, and the maximum amplitude found is determined as the modulation component amplitude. The ratio of the modulation component amplitude to the carrier amplitude is determined, and the ratio is defined as the modulation characteristic parameter.

[0009] In one implementation, determining the multi-dimensional modulation index based on the modulation feature parameters includes: The modulation characteristic parameters of each sub-band signal are traversed, and the maximum value among the modulation characteristic parameters of each sub-band signal is determined as the first dimension modulation index. Traverse the envelope amplitude spectrum corresponding to each envelope signal, extract the maximum amplitude at the first-order frequency conversion as the frequency conversion reference amplitude, and determine the ratio of the first-dimensional modulation index to the frequency conversion reference amplitude as the second-dimensional modulation index. For a target subband signal containing a target fault characteristic frequency peak, the amplitude at the target fault characteristic frequency peak in the envelope amplitude spectrum of the target subband signal is determined, and the ratio of the amplitude to the sum of the amplitudes of all frequency points in the envelope amplitude spectrum of the target subband signal is determined as the third-dimensional modulation index. The first-dimensional modulation index, the second-dimensional modulation index, and the third-dimensional modulation index are summarized to obtain the multi-dimensional modulation index.

[0010] In one embodiment, comparing the modulation index with a preset dynamic threshold to comprehensively determine and classify the fault state of the compressor based on the comparison result includes: If the third-dimensional modulation index is greater than or equal to the preset concentration threshold, the compressor is determined to be in a serious fault state. If the third dimension modulation index is less than the concentration threshold, then a first ratio of the first dimension modulation index to the preset first dynamic threshold is determined, and a second ratio of the second dimension modulation index to the preset second dynamic threshold is determined. The maximum value between the first ratio and the second ratio is used as a comprehensive severity index, and the failure status of the compressor is classified according to the comprehensive severity index.

[0011] In one implementation, the grading includes: When the comprehensive severity index is less than the first grading threshold, the compressor is determined to be in a normal fault state. When the comprehensive severity index is greater than or equal to the first grading threshold and less than the second grading threshold, the compressor is determined to be in a minor fault state. When the comprehensive severity index is greater than or equal to the second grading threshold and less than the third grading threshold, the compressor is determined to be in a medium fault condition. When the comprehensive severity index is greater than or equal to the third grading threshold, the compressor is determined to be in a serious fault state. Wherein, the first grading threshold < the second grading threshold < the third grading threshold.

[0012] Secondly, embodiments of this application provide a quantitative analysis device for compressor modulation characteristics, the device comprising: The frequency band decomposition unit is used to acquire the signal to be processed under the steady-state operation of the compressor, and to perform structured frequency band decomposition on the signal to be processed to obtain several sub-band signals. The envelope analysis unit is used to perform Hilbert envelope analysis on each of the plurality of sub-band signals to obtain the envelope signal corresponding to each of the sub-band signals. The spectrum analysis unit is used to perform spectrum analysis on the envelope signal to obtain the modulation characteristic parameters of each sub-band signal; The comprehensive judgment unit is used to determine multi-dimensional modulation indicators based on the modulation feature parameters, compare the modulation indicators with preset dynamic thresholds, and make a comprehensive judgment and classification of the fault status of the compressor based on the comparison results.

[0013] Thirdly, embodiments of this application provide a computer device, including: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes these computer instructions to perform the aforementioned quantitative analysis method for compressor modulation characteristics.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer instructions, which are used to cause a computer to execute the above-described method for quantitative analysis of compressor modulation characteristics.

[0015] The technical solutions provided by one or more embodiments of this application avoid non-stationary noise interference by limiting the compressor to a steady-state operating condition, and decouple the broadband complex vibration signal into several sub-band signals using structured frequency band decomposition. This effectively reduces the masking of weak fault characteristics by background noise, achieves physical isolation of composite fault components in different frequency bands, and provides a high signal-to-noise ratio data foundation for subsequent analysis. Furthermore, Hilbert envelope analysis is performed on each sub-band signal, which can strip away the high-frequency carrier and accurately extract the low-frequency modulation envelope, avoiding spurious modulation and endpoint effects caused by multi-frequency coupling, and truly restoring the periodic impact pattern caused by mechanical faults, significantly improving... This approach enhances the visibility of subtle fault characteristics. Furthermore, spectral analysis of the envelope signal transforms time-domain impulses into discrete frequency-domain spectra to accurately locate fault characteristic frequencies. Modulation depth parameters are calculated by extracting the DC component and peak energy of the fault frequency, effectively eliminating cross-band energy background interference and standardizing characteristics to obtain highly quantifiable core data with significant physical meaning. Finally, based on these parameters, a comprehensive evaluation index system covering multiple dimensions such as absolute modulation intensity, relative significance, and energy concentration is constructed. This system is then compared and judged using thresholds to achieve scientific classification and comprehensive diagnosis of compressor fault states, significantly improving the robustness, accuracy, and engineering practical value of fault diagnosis. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a method for quantitative analysis of compressor modulation characteristics provided in this application embodiment; Figure 2 A flowchart of step S1 provided in the embodiments of this application; Figure 3 A flowchart of step S3 provided in the embodiments of this application; Figure 4 A flowchart of step S5 provided in an embodiment of this application; Figure 5 A flowchart for determining multi-dimensional modulation indices provided for embodiments of this application; Figure 6 A flowchart for comprehensive judgment and classification is provided for embodiments of this application; Figure 7 A block diagram of a compressor modulation characteristic quantitative analysis device provided in this application embodiment; Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] As a core functional component of the air conditioning system, the electric compressor's offline testing and operational status monitoring aim to quickly identify typical faults such as bearing damage, valve leakage, and rotor-stationary rubbing, which directly affect the vehicle's sound quality, system reliability, and service life. The essence of these faults lies in the introduction of periodic impacts, gap changes, or nonlinear stiffness disturbances during compressor operation. These disturbances do not exist directly as independent components in the vibration signal but rather modulate the carrier signal originally generated by mechanical structure vibration in the form of amplitude modulation (AM) and frequency modulation (FM). Therefore, the accurate extraction and quantification of modulation characteristics are the core basis for fault identification and quality grading. The underlying logic of this technical approach is to establish a deterministic correspondence between "fault disturbance → modulation characteristic change → fault severity".

