Water hammer early warning method and device and storage medium
By deploying acoustic emission sensors in the water pipe system to obtain the spectral map and using Mahastellar distance for water hammer detection, the problem that water hammer detection in the prior art is difficult to identify multi-dimensional features, and early warning and accuracy of water hammer events are achieved Identification improves the adaptability and scalability of the early warning system.
Patent Information
- Application Number
- CN202411090877.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-12-10
AI Technical Summary
The existing water hammer detection methods lack the accurate identification ability of multi-dimensional features, and it is difficult to provide effective early warning before the water hammer occurs, especially in complex environments that may lead to false alarms or missed reports.
By deploying acoustic emission sensors at key nodes of the water pipe system, acoustic wave signals are obtained and converted to obtain a spectral map, key indicators are extracted from them, and early detection and early warning of water hammers are used using Mahayana distance.
It realizes early warning of water hammer events, breaks through the limitations of traditional monitoring technology, can identify abnormal voiceprint patterns before water hammer occurs, wins time for preventive measures, and improves detection through multi-eigenvector analysis of Mahastellar distance Accuracy and adaptability.
Smart Images
Figure CN119107975A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of industrial fault diagnosis, and in particular to a water hammer early warning method, equipment and storage medium. Background Art
[0002] Water hammer refers to the phenomenon of rapid pressure fluctuations caused by sudden changes in flow rate in closed or semi-closed fluid transmission systems (such as water pipes and hydraulic pipelines). This phenomenon not only causes serious damage to the pipeline structure, such as rupture and leakage, but may also trigger a chain reaction, affecting the stable operation of the entire water conservancy facility and even endangering public safety.
[0003] Traditionally, the monitoring and response of water hammer events mostly rely on physical quantity monitoring such as pressure sensors and vibration sensors, as well as post-analysis technology. Although these methods can record and analyze water hammer events to a certain extent, they often lack sufficient early warning capabilities, making it difficult to take effective measures before water hammer occurs to reduce losses. In addition, data analysis from a single sensor may be difficult to accurately judge complex water hammer phenomena due to limited information, especially the lack of ability to identify small changes in the early stages.
[0004] For example, Chinese invention patent CN202310489831.9 discloses a water hammer detection method, including: S1, collecting noise information to determine the time information of water hammer occurrence; S2, collecting pipeline vibration speed information during the water hammer occurrence period; S3, according to the water hammer occurrence time and pipeline vibration speed information, combined with equipment information, classify and grade the output of water hammer alarm information to achieve water hammer detection. The present invention can achieve real-time water hammer detection and alarm through real-time detection of pipeline noise and vibration information on the operation line, reducing a lot of labor costs, and can quickly and efficiently achieve water hammer detection; the present invention classifies and grades water hammer alarm information according to factors such as the intensity of water hammer pipeline action and the cause of water hammer, so that the staff can make quick decisions and formulate water hammer elimination strategies based on the alarm information, thereby reducing the negative impact of water hammer.
[0005] This method mentions classifying and grading water hammer alarm information based on factors such as the intensity of water hammer pipe action and the cause of water hammer. However, relying solely on pipeline noise and vibration information may not accurately determine the specific cause and intensity of water hammer, because this requires complex signal processing and pattern recognition technology. In addition, although this method emphasizes real-time detection, real-time processing of large amounts of noise and vibration data may place high demands on the processing system, especially when a quick response is required. If data processing is not fast enough or system resources are limited, it may cause alarm delays, thereby affecting real-time performance. Finally, in some special cases (such as complex internal structure of the pipeline, interference from multiple noise sources, etc.), it may be difficult to accurately detect water hammer relying solely on noise and vibration information. This method may lack countermeasures for such special situations, resulting in false alarms or missed alarms. Summary of the invention
[0006] The purpose of the present invention is to propose a water hammer early warning method, device and storage medium to solve the technical problem that the existing traditional water hammer detection method cannot accurately identify anomalies using multi-dimensional features.
