Time-frequency analysis method and device for dynamic pressure signal of gas compressor

Abstract

The invention relates to the technical field of non-stationary signal processing, and discloses a time-frequency analysis method and device for a dynamic pressure signal of a gas compressor, and the method comprises the steps: substituting an instantaneous frequency estimation operator in an original local maximum synchronous compression transformation algorithm into an instantaneous frequency independent variable of the instantaneous frequency estimation operator to obtain a double instantaneous frequency estimation operator; in each iteration process, the algorithm is utilized to execute the double local maximum synchronous compression transformation operation on the time-frequency distribution after the last iteration, and through multiple iterations, a multiple local maximum synchronous compression linear frequency modulation wavelet transform time-frequency distribution result can be obtained. According to the method, the time-frequency analysis result of the gas compressor can be accurately represented to represent the surge information of the gas compressor, so that the energy concentration ratio of the time-frequency result is improved, the phenomenon of energy leakage near an instantaneous frequency track is eliminated, and the high-precision time-frequency analysis result can be obtained while the amplitude information and the time-varying characteristic of the surge dynamic pressure signal of the gas compressor are accurately represented.

Description

Technical Field

[0001] The present invention relates to the technical field of non-stationary signal processing, and particularly relates to a time-frequency analysis method and device for compressor dynamic pressure signals. Background Art

[0002] As a key component in the operation of an aeroengine, the operating state of a compressor plays a crucial role in the health of the aeroengine. Therefore, the analysis of the compressor operating state is particularly important. The dynamic pressure signal is a crucial analysis parameter during the operation of the compressor and is an intuitive criterion for judging abnormal states such as compressor instability. However, under the complex operating conditions of the compressor, especially when the compressor surges, the collected dynamic pressure information has non-stationary characteristics.

[0003] When the existing time-frequency analysis methods are used to process the dynamic pressure signals collected when the compressor surges, the energy concentration of the time-frequency results is insufficient, and there is slight energy leakage near the instantaneous frequency trajectory, resulting in low energy on the instantaneous frequency trajectory and unable to accurately represent the amplitude information and time-varying characteristics of the compressor dynamic pressure signal. Summary of the Invention

[0004] In view of this, the present invention provides a time-frequency analysis method and device for compressor dynamic pressure signals to solve the problem that the existing time-frequency analysis methods cannot accurately represent the amplitude information and time-varying characteristics of the compressor dynamic pressure signals when processing the dynamic pressure signals collected when the compressor surges.

[0005] First aspect, the present invention provides a time-frequency analysis method for compressor dynamic pressure signals. The method includes: substituting the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm as a whole into the instantaneous frequency independent variable of the operator itself to obtain a double instantaneous frequency estimation operator; replacing the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm with the double instantaneous frequency estimation operator to obtain a double local maximum synchrosqueezing transform algorithm; acquiring a compressor dynamic pressure signal and performing a chirplet transform on the compressor dynamic pressure signal to obtain an initial LCT time-frequency distribution corresponding to the compressor dynamic pressure signal, and iteratively performing a double local maximum synchrosqueezing transform operation until the time-frequency distribution after the current iteration is the same as the time-frequency distribution after the previous iteration, completing the iteration, and obtaining a multiple local maximum synchrosqueezing chirplet transform time-frequency distribution result corresponding to the compressor dynamic pressure signal. The double local maximum synchrosqueezing transform operation includes: in the current iteration process, performing a synchrosqueezing transform operation on the time-frequency distribution after the previous iteration by using the double local maximum synchrosqueezing transform algorithm to obtain the time-frequency distribution after the current iteration, where when the current iteration is the first iteration, the time-frequency distribution after the previous iteration is the initial LCT time-frequency distribution; and identifying surge information of the compressor based on the multiple local maximum synchrosqueezing chirplet transform time-frequency distribution result corresponding to the compressor dynamic pressure signal.

[0006] The time-frequency analysis method for compressor dynamic pressure signals provided by the present invention, by substituting the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm as a whole into the instantaneous frequency independent variable of the operator itself to obtain a double instantaneous frequency estimation operator, and then obtaining a double local maximum synchrosqueezing transform algorithm. In the process of performing the double local maximum synchrosqueezing transform operation in each iteration, the synchrosqueezing transform operation is performed on the time-frequency distribution after the previous iteration by using the double local maximum synchrosqueezing transform algorithm to obtain the time-frequency distribution after the current iteration. When the time-frequency distribution after the current iteration is the same as the time-frequency distribution after the previous iteration, a multiple local maximum synchrosqueezing chirplet transform time-frequency distribution result corresponding to the compressor dynamic pressure signal can be obtained. Finally, the surge information of the compressor is characterized based on the time-frequency distribution result, significantly improving the energy concentration of the time-frequency result, concentrating the energy on the instantaneous frequency trajectory, eliminating the phenomenon of energy leakage near the instantaneous frequency trajectory, and being able to accurately characterize the amplitude information and time-varying characteristics of the compressor surge dynamic pressure signal while obtaining a high-precision time-frequency analysis result.