[0020] To achieve the aforementioned correspondence, the core prerequisite is to separate the independent carrier modulation characteristics corresponding to different fault sources from the mixed signal of multi-source vibrations and eliminate the interference of irrelevant factors on the modulation depth. However, the vibration excitation structure of the compressor itself naturally breaks this prerequisite: it contains multiple independent vibration sources such as bearings, valves, motors, and electromagnetic excitations. The characteristic carrier frequencies carried by each vibration source cover a wide frequency range from low to high frequencies, and there is a cross-modulation effect between different carriers. This makes the modulation structure of the vibration signal not a simple single-carrier single-modulation model, but a complex form of multi-carrier nested coupling. This directly breaks the default premise of single-carrier applicability of traditional modulation analysis methods.

[0021] When using traditional methods such as Hilbert envelope analysis and short-time Fourier transform (STFT) modulation bispectral analysis to process these multi-carrier coupled signals, the problems are amplified: First, the modulation sidebands generated by different carriers will overlap in the frequency domain. For example, the modulation sideband of the high-frequency fault carrier in the bearing overlaps with the modulation sideband of the low-frequency carrier in the valve in the envelope spectrum, making it impossible to distinguish the fault source corresponding to the sideband, and naturally, it is impossible to separate the independent modulation depth of different faults. Second, the strong background noise in the production line scenario will further submerge the weak real modulation sidebands, and traditional broadband envelope analysis has poor robustness to noise, which will significantly increase the false negative rate. In addition, the real-time constraint of ≤10 seconds for single-unit detection in the production line means that high-precision complex time-frequency algorithms cannot meet the production cycle requirements, and only fast algorithms with insufficient precision can be used, which further aggravates the distortion of feature extraction.

[0022] Even if the amplitude of the modulation sideband is barely extracted through filtering and other methods, there are still deeper quantization obstacles: the vibration energy background of different carrier frequencies can vary by several times or even tens of times. The sideband amplitude is not only related to the actual modulation depth, but is also affected by factors such as the amplitude fluctuation of the carrier itself, the selection of window function parameters, and spectral leakage. This results in a lack of stable mapping relationship between the modulation depth and the analysis results. Furthermore, the modulation characteristics of different frequency bands cannot be compared across frequency bands due to the inconsistency of dimensions. Ultimately, it is difficult to translate the modulation characteristics into a unified criterion for compressor quality classification that can be used in engineering.

[0023] In summary, overcoming the interference of multi-carrier frequency coupling and strong background noise in compressors, and achieving accurate extraction, stable quantization, and unified judgment of modulation features are the technical problems that urgently need to be solved.

[0024] According to an embodiment of this application, a method for quantitative analysis of compressor modulation characteristics is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0025] This embodiment provides a method for quantitative analysis of compressor modulation characteristics. Figure 1 A flowchart of a method for quantitative analysis of compressor modulation characteristics provided in this application embodiment is shown below. Figure 1 As shown, the process includes the following steps: Step S1: Obtain the signal to be processed under the steady-state operation of the compressor, and perform structured frequency band decomposition on the signal to be processed to obtain several sub-band signals.

[0026] Specifically, the key premise of steady-state operation is first defined to avoid non-stationary noise interference generated by the compressor during start-up, shutdown, or transient load changes, ensuring that the acquired signals can accurately and stably reflect the current mechanical health status of the equipment. Based on this, structured frequency band decomposition technology breaks through the limitations of traditional full-band mixed analysis, finely decoupling the broadband complex vibration signal into several sub-band signals. This process not only effectively reduces the masking effect of background noise on weak fault characteristics but also physically isolates composite fault components that may be mixed in different frequency bands. This provides high signal-to-noise ratio data input for subsequent high-precision envelope analysis, modulation feature extraction, and multi-dimensional fault severity determination for specific sub-bands, fundamentally guaranteeing the sensitivity and accuracy of the entire diagnostic system.

[0027] Step S3: Perform Hilbert envelope analysis on each of the several sub-band signals to obtain the envelope signal corresponding to each sub-band signal.

[0028] Specifically, Hilbert envelope analysis is performed on each of the obtained sub-band signals, effectively stripping away the high-frequency carrier signal and accurately extracting the low-frequency modulation envelope hidden within. Since each sub-band signal has been preprocessed into an approximately single-component state, the application of the Hilbert transform on this basis can minimize spurious modulation and endpoint effects caused by multi-frequency coupling, thus realistically reproducing the periodic impact patterns caused by mechanical faults such as bearing damage and valve leakage. This process transforms complex vibrations in the time domain into pure envelope signals characterizing the intensity of fault impacts, greatly improving the visibility of weak fault features and laying a solid and distortion-free data foundation for subsequent accurate quantification of modulation depth and calculation of multi-dimensional modulation indices in the frequency domain.

[0029] Step S5: Perform spectrum analysis on the envelope signal to obtain the modulation characteristic parameters of each sub-band signal.

[0030] Specifically, by performing spectral analysis on the envelope signals corresponding to each sub-band signal, periodic impulses that are difficult to quantify directly in the time domain can be transformed into clearly discernible discrete spectral lines in the frequency domain, thereby accurately locating the target fault characteristic frequency and its energy distribution. Based on this, by extracting the DC component from the envelope spectrum as the carrier energy reference and combining it with the peak energy at a specific fault frequency, modulation characteristic parameters characterizing the modulation depth are calculated. These parameters not only effectively eliminate energy background interference caused by differences in structural transmission paths between different frequency bands, achieving standardization of cross-band characteristics, but also provide highly physically meaningful and quantifiable core data support for subsequently constructing a multi-dimensional modulation index system and accurately assessing the severity of compressor faults.

[0031] Step S7: Determine multi-dimensional modulation indicators based on modulation feature parameters, compare the modulation indicators with preset dynamic thresholds, and make a comprehensive judgment and classification of the compressor's fault status based on the comparison results.

[0032] Specifically, the extracted discrete modulation feature parameters are globally integrated to construct a comprehensive evaluation index system covering multiple dimensions such as absolute modulation intensity, relative significance, and energy concentration, thereby comprehensively characterizing the evolution of complex internal compressor faults. Based on this, an adaptive dynamic threshold comparison mechanism is introduced to effectively overcome the shortcomings of traditional fixed thresholds, which are susceptible to environmental noise and fluctuations in operating parameters under complex conditions. Finally, based on the comprehensive comparison results of multi-dimensional indicators and dynamic thresholds, the compressor fault state is scientifically classified and determined. This logic not only significantly improves the robustness and accuracy of fault diagnosis but also provides an intuitive, reliable, and practically valuable decision-making basis for production line quality inspection and predictive maintenance of equipment.