[0007] The present invention provides a water hammer early warning method, comprising the following steps:
[0008] S1: Deploy acoustic emission sensors at key nodes of the water pipe system to obtain acoustic wave signals;
[0009] S2: converting and processing the sound wave signal to obtain a spectrogram;
[0010] S3: Extract key indicators for fault assessment from the spectrogram;
[0011] S4: Using Mahalanobis distance, early detection and warning of water hammer are completed based on key indicators.
[0012] A storage medium stores instructions and data for implementing an early warning method for water hammer.
[0013] A water hammer early warning device comprises: a processor and the storage medium; the processor loads and executes instructions and data in the storage medium to implement a water hammer early warning method.
[0014] The beneficial effects provided by the present invention are:
[0015] 1. Application in water hammer warning: Traditionally, the monitoring of water hammer events mostly relies on sound pressure or vibration signals, and often focuses on post-analysis rather than warning. This invention innovatively introduces voiceprint technology to achieve early warning of water hammer events by deeply analyzing the changing trend of the spectrogram characteristics of the acoustic emission signal. This method breaks through the limitations of traditional monitoring technology and can identify abnormal voiceprint patterns before water hammer occurs, thus buying valuable time for the implementation of preventive measures.
[0016] 2. Innovative application of Mahalanobis distance in multi-feature vector analysis: Although the Mahalanobis distance itself is not a new concept, this invention cleverly extends it from univariate anomaly detection to the field of multi-feature vector analysis. By using the Mahalanobis distance or similar statistics to identify outliers in acoustic emission signals, this technology provides a new perspective and method for complex signal processing. This expansion not only enriches the application scenarios of the Mahalanobis distance, but also opens up a new path for dealing with anomaly detection problems in high-dimensional data sets.
[0017] 3. Highly adaptable and configurable early warning system design: The flexibility of the early warning system is another major innovation. The system design allows the feature extraction method and Mahalanobis distance parameters to be adjusted according to different hydraulic system characteristics and actual operating environment. This design greatly improves the adaptability and scalability of the technical solution. This means that the technology can be customized and applied to a wide range of hydraulic engineering environments to meet the early warning needs under different conditions, demonstrating its strong potential and value in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 It is a schematic flow chart of the method of the present invention;
[0019] Figure 2 It is a spectrogram of an embodiment of the present invention;
[0020] Figure 3 It is a working schematic diagram of the hardware device of the present invention. DETAILED DESCRIPTION
[0021] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described below with reference to the accompanying drawings.
[0022] Before formally describing the present invention, the scheme of the present invention is first generally described for easy understanding.
[0023] Please refer to Figure 1 , Figure 1 It is a schematic flow diagram of the method of the present invention;
[0024] The present invention provides a water hammer early warning method, comprising the following steps:
[0025] S1: Deploy acoustic emission sensors at key nodes of the water pipe system to obtain acoustic wave signals;
[0026] It should be noted that the present invention uses a high-sensitivity acoustic emission sensor to collect the acoustic emission signal x(t) in real time at key locations of the hydraulic system (such as pipelines, pump rooms, etc.).
[0027] S2: converting and processing the sound wave signal to obtain a spectrogram;
[0028] It should be noted that step S2 is specifically as follows:
[0029] S21, performing short-time Fourier transform on the sound wave signal to obtain a spectrum;
[0030] As an example, for a given acoustic signal x(t), its short-time Fourier transform (STFT) s(t,f) is defined as:
[0031] ①
[0032] Where: x(t) is the original signal, ω(τ) is the window function, f is the frequency, and t is the time.
[0033] S22, converting the linear frequency scale in the spectrum into the Mel scale through a Mel filter bank;
[0034] As an embodiment, the conversion relationship between the linear frequency scale and the Mel scale in step S22 is as follows:
[0035]
[0036] Where m is the Mel value and f is the frequency.