[0007] In an optional implementation manner, performing a chirplet transform on the compressor dynamic pressure signal, and the initial LCT time-frequency distribution corresponding to the obtained compressor dynamic pressure signal is expressed by the following formula:

[0008]

[0009] Among them, represents the initial LCT time-frequency distribution; ω represents the instantaneous frequency of the signal; g represents the window function; f(u) represents the compressor dynamic pressure signal; u represents the time variable; β represents the chirp rate parameter.

[0010] In an alternative embodiment, a local maximum synchrosqueezing transform operation is performed on the initial LCT time-frequency distribution by using the initial local maximum synchrosqueezing transform algorithm, and the obtained local maximum synchrosqueezing linear chirplet transform time-frequency distribution is specifically expressed as:

[0011]

[0012] Among them, T1(t,η) represents the local maximum synchrosqueezing linear chirplet transform time-frequency distribution obtained by using the initial local maximum synchrosqueezing transform algorithm; η represents the instantaneous frequency variable; δ represents the Dirac function; ω m (t,ω) represents the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm, and its specific expression is:

[0013]

[0014] Among them, Δ represents the preset discrete frequency interval; represents the time-frequency spectrum of the linear chirplet transform.

[0015] In an alternative embodiment, the entire first instantaneous frequency estimation operator is substituted into the instantaneous frequency independent variable of the operator itself, and the obtained double instantaneous frequency estimation operator is specifically expressed as:

[0016]

[0017] Among them, represents the double instantaneous frequency estimation operator.

[0018] In an alternative embodiment, in the first iteration process, a synchrosqueezing transform operation is performed on the initial LCT time-frequency distribution by using the double local maximum synchrosqueezing transform algorithm, and the obtained time-frequency distribution after the first iteration is specifically expressed as:

[0019]

[0020] Among them, represents the time-frequency distribution after the first iteration.

[0021] In an alternative embodiment, during the nth iteration, the time-frequency distribution after the current iteration is specifically represented as:

[0022]

[0023] where n represents the number of iterations; represents the time-frequency distribution after the nth iteration; represents the time-frequency distribution after the (n - 1)th iteration; represents a double instantaneous frequency estimation operator; ω represents the instantaneous frequency of the signal; η represents the instantaneous frequency variable; δ represents the Dirac function.

[0024] In a second aspect, the present invention provides a time-frequency analysis device for a compressor dynamic pressure signal. The device includes: a double operator calculation module, configured to substitute the first instantaneous frequency estimation operator in the initial local maximum synchronous compression transform algorithm as a whole into the instantaneous frequency independent variable of the operator itself to obtain a double instantaneous frequency estimation operator; a double algorithm determination module, configured to replace the first instantaneous frequency estimation operator in the initial local maximum synchronous compression transform algorithm with the double instantaneous frequency estimation operator to obtain a double local maximum synchronous compression transform algorithm; a time-frequency distribution determination module, configured to acquire a compressor dynamic pressure signal and perform a chirplet transform on the compressor dynamic pressure signal to obtain an initial LCT time-frequency distribution corresponding to the compressor dynamic pressure signal, and iteratively perform a double local maximum synchronous compression transform operation until the time-frequency distribution after the current iteration is the same as the time-frequency distribution after the previous iteration, completing the iteration to obtain a time-frequency distribution result of multiple local maximum synchronous compression chirplet transforms corresponding to the compressor dynamic pressure signal. The double local maximum synchronous compression transform operation includes: during the current iteration, performing a synchronous compression transform operation on the time-frequency distribution after the previous iteration by using the double local maximum synchronous compression transform algorithm to obtain the time-frequency distribution after the current iteration, where when the current iteration is the first iteration, the time-frequency distribution after the previous iteration is the initial LCT time-frequency distribution; a surge information identification module, configured to identify the surge information of the compressor based on the time-frequency distribution result of multiple local maximum synchronous compression chirplet transforms corresponding to the compressor dynamic pressure signal.