[0033] This embodiment provides a quantitative analysis method for compressor modulation characteristics. By limiting the compressor to steady-state operation to avoid non-stationary noise interference, and employing structured frequency band decomposition to decouple broadband complex vibration signals into several sub-band signals, it effectively reduces the masking of weak fault characteristics by background noise and achieves effective isolation of composite fault components in different frequency bands, providing a high signal-to-noise ratio data foundation for subsequent analysis. Furthermore, Hilbert envelope analysis is performed on each sub-band signal, which can strip away high-frequency carriers and accurately extract low-frequency modulation envelopes, avoiding spurious modulation and endpoint effects caused by multi-frequency coupling, and truly restoring the periodic impact characteristics caused by mechanical faults, significantly… This approach enhances the visibility of subtle fault characteristics. Furthermore, spectral analysis of the envelope signal transforms time-domain impulses into discrete frequency-domain spectra to accurately pinpoint fault characteristic frequencies. Modulation depth parameters are calculated by extracting the DC component and peak energy of the fault frequency, effectively eliminating cross-band energy background interference and standardizing characteristics to obtain highly quantifiable core data with significant physical meaning. Finally, based on these parameters, a comprehensive evaluation index system is constructed, encompassing multiple dimensions such as absolute modulation intensity, relative significance, and energy concentration. This system is then compared and judged using thresholds to achieve scientific classification and comprehensive diagnosis of compressor fault states, significantly improving the robustness, accuracy, and engineering practical value of fault diagnosis.

[0034] Figure 2 The flowchart for step S1 provided in the embodiments of this application may include the following steps: Step S11: Perform steady-state monitoring on the original vibration signal collected during compressor operation. When the steady-state determination condition is met, the original vibration signal is truncated to obtain the signal to be processed.

[0035] Specifically, during compressor startup or under variable load conditions, the compressor's speed and load are dynamically changing, and the extracted modulation features lack a stable reference. Therefore, this embodiment first performs steady-state monitoring on the acquired raw vibration signal x(t). Specifically, the first-order rotational frequency f1(t) of the compressor is tracked in real time using a short-time Fourier transform (STFT). The steady-state determination condition is set as follows: the first-order rotational frequency f1(t) remains stable at the frequency f corresponding to the target speed. target Within a preset tolerance range (e.g., ±3%), and the duration of this state exceeds a preset duration (e.g., 1 second). When the above conditions are met, the compressor is determined to have entered a steady-state operating condition, and the original vibration signal segment after the determination time for a preset duration (e.g., 5 seconds) is automatically extracted as the signal to be processed, x. proc (t). This step ensures that the data foundation for subsequent analysis is under stable physical conditions.

[0036] Step S13: Perform structured frequency band decomposition on the signal to be processed to obtain multiple initial sub-band signals.

[0037] Specifically, regarding the intercepted signal x to be processed proc (t) In this embodiment, a structured frequency band decomposition strategy based on constant relative bandwidth is adopted to cover a wide frequency range of the compressor, from low-frequency mechanical vibration to high-frequency bearing impact. Specifically, the signal is bandpass filtered and extracted according to the 1 / 1 octave rule to obtain multiple initial sub-band signals x. i (t).

[0038] In this embodiment, the center frequency f of each initial sub-band c (n) increases in a fixed octave ratio (i.e., 2 times), and adjacent sub-bands do not overlap. The specific center frequencies of each sub-band are set as follows: 33Hz, 66Hz, 132Hz, 264Hz, 528Hz, 1056Hz, 2112Hz, 4224Hz, 8448Hz and 16896Hz.

[0039] To ensure the consistency of the physical meaning of signal characteristics within each frequency band, each sub-band is set to a constant relative bandwidth Br = 70.7%. Based on this relative bandwidth, the upper cutoff frequency f of the nth sub-band is... high (n) and lower cutoff frequency f low (n) is calculated using the following formulas: For example, when the center frequency f c When (n) = 1056Hz, the corresponding lower cutoff frequency is 746.7Hz, and the upper cutoff frequency is 1493.4Hz. Based on the above rules, the initial sub-band signal x of the corresponding frequency band is extracted sequentially. i (t), thus initially decoupling the complex vibration signal with a wide bandwidth into multiple narrowband signals with constant relative bandwidth.

[0040] Step S15: Perform single-component determination on each initial sub-band signal. If the determination is that the approximate single-component condition is not met, split the current initial sub-band signal into two new sub-band signals, and recursively perform single-component determination on the split sub-band signals until all sub-band signals meet the approximate single-component condition.

[0041] Specifically, due to the presence of multiple excitation sources within the compressor, some broadband initial sub-band signals may contain multiple modulation sources, failing to satisfy the single-component assumption required for the Hilbert transform. Therefore, this embodiment applies the following to each initial sub-band signal x i (t) performs single-component determination.

[0042] The core algorithm for single-component determination is based on the statistical characteristics of instantaneous frequency. The specific calculation process is as follows: 1. For the current initial sub-band signal x i (t) Perform Hilbert transform to construct the analytic signal z. i (t): in, t is the Hilbert transform operator; j is the imaginary unit; t is time.

[0043] 2. Extract the instantaneous phase of the analytical signal. : The instantaneous frequency f is obtained by differentiation. inst (t): 3. Calculate the standard deviation σ of the instantaneous frequency of the sub-band signal throughout the entire time window. f,i : Where T is the total time window length of the current sub-band signal; This is the average instantaneous frequency of the sub-band signal within the time window T.

[0044] The judgment criterion is set as follows: if the instantaneous frequency standard deviation σ f,i Less than or equal to the center frequency f of the sub-band c,i The product of σ and a preset proportional threshold α (e.g., α = 0.01) satisfies: f,i ≤α×f c,i If so, then the current sub-band signal is determined to satisfy the approximate single-component condition.

[0045] If the above conditions are not met (i.e., σ) f,i >α×f c,i This indicates that multiple frequency components are still coupled within the sub-band, triggering an adaptive splitting mechanism: the current sub-band signal is split according to the center frequency f. c,i / 2 1 / 4 and f c,i ·2 1 / 4 The signal is split into two new sub-band signals with their relative bandwidth halved. Then, the single-component determination process described above is recursively performed on the two new sub-band signals until all the split sub-band signals satisfy the approximate single-component condition.

[0046] Step S17: Summarize all sub-band signals that satisfy the approximate single-component condition to obtain several sub-band signals.

