Phase modulation machine equipment state online monitoring method, system, terminal and storage medium

Abstract

The application discloses a phase modulator equipment state online monitoring method and system, a terminal and a storage medium. The phase modulator equipment state online monitoring method comprises the following steps: firstly, collecting sound prints and vibration original data in different states of the phase modulator equipment; secondly, realizing accurate mapping of the sound prints and vibration one-dimensional time domain signals based on a high-resolution phase space mapping technology, and constructing two-dimensional atlas matrix elements of the sound prints and vibration signals based on a dynamic time warping method with high stability; then, realizing fusion of the sound-vibration two-dimensional atlas by using a Laplacian pyramid technology; and finally, realizing effective identification of different running states of the phase modulator equipment based on a convolutional neural network. The application realizes real-time monitoring and state perception of the phase modulator equipment state by using a device state perception method with high recognition degree and high reliability, and improves the stability of equipment operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of phase modifier equipment state monitoring, in particular to a phase modifier equipment state online monitoring method and system based on voiceprint vibration, a terminal and a storage medium. BACKGROUND

[0002] As a dynamic reactive power compensation device, the phase modifier equipment has the function of automatically and quickly adjusting reactive power, and stable operation of the equipment can improve the stability of the power system and the power supply quality of the system, and plays an important role in supporting the safe operation of the power grid. However, due to the large size and complex structure of the equipment, early defects of the phase modifier equipment are usually difficult to be found, so how to realize reliable perception of the real-time working state of the equipment is of great significance.

[0003] The traditional state monitoring method mainly collects based on vibration sensors, but this method needs to design and build an independent monitoring system for each type of equipment, and usually a large number of sensors need to be arranged to obtain complete monitoring equipment state parameters, which increases the workload and actual cost. Since abnormal vibration of the equipment will radiate noise outward, and the collection of noise is non-intrusive detection, low in collection cost and high in collection efficiency, it has also been applied in the state detection of power equipment. However, for the phase modifier equipment with complex system structure and complex working environment, it is difficult to realize comprehensive perception of the state parameters of the monitored equipment by using only the detection method of conventional single signal.

[0004] Therefore, it is necessary to provide a phase modifier equipment state online monitoring method which effectively fuses multi-parameter perception signals, obtains rich state parameters of the detected equipment, improves the operation efficiency and reliability of the equipment, reduces the production downtime and maintenance cost, and ultimately achieves the purpose of improving the stability of the power system. SUMMARY

[0005] In view of the above-mentioned shortcomings of the prior art, the purpose of the present application is to provide a phase modifier equipment state monitoring method and system based on voiceprint vibration, which uses an equipment state perception method with high recognition degree and strong reliability to realize real-time monitoring and state perception of the phase modifier equipment, and improves the stability of the equipment operation.

[0006] To achieve the above-mentioned purpose, a technical solution adopted by the present application is as follows: a phase modifier equipment state monitoring method, comprising:

[0007] First, collect voiceprint and vibration original data of the phase modifier equipment in different states;

[0008] Secondly, based on high-resolution phase space mapping technology, accurate mapping of one-dimensional time-domain signals of acoustic text and vibration is achieved, and based on a dynamic time warping method with strong stability, two-dimensional spectrum matrix elements of acoustic text and vibration signals are constructed.

[0009] Then, the fusion of two-dimensional acoustic and vibration spectra is achieved using the Laplace pyramid technique;

[0010] Finally, a convolutional neural network was used to achieve effective identification of different operating states of the camera adjustment device.

[0011] Furthermore, the phase space reconstruction of the one-dimensional time-domain acoustic signature signal x and the one-dimensional time-domain vibration signal y is achieved using phase space mapping technology as follows:

[0012]

[0013] In the formula, X i and Y i Let m and τ represent the reconstructed acoustic and vibration signal vectors, respectively, where m represents the embedding dimension, τ represents the delay time, and N represents the number of discrete points in the time domain signal.

[0014] Furthermore, the two-dimensional spectral matrix elements of the acoustic signature and vibration signal are constructed using a dynamic time warping method with strong stability. The calculation formula is as follows:

[0015]

[0016] In the formula, x i x j Representing the i-th and j-th voiceprint signal data respectively, y i y j These represent the i-th and j-th vibration signal data, respectively.

[0017] Furthermore, the Laplace pyramid technique is used to fuse the two-dimensional acoustic signature and vibration spectrum. The fusion calculation formula is as follows:

[0018] LF N (i,j)=w A LPA N +w B LPB N (3)

[0019] In the formula, LPA N The top layer of the N-layer Laplacian pyramid of the image in channel A; LPB N For the top layer of the N-layer Laplacian pyramid of the channel B image; LF N To merge the top layer of the Laplacian pyramid of the image; w A w B These are the weights for channels A and B, respectively.