[0025] In a third aspect, the present invention provides a computer device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the time-frequency analysis method for a compressor dynamic pressure signal according to the first aspect or any corresponding embodiment thereof.

[0026] Fourthly, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to execute the time-frequency analysis method for the dynamic pressure signal of the compressor according to the first aspect or any corresponding embodiment thereof.

[0027] Fifthly, the present invention provides a computer program product, including computer instructions, and the computer instructions are used to cause a computer to execute the time-frequency analysis method for the dynamic pressure signal of the compressor according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0029] Figure 1 is a schematic flowchart of the time-frequency analysis method for the dynamic pressure signal of the compressor according to an embodiment of the present invention;

[0030] Figure 2 is an example flowchart of the first double local maximum synchrosqueezing transform operation according to an embodiment of the present invention;

[0031] Figure 3 is an example flowchart of repeatedly performing the double local maximum synchrosqueezing transform operation according to an embodiment of the present invention;

[0032] Figure 4 is an example diagram of the dynamic pressure signal when the engine compressor component surges during the test according to an embodiment of the present invention;

[0033] Figure 5 is an example diagram of the effect difference between using the existing LMSCT algorithm and the MLMSCT algorithm for the compressor surge dynamic pressure signal;

[0034] Figure 6 is an example diagram of the time-frequency energy distribution when using the existing LMSCT algorithm and the MLMSCT algorithm respectively for processing non-stationary signals;

[0035] Figure 7 is a structural block diagram of the time-frequency analysis device for the dynamic pressure signal of the compressor according to an embodiment of the present invention;

[0036] Figure 8 is a schematic hardware structure diagram of the computer device according to an embodiment of the present invention. Detailed implementation mode

[0037] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0038] As a key component in the operation of an aeroengine, the operating state of the compressor plays a crucial role in the health of the aeroengine. Therefore, the analysis of the compressor operating state is particularly important. The dynamic pressure signal is a crucial analysis parameter during the operation of the compressor and is an intuitive criterion for judging abnormal states such as compressor instability. However, under the complex operating conditions of the compressor, especially when the compressor surges, the collected dynamic pressure signal has non-stationary characteristics. How to extract accurate time-frequency information from non-stationary signals has always been a highly challenging task in the field of signal processing.

[0039] According to an embodiment of the present invention, an embodiment of a time-frequency analysis method for compressor dynamic pressure signals is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0040] In this embodiment, a time-frequency analysis method for compressor dynamic pressure signals is provided, which can be used in the above computer device. Figure 1 is a flowchart of the time-frequency analysis method for compressor dynamic pressure signals according to an embodiment of the present invention, as Figure 1 shown, the process includes the following steps:

[0041] Step S101, substitute the first instantaneous frequency estimator in the initial local maximum synchrosqueezing transform algorithm as a whole into the instantaneous frequency independent variable of the operator itself to obtain a double instantaneous frequency estimator.

[0042] Considering that in the existing time-frequency methods, the reason for the insufficient energy concentration of the time-frequency results is that the accuracy of the instantaneous frequency estimator in the local maximum synchrosqueezing transform algorithm is insufficient, resulting in low energy on the instantaneous frequency trajectory in the time-frequency results. In order to obtain a higher-precision instantaneous frequency estimator, an embodiment of the present invention can substitute the first instantaneous frequency estimator in the initial local maximum synchrosqueezing transform algorithm as a whole into the instantaneous frequency independent variable of the operator itself, and then a double instantaneous frequency estimator can be obtained.

[0043] The first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm can be expressed as:

[0044]

[0045] Where, Δ represents a preset discrete frequency interval; represents the time-frequency spectrum of the chirplet transform; substituting the whole first instantaneous frequency estimation operator into the instantaneous frequency independent variable of the operator itself, the double instantaneous frequency estimation operator can be obtained, and its specific expression is:

[0046]

[0047] Step S102, replacing the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm with the double instantaneous frequency estimation operator to obtain the double local maximum synchrosqueezing transform algorithm.

[0048] In the embodiment of the present invention, the double instantaneous frequency estimation operator obtained in the above steps can be used to replace the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm to obtain the double local maximum synchrosqueezing transform algorithm. Subsequently, using the double local maximum synchrosqueezing transform algorithm to process the time-frequency distribution not only improves the accuracy of the time-frequency result, but also, due to the double processing of the instantaneous frequency estimation operator of the local maximum synchrosqueezing transform algorithm, can reduce the number of iterations to the greatest extent on the premise of ensuring the optimal time-frequency result, and improve the efficiency of the time-frequency analysis of the compressor dynamic pressure signal. For example, the existing number of iterations is six times, and the present invention using the double local maximum synchrosqueezing transform algorithm can reduce the number of iterations to three times.