[0047] Specifically, after recursive decomposition, all initial sub-bands that do not meet the conditions are refined into narrowband signals that satisfy the single-component assumption. All sub-band signals that satisfy the approximate single-component condition are then aggregated to obtain the final set of several sub-band signals {x}.sub,k (t)}. Each sub-band signal in this set represents an approximate single-carrier modulated signal of an independent vibration source (such as a specific bearing or valve) inside the compressor, completely eliminating the multi-carrier coupling effect and laying a solid data foundation for accurately extracting the modulation depth in subsequent steps.

[0048] This embodiment performs steady-state monitoring of the raw vibration signal collected during compressor operation. When the first-order rotational frequency remains stable within a preset tolerance range corresponding to the target speed for a duration exceeding a preset time, the raw vibration signal is intercepted to obtain the signal to be processed. This ensures that the data basis for subsequent analysis is under stable physical conditions, avoiding the problem of lack of a stable reference standard for modulation characteristics caused by dynamic changes in speed and load during startup or variable load conditions. Based on this, the signal to be processed is bandpass filtered and extracted according to a structured frequency band decomposition strategy with constant relative bandwidth. This yields multiple initial sub-band signals with constant relative bandwidth and non-overlapping adjacent sub-bands, thus initially decoupling the complex broadband vibration signal into multiple narrowband signals, covering... This method covers a wide frequency range of the compressor, from low-frequency mechanical vibration to high-frequency bearing impact, and ensures that the physical meaning of the signal characteristics within each frequency band is consistent. Furthermore, for each initial sub-band signal, a single-component determination based on instantaneous frequency statistical characteristics is performed. Sub-band signals that do not meet the approximate single-component condition are adaptively split into two new sub-band signals with half the relative bandwidth, and the single-component determination is recursively performed until all sub-band signals meet the approximate single-component condition. Finally, all sub-band signals that meet the approximate single-component condition are summarized to obtain several sub-band signals, ensuring that each sub-band signal represents an approximate single-carrier modulation signal from an independent vibration source inside the compressor. This completely eliminates the multi-carrier coupling effect and lays a solid data foundation for the subsequent accurate extraction of modulation depth.

[0049] Figure 3 The flowchart for step S3 provided in the embodiments of this application may include the following steps: Step S31: Perform mirror extension processing on the beginning and end of each sub-band signal to generate extended sub-band signals.

[0050] Specifically, when performing Hilbert transform on sub-band signals, due to the signal truncation effect, spurious boundary fluctuations (i.e., boundary effects) are easily generated at both ends, leading to envelope extraction distortion. To solve this problem, this embodiment performs Hilbert transform on each sub-band signal x. sub,kThe first and last ends of (t) are mirror-extended. Specifically, using the data at the beginning of the sub-band signal as the axis of symmetry, the first half of the signal is mirror-flipped and spliced ​​to the beginning of the signal; simultaneously, using the data at the end of the sub-band signal as the axis of symmetry, the second half of the signal is mirror-flipped and spliced ​​to the end of the signal. Let the length of the original sub-band signal be N, and the length of the mirror extension be set to L (e.g., L=N / 10), then the generated extended sub-band signal x ext,k The total length of (t) becomes N+2L. This step effectively suppresses boundary effects in subsequent transformation processes.

[0051] Step S33: Perform Hilbert transform on the extended subband signal to construct an analytic signal, determine the magnitude of the analytic signal, and obtain the initial envelope signal.

[0052] Specifically, for the extended subband signal x ext,k (t) is subjected to Hilbert transform to construct an analytic signal. The calculation formula is: Subsequently, the magnitude of the analytic signal is determined, and the initial envelope signal E is obtained. init,k (t). Its calculation formula is: Step S35: Remove the extended segments in the initial envelope signal that correspond to the mirror extension processing to obtain the original envelope signal.

[0053] Specifically, due to the obtained initial envelope signal E init,k (t) includes an artificially added extension portion with a length of N+2L. To ensure the accuracy of subsequent feature extraction, this embodiment directly removes the extension segments with lengths L at the beginning and end of the initial envelope signal, retaining only the effective segment with a length of N in the middle, thus obtaining the original envelope signal E with the same length as the original subband signal. raw,k (t).

[0054] Step S37: Filter the original envelope signal to obtain the envelope signal corresponding to each sub-band signal.

[0055] Specifically, affected by background noise and high-frequency impact spikes, the original envelope signal E raw,k (t) often contains a large amount of high-frequency noise, which cannot truly reflect the low-frequency modulation trend. Therefore, this embodiment performs low-pass filtering on the original envelope signal. A cutoff frequency of f is used. cut Butterworth low-pass filter for E raw,k (t) is used for smoothing filtering. Cutoff frequency f cutThe setting is based on 1.5 to 2 times the compressor's highest possible fault characteristic frequency (e.g., set to 500Hz). After filtering, high-frequency impact glitches are removed, resulting in a smooth and true envelope signal E corresponding to each sub-band signal. k (t). This envelope signal reflects the amplitude modulation (AM) characteristics caused by mechanical faults within this frequency band.

[0056] This embodiment generates extended subband signals by performing mirror extension processing on both ends of each subband signal. This effectively suppresses false boundary fluctuations at both ends caused by signal truncation during the subsequent Hilbert transform, thus avoiding envelope extraction distortion. Based on this, a Hilbert transform is performed on the extended subband signal to construct an analytic signal, and the magnitude of the analytic signal is determined to obtain the initial envelope signal. Then, the extended segment corresponding to the mirror extension processing in the initial envelope signal is removed to obtain the original envelope signal of the same length as the original subband signal, thereby ensuring the accuracy of subsequent feature extraction. Finally, the original envelope signal is low-pass filtered to remove background noise and high-frequency impact spikes, resulting in a smooth and realistic envelope signal corresponding to each subband signal. This envelope signal can centrally reflect the amplitude modulation characteristics caused by mechanical faults within the frequency band.

[0057] Figure 4 The flowchart for step S5 provided in the embodiments of this application may include the following steps: Step S51: Determine the envelope amplitude spectrum corresponding to each envelope signal.

[0058] Specifically, in order to convert the time-domain envelope signal into frequency-domain features, this embodiment calculates the envelope signal E corresponding to each sub-band signal. k (t) is subjected to Fast Fourier Transform (FFT) to obtain the envelope amplitude spectrum E. k (f).

[0059] Step S53: Extract the DC component amplitude at zero frequency from the envelope amplitude spectrum and determine the DC component amplitude as the carrier amplitude.