[0037] S23, according to the Mel scale, accumulating the spectrum energy within the frequency range covered by each Mel filter in the Mel filter bank to obtain a Mel spectrum;
[0038] As an example, in step S23, the expression of the mel spectrum is as follows:
[0039]
[0040] The Mel spectrum m(t, k) is obtained by passing the short-time Fourier transform spectrum |S(t, f)| through the Mel filter bank and summing them up. It should be noted that the Mel filter bank consists of a series of triangular filters, and the center frequency of each filter is equally spaced on the Mel scale. k (f) is defined as:
[0041]
[0042] where f k is the center frequency of the kth Mel filter.
[0043] S24. Obtain a spectrogram based on the mel spectrum.
[0044] It should be noted that the spectrogram is s(t,f) 2 For a visual representation of Figure 2 , Figure 2 It is an embodiment of the spectrogram in the present invention. Figure 2 In the figure, the upper part is the sound signal x(t), and the lower part is the spectrogram, which is |s(t,f) 2 , where the color depth indicates the energy level.
[0045] S3: Extract key indicators for fault assessment from the spectrogram;
[0046] It should be noted that common features of spectrograms include energy distribution, resonance peaks, fundamental frequency, harmonic structure, noise and interference.
[0047] The above common characteristic indicators can be used for fault assessment, but the key indicators used in the present invention are the voiceprint deviation vector and the standard deviation vector.
[0048] As other extended embodiments, the present invention specifically provides several characteristic calculation methods that are strongly correlated with water hammer noise.
[0049] For example, the energy signature:
[0050] For the kth Mel band, its energy characteristic E k Defined as:
[0051]
[0052] Where T is the total number of time frames.
[0053] For example, peak frequency and rate of change of frequency:
[0054] Peak frequency f peak (t) is the center frequency of the frequency band with the highest energy in the Mel spectrum at time t:
[0055]
[0056] The frequency change rate Δf(t) is the rate of change of the frequency peak over time: where Δt is the interval between time frames.
[0057] The above embodiments are for reference only. As a preferred embodiment, the calculation method of the standard deviation vector in the present invention is as follows:
[0058] Eigenvector X i It can be expressed as:
[0059]
[0060] Where E 1 ,E 2 ,....E N It is energy in different frequency bands.
[0061] S4: Using Mahalanobis distance, early detection and warning of water hammer are completed based on key indicators.
[0062] Step S4 is specifically as follows:
[0063] S41, calculating the Mahalanobis distance according to the voiceprint feature vector;
[0064] As an embodiment, the voiceprint feature vector is used to calculate the Mahalanobis distance: D M (X i )
[0065]
[0066] in: Eigenvector X i The mean of Σ eigenvector X i The covariance matrix of .
[0067] S42. If the Mahalanobis distance is greater than a preset critical value, the corresponding feature is deemed to be abnormal, and an early warning of water hammer is issued.
[0068] As an example, according to a preset significance level α (usually 0.05 or 0.01), the critical value D of the Mahalanobis distance is found. Mcrit This critical value can usually be obtained from a statistical table or calculated using a specific distribution function.
[0069] If the calculated Mahalanobis distance D M (X i ) is greater than the critical value D Mcrit :
[0070] D M (X i )>D Mcrit (10)
[0071] The i-th feature vector is considered abnormal. In the water hammer warning system, this anomaly may indicate an early sign of a water hammer event. In this way, the Mahalanobis distance is applied to anomaly detection in multidimensional feature space, effectively identifying those soundprint patterns that deviate significantly from the normal pattern, providing key decision support for the water hammer warning system.
[0072] See also Figure 3 , Figure 3 4 is a schematic diagram of the working of the hardware device of an embodiment of the present invention, wherein the hardware device specifically comprises: a water hammer early warning device 401, a processor 402 and a storage medium 403.
[0073] A water hammer early warning device 401: The water hammer early warning device 401 implements the water hammer early warning method.
[0074] Processor 402: The processor 402 loads and executes instructions and data in the storage medium 403 to implement the water hammer early warning method.
[0075] Storage medium 403: The storage medium 403 stores instructions and data; the storage medium 403 is used to implement the water hammer early warning method.