[0049] Step S103, obtaining the compressor dynamic pressure signal, performing a chirplet transform on the compressor dynamic pressure signal to obtain the initial LCT time-frequency distribution corresponding to the compressor dynamic pressure signal, and iteratively performing the double local maximum synchrosqueezing transform operation until the time-frequency distribution after the current iteration is the same as the time-frequency distribution after the previous iteration, completing the iteration, and obtaining the time-frequency distribution result of the multiple local maximum synchrosqueezing chirplet transform corresponding to the compressor dynamic pressure signal.

[0050] Where, the double local maximum synchrosqueezing transform operation includes: in the current iteration process, using the double local maximum synchrosqueezing transform algorithm to perform a synchrosqueezing transform operation on the time-frequency distribution after the previous iteration to obtain the time-frequency distribution after the current iteration, where, when the current iteration is the first iteration, the time-frequency distribution after the previous iteration is the initial LCT time-frequency distribution.

[0051] The embodiments of the present invention do not limit the manner of obtaining the dynamic pressure signal of the compressor. The dynamic pressure signal of the compressor can be obtained by means such as a pressure sensor, a pressure measuring instrument, or computational fluid dynamics (CFD). For example, Figure 2 As shown, the linear frequency modulation wavelet transform can be performed on the dynamic pressure signal of the compressor (i.e., the non-stationary signal) to obtain the initial LCT time-frequency distribution corresponding to the dynamic pressure signal of the compressor, so that it can better match the frequency change characteristics in the dynamic pressure signal of the compressor and effectively capture the dynamic change of the pressure signal over time.

[0052] The embodiments of the present invention start to perform the iterative operation of the double local maximum synchronous compression transform. As Figure 2 shown, in the first iteration process, the synchronous compression transform operation is performed on the initial LCT time-frequency distribution by using the double local maximum synchronous compression transform algorithm to obtain the time-frequency distribution after the first iteration. In the second iteration process, the synchronous compression transform operation can be performed on the time-frequency distribution after the first iteration by using the double local maximum synchronous compression transform algorithm to obtain the time-frequency distribution after the second iteration. At this time, it can be judged whether the time-frequency distribution after the second iteration is consistent with the time-frequency distribution after the first iteration. If not, the iterative operation of the double local maximum synchronous compression transform is continued until the time-frequency distribution after the current iteration is consistent with the time-frequency distribution after the previous iteration; if the time-frequency distribution after the second iteration is consistent with the time-frequency distribution after the first iteration, the iteration is ended, and the time-frequency distribution after the second iteration is output and determined as the time-frequency distribution result of the multiple local maximum synchronous compression linear frequency modulation wavelet transform corresponding to the dynamic pressure signal of the compressor.

[0053] Step S104, based on the time-frequency distribution result of the multiple local maximum synchronous compression linear frequency modulation wavelet transform corresponding to the dynamic pressure signal of the compressor, identify the surge information of the compressor.

[0054] The embodiments of the present invention can identify the amplitude information and time-varying characteristics in the time-frequency distribution result of the multiple local maximum synchronous compression linear frequency modulation wavelet transform corresponding to the dynamic pressure signal of the compressor to characterize the surge information of the compressor, and can more specifically judge the surge information of the compressor.

[0055] The time-frequency analysis method for the dynamic pressure signal of a compressor provided in this embodiment obtains a double instantaneous frequency estimation operator by substituting the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm as a whole into the instantaneous frequency independent variable of the operator itself, and then obtains a double local maximum synchrosqueezing transform algorithm. During the process of performing the double local maximum synchrosqueezing transform operation in each iteration, the double local maximum synchrosqueezing transform algorithm is used to perform a synchrosqueezing transform operation on the time-frequency distribution after the previous iteration to obtain the time-frequency distribution after the current iteration. When the time-frequency distribution after the current iteration is consistent with the time-frequency distribution after the previous iteration, the time-frequency distribution results of the multiple local maximum synchrosqueezing linear chirplet transform corresponding to the dynamic pressure signal of the compressor can be obtained. Finally, the surge information of the compressor is characterized based on the time-frequency distribution results, which significantly improves the energy concentration of the time-frequency results, concentrates the energy on the instantaneous frequency trajectory, and eliminates the phenomenon of energy leakage near the instantaneous frequency trajectory. While accurately characterizing the amplitude information and time-varying characteristics of the dynamic pressure signal of the compressor surge, a high-precision time-frequency analysis result can also be obtained.