[0060] Specifically, in the envelope amplitude spectrum, the DC component at zero frequency (i.e., f=0Hz) represents the average energy level of the sub-band signal, which physically corresponds to the original carrier energy unaffected by fault modulation. Therefore, this embodiment directly uses the envelope amplitude spectrum E k (f) Extract the amplitude at 0Hz and determine it as the carrier amplitude A0.

[0061] Step S55: Search the envelope amplitude spectrum based on the preset target fault characteristic frequency peak value, and determine the maximum amplitude found as the modulation component amplitude.

[0062] Specifically, different types of compressor faults (such as bearing damage, valve leakage, etc.) will produce periodic impact modulations at specific fault characteristic frequencies. This embodiment pre-sets target fault characteristic frequencies f corresponding to various typical faults. target (For example, the bearing outer ring fault frequency BPFO, the inner ring fault frequency BPFI, etc.). In the envelope amplitude spectrum E k In (f), the target fault characteristic frequency f target Centered on a preset search bandwidth Δf (e.g., ±2Hz), a peak search is performed. The maximum amplitude within this search interval is extracted and determined as the modulation component amplitude A. m This amplitude accurately reflects the modulation energy intensity caused by the target fault.

[0063] Step S57: Determine the ratio of the modulation component amplitude to the carrier amplitude, and use the ratio as a modulation characteristic parameter.

[0064] Specifically, in order to eliminate interference caused by differences in carrier energy background between different subbands and to achieve standardized comparison of cross-band characteristics, this embodiment calculates the modulation component amplitude A. m The ratio of the amplitude to the carrier amplitude A0 is defined as the modulation characteristic parameter (i.e., modulation depth m). Its calculation formula is: m = A0 / A0. m / A0, where the modulation depth m is a dimensionless relative value, can stably and accurately quantify the modulation severity of the target fault within the sub-band, providing a reliable data foundation for subsequent multi-dimensional index calculations.

[0065] This embodiment transforms the time-domain envelope signal into frequency-domain features by determining the envelope amplitude spectrum corresponding to each envelope signal. Based on this, the DC component amplitude at zero frequency is extracted from the envelope amplitude spectrum and determined as the carrier amplitude to characterize the original carrier energy unaffected by fault modulation. Then, the envelope amplitude spectrum is searched based on the preset target fault characteristic frequency peak, and the maximum amplitude found is determined as the modulation component amplitude to truly reflect the modulation energy intensity caused by the target fault. Finally, the ratio of the modulation component amplitude to the carrier amplitude is determined and determined as the modulation characteristic parameter, thereby eliminating interference caused by the difference in carrier energy background between different subbands, realizing standardized comparison of cross-band features, and stably and accurately quantifying the modulation severity of the target fault within the subband, providing a reliable data foundation for subsequent multi-dimensional index calculations.

[0066] Figure 5 The flowchart for determining multi-dimensional modulation indices provided in the embodiments of this application may include the following steps: Step S711: Traverse the modulation characteristic parameters of each sub-band signal, and determine the maximum value among the modulation characteristic parameters of each sub-band signal as the first dimension modulation index.

[0067] Specifically, in order to comprehensively assess the most severe modulation level among all potential fault sources within the compressor, this embodiment traverses the modulation characteristic parameters (i.e., modulation depth m) of all sub-band signals. k The maximum values ​​from these modulation depths are extracted and determined as the first-dimensional modulation index M1. Its calculation formula is: M1 = max(m1, m2, ..., m...). K ), where M1 is the first-dimensional modulation index; m k Let be the modulation depth of the k-th subband signal; K represents the total number of subband signals. This first-dimensional modulation index reflects the absolute modulation intensity of the most severe fault inside the compressor.

[0068] Step S713: Traverse the envelope amplitude spectrum corresponding to each envelope signal, extract the maximum amplitude at the first-order frequency conversion as the frequency conversion reference amplitude, and determine the ratio of the first-dimensional modulation index to the frequency conversion reference amplitude as the second-dimensional modulation index.

[0069] Specifically, in order to eliminate the influence of compressor operating speed fluctuations on modulation depth, this embodiment traverses the envelope amplitude spectrum E corresponding to each sub-band signal. k (f) Extract the first-order rotational frequency (i.e., the base frequency f corresponding to the current compressor speed). rot The maximum amplitude at point () is determined as the frequency reference amplitude A. rot Subsequently, the first-dimensional modulation index M1 and the frequency conversion reference amplitude A are compared. rot The ratio is calculated to determine the second-dimensional modulation index, M2. Its calculation formula is: M2 = M1 / A rot Where M2 is the second-dimensional modulation index; A rot This is the reference amplitude for the frequency conversion. This second-dimensional modulation index characterizes the relative significance of the fault modulation intensity relative to the compressor operating baseline energy.

[0070] Step S715: For the target sub-band signal containing the target fault characteristic frequency peak, determine the amplitude at the target fault characteristic frequency peak in the envelope amplitude spectrum of the target sub-band signal, and determine the ratio of the amplitude to the sum of the amplitudes of all frequency points in the envelope amplitude spectrum of the target sub-band signal as the third-dimensional modulation index.

[0071] Specifically, in order to assess the energy concentration of the target fault features within a specific frequency band and eliminate interference from background broadband noise, this embodiment extracts the amplitude A at the target fault feature frequency peak in the envelope amplitude spectrum of the target sub-band signal containing the target fault feature frequency peak. target Subsequently, the sum of the amplitudes at all frequency points in the envelope amplitude spectrum of the target sub-band signal is calculated (i.e., the total energy E of the sub-band). totalThe ratio of the target amplitude to the total energy is defined as the third-dimensional modulation index M3. Its calculation formula is as follows: Where M3 is the third-dimensional modulation index; A target The amplitude at the peak value of the target fault characteristic frequency; This is the sum of the amplitudes at all frequency points in the target subband envelope amplitude spectrum. This third-dimensional modulation index characterizes the concentration of fault energy within this frequency band; a larger value indicates a more prominent fault characteristic.

[0072] Step S717: Summarize the first-dimensional modulation index, the second-dimensional modulation index, and the third-dimensional modulation index to obtain a multi-dimensional modulation index.

[0073] Specifically, the first-dimensional modulation index M1, the second-dimensional modulation index M2, and the third-dimensional modulation index M3 obtained above are summarized and constructed into a three-dimensional feature vector M=[M1,M2,M3], which serves as the input basis for subsequent comparison with preset dynamic thresholds and for comprehensive judgment and classification of compressor fault status.