[0076] The beneficial effects of the present invention are:
[0077] 1. Application in water hammer early warning: Traditionally, the monitoring of water hammer events mostly relies on sound pressure or vibration signals, and often focuses on post-event analysis rather than early warning. This invention innovatively introduces voiceprint technology to achieve early warning of water hammer events by deeply analyzing the changing trend of the spectrogram characteristics of the acoustic emission signal. This method breaks through the limitations of traditional monitoring technology and can identify abnormal voiceprint patterns before water hammer occurs, thus buying valuable time for the implementation of preventive measures.
[0078] 2. Innovative application of Mahalanobis distance in multi-feature vector analysis: Although the Mahalanobis distance itself is not a new concept, this invention cleverly extends it from univariate anomaly detection to the field of multi-feature vector analysis. By using the Mahalanobis distance or similar statistics to identify outliers in acoustic emission signals, this technology provides a new perspective and method for complex signal processing. This expansion not only enriches the application scenarios of the Mahalanobis distance, but also opens up a new path for dealing with anomaly detection problems in high-dimensional data sets.
[0079] 3. Highly adaptable and configurable early warning system design: The flexibility of the early warning system is another major innovation. The system design allows the feature extraction method and Mahalanobis distance parameters to be adjusted according to different hydraulic system characteristics and actual operating environment. This design greatly improves the adaptability and scalability of the technical solution. This means that the technology can be customized and applied to a wide range of hydraulic engineering environments to meet the early warning needs under different conditions, demonstrating its strong potential and value in practical applications.
[0080] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A water hammer early warning method, characterized in that: The method comprises the following steps: S1: Deploy acoustic emission sensors at key nodes of the water pipe system to obtain acoustic wave signals; S2: converting and processing the sound wave signal to obtain a spectrogram; S3: Extract key indicators for fault assessment from the spectrogram; S4: Using Mahalanobis distance, early detection and warning of water hammer are completed based on key indicators.
2. A water hammer early warning method according to claim 1, characterized in that: Step S2 is specifically as follows: S21, performing short-time Fourier transform on the sound wave signal to obtain a spectrum; S22, converting the linear frequency scale in the spectrum into the Mel scale through a Mel filter bank; S23, according to the Mel scale, accumulating the spectrum energy within the frequency range covered by each Mel filter in the Mel filter bank to obtain a Mel spectrum; S24. Obtain a spectrogram based on the mel spectrum.
3. A water hammer early warning method as claimed in claim 2, characterized in that: The spectrum in step S21 is as follows: Where: x(t) is the original signal, ω(τ) is the window function, f is the frequency, and t is the time.
4. A water hammer early warning method as claimed in claim 3, characterized in that: The conversion relationship between the linear frequency scale and the Mel scale in step S22 is as follows: Where m is the Mel value and f is the frequency.
5. A water hammer early warning method as claimed in claim 4, characterized in that: In step S23, the expression of the mel spectrum is as follows: The Mel spectrum m(t, k) is obtained by passing the short-time Fourier transform spectrum |S(t, f)| through the Mel filter bank and summing them up; H k (f) is defined as: where f k is the center frequency of the kth Mel filter.
6. A water hammer early warning method as claimed in claim 5, characterized in that: The spectrogram in step S24 is |s(t,f) 2 Visual representation of .
7. A water hammer early warning method as claimed in claim 1, characterized in that: The key indicators in step S3 include: voiceprint deviation vector and standard deviation vector.
8. A water hammer early warning method as claimed in claim 7, characterized in that: Step S4 is specifically as follows: S41, calculating the Mahalanobis distance according to the voiceprint deviation vector and the standard deviation vector; S42. If the Mahalanobis distance is greater than a preset critical value, the corresponding feature is deemed to be abnormal, and an early warning of water hammer is issued.
9. A storage medium, characterized in that: The storage medium stores instructions and data for implementing a water hammer early warning method as described in any one of claims 1 to 8.
10. A water hammer early warning device, characterized in that: include: Processor and storage medium; the processor loads and executes instructions and data in the storage medium to implement a water hammer early warning method as described in any one of claims 1 to 8.