[0056] In a specific embodiment, as Figure 3 shown, the embodiment of the present invention can pre-define the initial value of the number of iterations as 2, obtain the dynamic pressure signal (non-stationary signal) of the compressor, and perform a linear chirplet transform (LCT) on the dynamic pressure signal of the compressor to obtain the initial LCT time-frequency distribution corresponding to the dynamic pressure signal of the compressor Its specific expression is:

[0057]

[0058] Wherein, represents the initial LCT time-frequency distribution; ω represents the instantaneous frequency of the signal; g represents the window function; f(u) represents the dynamic pressure signal of the compressor; u represents the time variable; β represents the chirp rate parameter.

[0059] In one iteration process, specifically, the double local maximum synchrosqueezing transform algorithm is used to perform a synchrosqueezing transform operation on the initial LCT time-frequency distribution. The principle operation of the time-frequency distribution after the first iteration obtained is: First, the initial local maximum synchrosqueezing transform algorithm is used to perform a local maximum synchrosqueezing transform operation on the initial LCT time-frequency distribution, and the time-frequency distribution result of the local maximum synchrosqueezing linear chirplet transform obtained is:

[0060]

[0061] Among them, \(T1(t,η)\) represents the time-frequency distribution of the local maximum synchrosqueezing linear frequency modulation wavelet transform obtained by using the initial local maximum synchrosqueezing transform algorithm; \(η\) represents the instantaneous frequency variable; \(δ\) represents the Dirac function; \(ω\) m (t,ω) represents the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm, and its specific expression is:

[0062]

[0063] Among them, \(Δ\) represents the preset discrete frequency interval; represents the time-frequency spectrum of the linear frequency modulation wavelet transform. In order to obtain a higher-precision instantaneous frequency estimation operator in the embodiments of the present invention, the entire first instantaneous frequency estimation operator can be substituted into the instantaneous frequency independent variable of the operator itself to obtain a double instantaneous frequency estimation operator, which is specifically expressed as:

[0064]

[0065] Among them, represents the double instantaneous frequency estimation operator; and by replacing the original first instantaneous frequency estimation operator with the double instantaneous frequency estimation operator, the double local maximum synchrosqueezing transform algorithm can be obtained.

[0066] In the embodiments of the present invention, the obtained double local maximum synchrosqueezing transform algorithm is used to perform a synchrosqueezing transform operation on the initial LCT time-frequency distribution, and the time-frequency distribution after the first iteration can be obtained, which is specifically expressed as:

[0067]

[0068] Among them, represents the time-frequency distribution after the first iteration; in subsequent iterations, the double local maximum synchrosqueezing transform algorithm can be directly used to perform operations.

[0069] In the \(n\)th iteration process of the embodiments of the present invention, the double local maximum synchrosqueezing transform algorithm can be used to perform a synchrosqueezing transform operation on the time-frequency distribution after the previous iteration to obtain the time-frequency distribution after the current iteration, and it can be determined whether the time-frequency distribution after the current iteration is consistent with the time-frequency distribution after the previous iteration. If not, the double local maximum synchrosqueezing transform algorithm continues to be used for operations. If they are consistent, the time-frequency distribution after the current iteration is output and determined as the time-frequency distribution result of the multiple local maximum synchrosqueezing linear frequency modulation wavelet transform corresponding to the dynamic pressure signal of the compressor.

[0070] In a specific embodiment, the time-frequency distribution after the first iteration is continuously subjected to the double local maximum synchrosqueezing transform operation to obtain the time-frequency distribution after the second iteration, which is specifically expressed as:

[0071]

[0072] Similarly, through iteration, the time-frequency distribution after the third iteration can be obtained as follows:

[0073]

[0074] Generally, after continuously performing the double local maximum synchronous compression transform operation, the time-frequency distribution after the nth iteration can be obtained as follows:

[0075]

[0076] where n represents the number of iterations; represents the time-frequency distribution after the nth iteration, n ≥ 2; represents the time-frequency distribution after the (n - 1)th iteration; represents the double instantaneous frequency estimation operator; ω represents the instantaneous frequency of the signal; η represents the instantaneous frequency variable; δ represents the Dirac function.