[0074] This embodiment uses the modulation characteristic parameters of each sub-band signal as the first-dimensional modulation index to reflect the absolute modulation intensity of the most severe fault inside the compressor. Based on this, the maximum amplitude at the first-order rotation frequency is extracted by traversing the envelope amplitude spectrum corresponding to each envelope signal as the rotation frequency reference amplitude. The ratio of the first-dimensional modulation index to the rotation frequency reference amplitude is determined as the second-dimensional modulation index, thereby eliminating the influence of compressor operating speed fluctuations on the modulation depth and characterizing the relative significance of the fault modulation intensity relative to the compressor operating baseline energy. Furthermore, for the target sub-band signal containing the target fault characteristic frequency peak, the ratio of the amplitude at the target fault characteristic frequency peak to the sum of the amplitudes of all frequency points in the envelope amplitude spectrum of the target sub-band signal is determined as the third-dimensional modulation index, thereby assessing the energy concentration of the target fault characteristic in a specific frequency band and eliminating the interference of background broadband noise. Finally, the first-dimensional, second-dimensional, and third-dimensional modulation indices are summarized to obtain a multi-dimensional modulation index, providing input for subsequent comparison with a preset dynamic threshold to comprehensively determine and classify the compressor fault state.

[0075] Figure 6 The flowchart for comprehensive judgment and classification provided in this application embodiment may include the following steps: Step S731: If the third-dimensional modulation index is greater than or equal to the preset concentration threshold, the compressor is determined to be in a serious fault state.

[0076] Specifically, since the third-dimensional modulation index M3 characterizes the concentration of fault energy in the spectrum, if this index is too high, it indicates that the fault characteristics are extremely significant, and no further complex comprehensive evaluation is needed. Therefore, this embodiment first compares the third-dimensional modulation index M3 with a preset concentration threshold T3. If M3 ≥ T3, the compressor's fault state is directly determined to be a severe fault, and the current evaluation process ends.

[0077] Step S733: If the third dimension modulation index is less than the concentration threshold, then determine the first ratio of the first dimension modulation index to the preset first dynamic threshold, and determine the second ratio of the second dimension modulation index to the preset second dynamic threshold.

[0078] Specifically, if the third-dimensional modulation index M3 is less than the concentration threshold T3, it indicates that the fault energy is not yet highly concentrated, and a comprehensive evaluation combining the absolute modulation intensity and the operating baseline is required. In this case, the first ratio R1 between the first-dimensional modulation index M1 and the preset first dynamic threshold T1, and the second ratio R2 between the second-dimensional modulation index M2 and the preset second dynamic threshold T2 are determined. The calculation formulas are: R1 = M1 / T1, R2 = M2 / T2, where R1 represents the degree of deviation of the first-dimensional modulation index from its normal baseline; and R2 represents the degree of deviation of the second-dimensional modulation index from its normal baseline.

[0079] It should be noted that, in order to avoid misjudgment and missed judgment caused by fixed thresholds in different batches of products or complex working conditions, this embodiment proposes a dynamic judgment threshold establishment method based on large-sample statistics of qualified products of the same model. The specific implementation process is as follows: 1) Initial Threshold Establishment: In the initial stage of production line testing, vibration signals from a sufficient number (e.g., no less than 100) of qualified compressors of the same model are collected. Following the same process as steps S1 to S7 above, multi-dimensional modulation indices (i.e., first-dimensional modulation index M1, second-dimensional modulation index M2, and third-dimensional modulation index M3) are calculated for each qualified compressor. Subsequently, statistical analysis is performed on the sample distribution of each index, and the mean μ and standard deviation σ of each index are calculated. Based on the 3σ principle of normal distribution, the initial dynamic judgment threshold is set to μ+3σ, which serves as the initial interception benchmark for qualified and defective products.

[0080] 2) Post-production threshold correction: As production line data continues to accumulate, new qualified product samples and confirmed faulty samples will be continuously introduced. When the sample size reaches the preset update cycle, the mean μ and standard deviation σ are automatically recalculated to iteratively correct the dynamic judgment threshold, thereby further improving the threshold's adaptability to fluctuations in the production environment and the accuracy of the interception indicators.

[0081] Step S735: The maximum value of the first ratio and the second ratio is used as the comprehensive severity index, and the failure status of the compressor is classified according to the comprehensive severity index.

[0082] Specifically, in order to comprehensively reflect the severity of the fault, this embodiment takes the maximum value of the first ratio R1 and the second ratio R2 as the comprehensive severity index S. The calculation formula is: S=max(R1,R2).

[0083] Subsequently, based on the magnitude of the comprehensive severity index S, the compressor's fault condition is classified into four levels. In a preferred embodiment, the classification includes: When the overall severity index is less than the first level threshold, the compressor's fault status is determined to be normal. When the comprehensive severity index is greater than or equal to the first level threshold and less than the second level threshold, the compressor is judged to be in a minor fault state. When the comprehensive severity index is greater than or equal to the second-level threshold and less than the third-level threshold, the compressor is judged to be in a medium-level fault condition. When the comprehensive severity index is greater than or equal to the third level threshold, the compressor is judged to be in a serious fault state. Among them, the first level threshold is less than the second level threshold and the third level threshold.

[0084] Specifically, when the comprehensive severity index S <T level1 When T is reached, the compressor's fault condition is determined to be normal; when T... level1 ≤S <T level2 When T is reached, the compressor's fault condition is determined to be a minor fault; when T... level2 ≤S <T level3 When S≥T, the compressor's fault condition is determined to be a medium fault; when S≥T level3 At that time, the compressor's fault condition was determined to be a serious fault. level1 T level2 T level3 These are the first-level threshold, the second-level threshold, and the third-level threshold, respectively, and they satisfy T. level1 <T level2 <T level3 In this embodiment, T is preferably set. level1 =1.0, T level2 =1.2, T level3 =1.5.

[0085] Based on the modulation indices and dynamic judgment results obtained above, two types of results with different dimensions are finally output to serve the different needs of automated rapid screening in production lines and in-depth R&D analysis: 1) Numerical Output (Comprehensive Severity Index S): This output converts the overall judgment result into a single numerical comprehensive severity index S. This value directly corresponds to the aforementioned fault severity classification (e.g., normal, minor, moderate, severe), and its specific value needs to be calibrated in conjunction with the vehicle verification results. This output mode is mainly used in production line rapid screening scenarios, and can seamlessly connect to the production line offline automated testing system to achieve rapid and automated quality judgment of the compressor.

[0086] 2) Two-dimensional modulation spectrum output: A two-dimensional modulation spectrum is generated simultaneously for in-depth analysis. In this two-dimensional modulation spectrum, the horizontal axis represents the carrier frequency, the vertical axis represents the modulation frequency, and the amplitude (usually represented by color intensity) represents the modulation depth. This two-dimensional modulation spectrum can intuitively display the overall distribution of modulation characteristics of each frequency band inside the compressor. It not only provides engineers with a visual basis for troubleshooting difficult faults and analyzing fault evolution mechanisms, but also provides valuable data accumulation for subsequent dynamic threshold correction and algorithm model optimization.

[0087] This embodiment compares the third-dimensional modulation index with a preset concentration threshold. When the third-dimensional modulation index is greater than or equal to the concentration threshold, the compressor's fault state is directly determined to be a severe fault, thus achieving rapid interception without complex comprehensive evaluation when the fault energy is highly concentrated. When the third-dimensional modulation index is less than the concentration threshold, the first ratio of the first-dimensional modulation index to the preset first dynamic threshold and the second ratio of the second-dimensional modulation index to the preset second dynamic threshold are determined. These are then combined with the absolute modulation intensity and the operating baseline for comprehensive evaluation. The maximum value of the first and second ratios is then used as the comprehensive severity index, and the compressor's fault state is graded according to the comprehensive severity index, thereby comprehensively reflecting the severity of the fault and achieving a scientific classification and determination of the fault state.

[0088] Accordingly, please refer to Figure 7 A block diagram of a compressor modulation feature quantitative analysis device provided in this application embodiment, the device comprising: The frequency band decomposition unit 101 is used to acquire the signal to be processed under the steady-state operation of the compressor, and to perform structured frequency band decomposition on the signal to be processed to obtain several sub-band signals. The envelope analysis unit 103 is used to perform Hilbert envelope analysis on each of the several sub-band signals to obtain the envelope signal corresponding to each sub-band signal. The spectrum analysis unit 105 is used to perform spectrum analysis on the envelope signal to obtain the modulation characteristic parameters of each sub-band signal; The comprehensive judgment unit 107 is used to determine multi-dimensional modulation indicators based on modulation feature parameters, compare the modulation indicators with preset dynamic thresholds, and make a comprehensive judgment and classification of the compressor's fault status based on the comparison results.

[0089] In some alternative implementations, the frequency band decomposition unit 101 includes: Steady-state monitoring is performed on the original vibration signal collected during compressor operation. When the steady-state judgment condition is met, the original vibration signal is truncated to obtain the signal to be processed. The signal to be processed is decomposed into structured frequency bands to obtain multiple initial sub-band signals; For each initial sub-band signal, perform single-component determination. If it is determined that the approximate single-component condition is not met, then split the current initial sub-band signal into two new sub-band signals, and recursively perform single-component determination on the split sub-band signals until all sub-band signals meet the approximate single-component condition. All sub-band signals that satisfy the approximate single-component condition are summarized to obtain several sub-band signals.

[0090] In some alternative implementations, the envelope analysis unit 103 includes: Mirroring and extending the first and last ends of each sub-band signal are performed to generate extended sub-band signals. The Hilbert transform is applied to the extended subband signal to construct an analytic signal, and the magnitude of the analytic signal is determined to obtain the initial envelope signal. The original envelope signal is obtained by removing the extended segments corresponding to the mirror extension processing from the initial envelope signal; The original envelope signal is filtered to obtain the envelope signal corresponding to each sub-band signal.

[0091] In some alternative implementations, the spectrum analysis unit 105 includes: Determine the envelope amplitude spectrum corresponding to each envelope signal; Extract the DC component amplitude at zero frequency from the envelope amplitude spectrum and determine the DC component amplitude as the carrier amplitude; The envelope amplitude spectrum is searched based on the preset target fault characteristic frequency peak value, and the maximum amplitude found is determined as the modulation component amplitude. Determine the ratio of the modulation component amplitude to the carrier amplitude, and use this ratio as the modulation characteristic parameter.

[0092] In some optional implementations, the comprehensive determination unit 107 includes: Iterate through the modulation characteristic parameters of each sub-band signal, and determine the maximum value among the modulation characteristic parameters of each sub-band signal as the first dimension modulation index. Traverse the envelope amplitude spectrum corresponding to each envelope signal, extract the maximum amplitude at the first-order frequency conversion as the frequency conversion reference amplitude, and determine the ratio of the first-dimensional modulation index to the frequency conversion reference amplitude as the second-dimensional modulation index. For a target subband signal containing the peak value of the target fault characteristic frequency, the amplitude at the peak value of the target fault characteristic frequency in the envelope amplitude spectrum of the target subband signal is determined, and the ratio of the amplitude to the sum of the amplitudes of all frequency points in the envelope amplitude spectrum of the target subband signal is determined as the third-dimensional modulation index. By summing the first-dimensional modulation index, the second-dimensional modulation index, and the third-dimensional modulation index, a multi-dimensional modulation index is obtained.

[0093] In some optional implementations, the comprehensive determination unit 107 includes: If the third-dimensional modulation index is greater than or equal to the preset concentration threshold, the compressor's fault status is determined to be a serious fault. If the third dimension modulation index is less than the concentration threshold, then the first ratio of the first dimension modulation index to the preset first dynamic threshold is determined, and the second ratio of the second dimension modulation index to the preset second dynamic threshold is determined. The maximum value between the first ratio and the second ratio is used as the comprehensive severity index, and the compressor's failure status is classified according to the comprehensive severity index.

[0094] In some alternative implementations, the grading includes: When the overall severity index is less than the first level threshold, the compressor's fault status is determined to be normal. When the comprehensive severity index is greater than or equal to the first level threshold and less than the second level threshold, the compressor is judged to be in a minor fault state. When the comprehensive severity index is greater than or equal to the second-level threshold and less than the third-level threshold, the compressor is judged to be in a medium-level fault condition. When the comprehensive severity index is greater than or equal to the third level threshold, the compressor is judged to be in a serious fault state. Among them, the first level threshold is less than the second level threshold and the third level threshold.

[0095] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0096] In this embodiment, a compressor modulation feature quantitative analysis device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0097] Please see Figure 8 , Figure 8 This application provides a schematic diagram of the structure of a computer device, as shown in the embodiment of the present application. Figure 8 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 8 Take a processor 10 as an example.