[0077] The iteration termination condition is: when it indicates that the optimal solution has been found in the iteration process, and the iteration operation terminates, outputting the optimal iteration result

[0078] Based on the local maximum synchronous compression linear frequency modulation wavelet transform algorithm, the present invention creatively constructs a double instantaneous frequency estimation operator to replace the original operator, optimizes the accuracy of the instantaneous frequency estimation operator, and then improves the original local maximum synchronous compression transform algorithm into a double local maximum synchronous compression transform algorithm. In addition, an iterative step of continuously performing the double local maximum synchronous compression transform operation on the time-frequency distribution of the linear frequency modulation wavelet transform (LCT) is innovatively designed to seek the optimal solution. Through the process of redistributing the time-frequency coefficients through multiple iterations, the diverging time-frequency energy is gradually concentrated, and then the optimal time-frequency result is iteratively obtained, accurately characterizing the amplitude information and time-varying characteristics of the compressor surge dynamic pressure signal.

[0079] In specific applications, as Figure 4 shown, it is the dynamic pressure signal when a certain type of aero-engine compressor component surges during the test. This dynamic pressure signal is collected by a high-frequency dynamic pressure sensor, and the sampling frequency is 20 kHz. The MLMSCT algorithm (multiple local maximum synchronous compression linear frequency modulation wavelet transform algorithm) proposed by the present invention and the existing LMSCT algorithm (local maximum synchronous compression linear frequency modulation wavelet transform algorithm) are respectively used to process the compressor surge dynamic pressure signal, as Figure 5The figure shows the processing result diagrams of the compressor surge dynamic pressure signal using the MLMSCT algorithm (Multiple Local Maximum Synchronous Compression Linear Frequency Modulation Wavelet Transform algorithm) and the existing LMSCT algorithm (Local Maximum Synchronous Compression Linear Frequency Modulation Wavelet Transform algorithm) respectively. From the visual effect evaluation, the time-frequency result of MLMSCT has less noise near the instantaneous frequency trajectory, while the time-frequency result of LMSCT has a slight energy leakage near the instantaneous frequency trajectory, as Figure 6 shown. In order to more accurately analyze the energy distribution of the time-frequency results, the time-frequency energy slice at a certain moment is plotted. The amplitude of the main frequency band needs to be focused on. The peak energy of the time-frequency slice of the MLMSCT algorithm is significantly higher than that of the LMSCT algorithm, indicating that the MLMSCT algorithm has a higher time-frequency energy concentration in processing non-stationary signals; on both sides of the main frequency band, where there should be no energy distribution, the time-frequency slice of the LMSCT algorithm has a weak energy distribution, which quantitatively shows that there is a slight energy leakage in the LMSCT result near the instantaneous frequency trajectory, while the time-frequency result of the present invention has no energy leakage phenomenon near the instantaneous frequency trajectory.

[0080] In order to quantitatively evaluate the time-frequency energy concentration of the present invention, the concept of Rényi entropy is introduced here. Rényi entropy is an index used to objectively evaluate the energy concentration of the time-frequency trajectory. The smaller the Rényi entropy value, the higher the energy concentration of the time-frequency trajectory; conversely, it indicates a lower energy concentration of the time-frequency trajectory. The calculation formula of the Rényi entropy of the time-frequency distribution result is:

[0081]

[0082] In the formula: TFR(t,ω) is the time-frequency distribution result of the signal. The Rényi entropy values corresponding to the LMSCT algorithm and the MLMSCT algorithm calculated according to the formula are shown in Table 1 below:

[0083] Table 1

[0084] Time-frequency analysis algorithm LMSCT MLMSCT Rényi entropy 12.6495 11.9783

[0085] It can be seen that MLMSCT has a lower Rényi entropy, which quantitatively shows that the energy concentration of the time-frequency trajectory of the present invention is significantly higher than that of the LMSCT algorithm. Therefore, MLMSCT can more accurately characterize the amplitude information and time-varying characteristics of the compressor surge dynamic pressure signal, providing an effective technical means for the analysis of the compressor surge signal.