[0098] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0099] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0100] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0101] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0102] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0103] This application also provides a computer-readable storage medium. The methods described in this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the methods shown in the above embodiments are implemented.

[0104] The apparatus and units described in the above embodiments can be implemented by a computer chip or physical entity, or by a product with a certain function. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0105] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0106] Those skilled in the art will understand that the embodiments of this application can be provided as methods or apparatus. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatuses, and devices according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0108] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0109] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0110] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0111] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0112] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0113] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for quantitative analysis of compressor modulation characteristics, characterized in that, The method includes: The signal to be processed under the steady-state operation of the compressor is acquired, and the signal to be processed is decomposed into a structured frequency band to obtain several sub-band signals. Perform Hilbert envelope analysis on each of the sub-band signals to obtain the envelope signal corresponding to each sub-band signal; Spectral analysis is performed on the envelope signal to obtain the modulation characteristic parameters of each sub-band signal; Based on the modulation feature parameters, multi-dimensional modulation indices are determined, and the modulation indices are compared with preset dynamic thresholds to comprehensively determine and classify the fault status of the compressor based on the comparison results.

2. The method according to claim 1, characterized in that, The process involves acquiring the signal to be processed from the compressor under steady-state operating conditions, and performing structured frequency band decomposition on the signal to obtain several sub-band signals, including: Steady-state monitoring is performed on the raw vibration signal collected during compressor operation. When the steady-state determination condition is met, the raw vibration signal is truncated to obtain the signal to be processed. The signal to be processed is subjected to structured frequency band decomposition to obtain multiple initial sub-band signals; For each of the initial sub-band signals, a single-component determination is performed. If the determination is that the approximate single-component condition is not met, the current initial sub-band signal is split into two new sub-band signals, and the single-component determination is recursively performed on the split sub-band signals until all sub-band signals meet the approximate single-component condition. All sub-band signals that satisfy the approximate single-component condition are summarized to obtain the plurality of sub-band signals.

3. The method according to claim 1, characterized in that, The step of performing Hilbert envelope analysis on each of the plurality of sub-band signals to obtain the envelope signal corresponding to each sub-band signal includes: The beginning and end of each sub-band signal are mirrored and extended to generate extended sub-band signals. The extended subband signal is subjected to Hilbert transform to construct an analytic signal, and the magnitude of the analytic signal is determined to obtain the initial envelope signal; The original envelope signal is obtained by removing the extended segment corresponding to the mirror extension process from the initial envelope signal; The original envelope signal is filtered to obtain the envelope signal corresponding to each of the sub-band signals.

4. The method according to claim 1, characterized in that, The step of performing spectral analysis on the envelope signal to obtain the modulation characteristic parameters of each sub-band signal includes: Determine the envelope amplitude spectrum corresponding to each envelope signal; The DC component amplitude at zero frequency is extracted from the envelope amplitude spectrum, and the DC component amplitude is determined as the carrier amplitude. The envelope amplitude spectrum is searched based on the preset target fault characteristic frequency peak value, and the maximum amplitude found is determined as the modulation component amplitude. The ratio of the modulation component amplitude to the carrier amplitude is determined, and the ratio is defined as the modulation characteristic parameter.

5. The method according to claim 1, characterized in that, The determination of multi-dimensional modulation indices based on the modulation feature parameters includes: The modulation characteristic parameters of each sub-band signal are traversed, and the maximum value among the modulation characteristic parameters of each sub-band signal is determined as the first dimension modulation index. Traverse the envelope amplitude spectrum corresponding to each envelope signal, extract the maximum amplitude at the first-order frequency conversion as the frequency conversion reference amplitude, and determine the ratio of the first-dimensional modulation index to the frequency conversion reference amplitude as the second-dimensional modulation index. For a target subband signal containing a target fault characteristic frequency peak, the amplitude at the target fault characteristic frequency peak in the envelope amplitude spectrum of the target subband signal is determined, and the ratio of the amplitude to the sum of the amplitudes of all frequency points in the envelope amplitude spectrum of the target subband signal is determined as the third-dimensional modulation index. The first-dimensional modulation index, the second-dimensional modulation index, and the third-dimensional modulation index are summarized to obtain the multi-dimensional modulation index.

6. The method according to claim 5, characterized in that, The step of comparing the modulation index with a preset dynamic threshold to comprehensively determine and classify the fault state of the compressor based on the comparison result includes: If the third-dimensional modulation index is greater than or equal to the preset concentration threshold, the compressor is determined to be in a serious fault state. If the third dimension modulation index is less than the concentration threshold, then a first ratio of the first dimension modulation index to the preset first dynamic threshold is determined, and a second ratio of the second dimension modulation index to the preset second dynamic threshold is determined. The maximum value between the first ratio and the second ratio is used as a comprehensive severity index, and the failure status of the compressor is classified according to the comprehensive severity index.

7. The method according to claim 6, characterized in that, The classification includes: When the comprehensive severity index is less than the first grading threshold, the compressor is determined to be in a normal fault state. When the comprehensive severity index is greater than or equal to the first grading threshold and less than the second grading threshold, the compressor is determined to be in a minor fault state. When the comprehensive severity index is greater than or equal to the second grading threshold and less than the third grading threshold, the compressor is determined to be in a medium fault condition. When the comprehensive severity index is greater than or equal to the third grading threshold, the compressor is determined to be in a serious fault state. Wherein, the first grading threshold < the second grading threshold < the third grading threshold.

8. A device for quantitative analysis of compressor modulation characteristics, characterized in that, The device includes: The frequency band decomposition unit is used to acquire the signal to be processed under the steady-state operation of the compressor, and to perform structured frequency band decomposition on the signal to be processed to obtain several sub-band signals. The envelope analysis unit is used to perform Hilbert envelope analysis on each of the plurality of sub-band signals to obtain the envelope signal corresponding to each of the sub-band signals. The spectrum analysis unit is used to perform spectrum analysis on the envelope signal to obtain the modulation characteristic parameters of each sub-band signal; The comprehensive judgment unit is used to determine multi-dimensional modulation indicators based on the modulation feature parameters, compare the modulation indicators with preset dynamic thresholds, and make a comprehensive judgment and classification of the fault status of the compressor based on the comparison results.

9. A computer device, characterized in that it comprises: A memory and a processor are interconnected, the memory storing computer instructions, and the processor executing the computer instructions to perform the compressor modulation feature quantitative analysis method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the compressor modulation feature quantitative analysis method according to any one of claims 1 to 7.

Images