[0086] In this embodiment, a time-frequency analysis device for compressor dynamic pressure signals is further provided. This device is used to implement the above-mentioned embodiments and preferred implementation manners, and those that have been described will not be elaborated again. As used hereinafter, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0087] This embodiment provides a time-frequency analysis device for compressor dynamic pressure signals. As Figure 7 shown, it includes: a double-operator calculation module 701, which is used to substitute the first instantaneous frequency estimation operator in the initial local maximum synchronous compression transform algorithm into the instantaneous frequency independent variable of the operator itself to obtain a double instantaneous frequency estimation operator; a double-algorithm determination module 702, which is used to replace the first instantaneous frequency estimation operator in the initial local maximum synchronous compression transform algorithm with the double instantaneous frequency estimation operator to obtain a double local maximum synchronous compression transform algorithm; a time-frequency distribution determination module 703, which is used to acquire the compressor dynamic pressure signal, perform a linear frequency modulation wavelet transform on the compressor dynamic pressure signal to obtain the initial LCT time-frequency distribution corresponding to the compressor dynamic pressure signal, and iteratively perform the double local maximum synchronous compression transform operation until the time-frequency distribution after the current iteration is the same as the time-frequency distribution after the previous iteration, completing the iteration and obtaining the time-frequency distribution result of the multiple local maximum synchronous compression linear frequency modulation wavelet transform corresponding to the compressor dynamic pressure signal. The double local maximum synchronous compression transform operation includes: in the current iteration process, performing a synchronous compression transform operation on the time-frequency distribution after the previous iteration using the double local maximum synchronous compression transform algorithm to obtain the time-frequency distribution after the current iteration, where when the current iteration is the first iteration, the time-frequency distribution after the previous iteration is the initial LCT time-frequency distribution; a surge information recognition module 704, which is used to recognize the surge information of the compressor based on the time-frequency distribution result of the multiple local maximum synchronous compression linear frequency modulation wavelet transform corresponding to the compressor dynamic pressure signal.

[0088] The further functional descriptions of the above-mentioned various modules and units are the same as those in the corresponding embodiments above and will not be elaborated here.

[0089] The time-frequency analysis device for compressor dynamic pressure signals in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0090] This embodiment of the present invention also provides a computer device having the aboveFigure 7 A time-frequency analysis device for a compressor dynamic pressure signal as shown.

[0091] Please refer to Figure 8 , Figure 8 which is a schematic structural diagram of a computer device provided by an alternative embodiment of the present invention. As Figure 8 shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting the components, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 8 In

[0092] FIG. 11, one processor 10 is taken as an example.

[0093] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above hardware chip can be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above programmable logic device can be a complex programmable logic device, a field programmable gate array, a general array logic, or any combination thereof.

[0094] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiments.

[0095] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, a hard disk, or a solid state drive; the memory 20 may further include a combination of the above types of memory.

[0096] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 may be connected through a bus or other means. Figure 8 Taking connection through a bus as an example.

[0097] The input device 30 can receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (such as an LED), and a haptic feedback device (such as a vibration motor), etc. The above display device includes but is not limited to a liquid crystal display, a light emitting diode, a display, and a plasma display. In some alternative embodiments, the display device may be a touch screen.

[0098] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processes on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium may be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state drive, etc.; further, the storage medium may further include a combination of the above types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiment is implemented.

[0099] A part of the present invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the present invention through the operations of the computer. Those skilled in the art should understand that the forms in which computer program instructions exist in a computer-readable medium include but are not limited to source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include but are not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0100] Although embodiments of the present invention 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 the present invention, and such modifications and variations all fall within the defined scope.

Claims

1. A time-frequency analysis method for the dynamic pressure signal of a compressor, characterized in that The method includes: Substituting the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm as a whole into the instantaneous frequency independent variable of the operator itself to obtain a double instantaneous frequency estimation operator; Replacing the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm with the double instantaneous frequency estimation operator to obtain a double local maximum synchrosqueezing transform algorithm; Obtaining a compressor dynamic pressure signal, performing a chirplet transform on the compressor dynamic pressure signal to obtain an initial LCT time-frequency distribution corresponding to the compressor dynamic pressure signal, and iteratively performing a double local maximum synchrosqueezing transform operation until the time-frequency distribution after the current iteration is the same as the time-frequency distribution after the previous iteration, completing the iteration, and obtaining a multiple local maximum synchrosqueezing chirplet transform time-frequency distribution result corresponding to the compressor dynamic pressure signal. The double local maximum synchrosqueezing transform operation includes: In the current iteration process, performing a synchrosqueezing transform operation on the time-frequency distribution after the previous iteration by using the double local maximum synchrosqueezing transform algorithm to obtain the time-frequency distribution after the current iteration. Wherein, when the current iteration is the first iteration, the time-frequency distribution after the previous iteration is the initial LCT time-frequency distribution; Identifying the surge information of the compressor based on the multiple local maximum synchrosqueezing chirplet transform time-frequency distribution result corresponding to the compressor dynamic pressure signal.

2. The method according to claim 1, wherein Performing a chirplet transform on the compressor dynamic pressure signal, and the initial LCT time-frequency distribution corresponding to the obtained compressor dynamic pressure signal is expressed by the following formula: Among them, represents the initial LCT time-frequency distribution; ω represents the instantaneous frequency of the signal; g represents the window function; f(u) represents the compressor dynamic pressure signal; u represents the time variable; β represents the chirp rate parameter.

3. The method according to claim 2, wherein Performing a local maximum synchrosqueezing transform operation on the initial LCT time-frequency distribution by using the initial local maximum synchrosqueezing transform algorithm, and the local maximum synchrosqueezing chirplet transform time-frequency distribution obtained is specifically expressed as: Among them, T1(t,η) represents the time-frequency distribution of the local maximum synchrosqueezing linear frequency modulation wavelet transform obtained by using the initial local maximum synchrosqueezing transform algorithm; η represents the instantaneous frequency variable; δ represents the Dirac function; ω m (t,ω) represents the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm, and its specific expression is: where, Δ represents a preset discrete frequency interval; represents the time-frequency spectrum of the chirplet transform.

4. The method according to claim 3, characterized in that, Substituting the first instantaneous frequency estimation operator as a whole into the instantaneous frequency independent variable of the operator itself, and the obtained double instantaneous frequency estimation operator is specifically expressed as: Among them, represents a double instantaneous frequency estimation operator.

5. The method according to claim 4, wherein In the first iteration process, performing a synchrosqueezing transform operation on the initial LCT time-frequency distribution by using the double local maximum synchrosqueezing transform algorithm, and the time-frequency distribution after the first iteration obtained is specifically expressed as: Among them, represents the time-frequency distribution after the first iteration.

6. The method according to claim 1 or 5, characterized in that In the process of the nth iteration, the time-frequency distribution after the current iteration is specifically expressed as: where n represents the number of iterations; represents the time-frequency distribution after the n-th iteration; represents the time-frequency distribution after the (n - 1)-th iteration; represents the double instantaneous frequency estimation operator; ω represents the instantaneous frequency of the signal; η represents the instantaneous frequency variable; δ represents the Dirac function.

7. A time-frequency analysis device for compressor dynamic pressure signals, characterized in that, The device includes: A double operator calculation module, configured to substitute the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm as a whole into the instantaneous frequency independent variable of the operator itself to obtain a double instantaneous frequency estimation operator; A double algorithm determination module, configured to replace the first instantaneous frequency estimation operator in the initial local maximum synchrosqueezing transform algorithm with the double instantaneous frequency estimation operator to obtain a double local maximum synchrosqueezing transform algorithm; A time-frequency distribution determination module, configured to obtain a compressor dynamic pressure signal, perform a linear frequency modulation wavelet transform on the compressor dynamic pressure signal to obtain an initial LCT time-frequency distribution corresponding to the compressor dynamic pressure signal, and iteratively perform a double local maximum synchronous compression transform operation until the time-frequency distribution after the current iteration is the same as the time-frequency distribution after the previous iteration, completing the iteration and obtaining a time-frequency distribution result of multiple local maximum synchronous compression linear frequency modulation wavelet transforms corresponding to the compressor dynamic pressure signal. The double local maximum synchronous compression transform operation includes: In the current iteration process, perform a synchronous compression transform operation on the time-frequency distribution after the previous iteration by using the double local maximum synchronous compression transform algorithm to obtain the time-frequency distribution after the current iteration. Wherein, when the current iteration is the first iteration, the time-frequency distribution after the previous iteration is the initial LCT time-frequency distribution; A surge information identification module, configured to identify surge information of the compressor based on the time-frequency distribution result of multiple local maximum synchronous compression linear frequency modulation wavelet transforms corresponding to the compressor dynamic pressure signal.

8. A computer device, characterized in that, including: A memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the time-frequency analysis method for the compressor dynamic pressure signal according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to cause a computer to execute the time-frequency analysis method for the compressor dynamic pressure signal according to any one of claims 1 to 6.

10. A computer program product, characterized in that, including computer instructions, and the computer instructions are used to cause a computer to execute the time-frequency analysis method for the compressor dynamic pressure signal according to any one of claims 1 to 6.

Images