[0020] Furthermore, the LeNet-5 convolutional neural network is used to achieve effective recognition under different operating conditions of different camera adjustment devices. This LeNet-5 convolutional neural network is an intelligent self-learning model, including ensemble learning, backpropagation, and selection optimization. The calculation formulas for its convolutional layers and pooling layers are as follows:

[0021]

[0022] In the formula, f() represents the activation function, X i,j Let Y represent the input element in the i-th row and j-th column, σ represent the convolution kernel element, m and n represent the width and height of the convolution kernel respectively, δ represent the error offset, D() represent downsampling, and Y i,j M represents the element in the pooling region, and M represents the coordinate position of the convolution kernel on the input feature map, that is, the spatial position of the filter during the convolution process.

[0023] Another technical solution adopted by the present invention is as follows: an online monitoring system for the status of a synchronous condenser, which includes a data acquisition unit, a data processing unit, and an acoustic-vibration spectrum fusion unit;

[0024] The data acquisition unit is used to acquire the acoustic and vibration signals of the synchronous condenser equipment;

[0025] The data processing unit is used to process the raw acoustic signature and vibration signal data and transmit the data to the acoustic-vibration spectrum fusion unit;

[0026] The aforementioned acoustic-vibration spectrum fusion unit is used to process and calculate acoustic signatures and vibration data to achieve the fusion of two-dimensional acoustic-vibration spectra;

[0027] The specific processing procedure of the acoustic vibration spectrum fusion unit is as follows:

[0028] The phase space mapping technique is used to achieve accurate mapping of one-dimensional time-domain signals of acoustic text and vibration. The dynamic time warping method is used to construct two-dimensional spectrum matrix elements of acoustic text and vibration. The Laplace pyramid technique is used to achieve the fusion of two-dimensional acoustic and vibration images to obtain a fused image that can fully represent the current state of the camera condenser. Finally, the convolutional neural network is used to achieve intelligent identification of the camera condenser's equipment state.

[0029] Furthermore, the online monitoring system for the status of the synchronous condenser equipment also includes a result display unit, which is used to display the acoustic-vibration fusion spectrum of the synchronous condenser equipment under different states and the output of status identification / diagnosis results.

[0030] Furthermore, the data processing unit includes: collecting the acoustic and vibration signals obtained by the data acquisition unit and amplifying the signals, performing ADC acquisition, and storing the signals.

[0031] Corresponding to the online monitoring method for the status of a synchronous condenser device of the present invention, the present invention also provides a computer monitoring terminal, which includes: one or more processors; a memory coupled to the processors for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the online monitoring method for the status of a synchronous condenser device as described above.

[0032] Corresponding to the online monitoring method for the status of a synchronous condenser device of the present invention, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the online monitoring method for the status of a synchronous condenser device as described above.

[0033] Compared with existing conventional technologies, this invention has the following advantages: The online monitoring method for the status of a phase-shifting camera based on acoustic signature vibration proposed in this invention overcomes the shortcomings of conventional monitoring methods in comprehensively sensing the monitoring equipment, and can effectively sense the degree of change in the working status of the phase-shifting camera. It fully mines the potential state parameter information of the one-dimensional time-domain acoustic and vibration data of the phase-shifting camera using high-precision phase space reconstruction technology, constructs two-dimensional visualization matrix elements of acoustic signature and vibration using stable dynamic time warping technology, and fully fuses the two-dimensional acoustic and vibration images using Laplace pyramid technology, ultimately obtaining a fused image that can fully represent the current state of the phase-shifting camera. The method described in this invention is computationally efficient and easy to implement. Based on obtaining a fused image that can fully represent the current working state of the phase-shifting camera, a convolutional neural network is used to finally achieve reliable identification of the phase-shifting camera's status.

[0034] In summary, this invention is computationally efficient, easy to implement, and provides intuitive results. The resulting fusion map contains rich information on the status of the synchronous condenser equipment, providing reliable data support for status identification of the synchronous condenser equipment under different operating conditions.

[0035] Therefore, the online monitoring system, terminal and storage medium for synchronous condenser equipment status of the present invention also have the above-mentioned beneficial effects, which will not be repeated here. Attached Figure Description

[0036] Figure 1 This is a flowchart illustrating the online monitoring method for the status of a synchronous condenser device according to the present invention.

[0037] Figure 2 This is a schematic diagram of the online monitoring system for the status of the synchronous condenser equipment of the present invention;

[0038] Figure 3 This is a schematic diagram of the structure of the online monitoring terminal for the status of the synchronous condenser equipment of the present invention;

[0039] Figure 4This is a schematic diagram of a two-dimensional image of the camera in normal-fusion state according to the present invention;

[0040] Figure 5 This is a schematic diagram of a two-dimensional image under the abnormal-fusion state of the camera adjustment device according to the present invention;

[0041] Figure 6 The figure shows the performance curve of the method described in this invention during the training process using actual test data.

[0042] Figure 7 This is a schematic diagram of the confusion matrix calculation results obtained using the method described in this invention. Detailed Implementation

[0043] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification.

[0044] It should be noted that the structures, proportions, sizes, etc., illustrated in the accompanying drawings of this specification are only used to complement the content disclosed in the specification for those skilled in the art to understand and read, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0045] Example

[0046] This embodiment provides a method, system, terminal, and storage medium for online monitoring of the status of a synchronous condenser based on acoustic vibration.

[0047] Firstly, a method for online monitoring of the status of a camera adjustment device is provided, such as... Figure 1 As shown, the steps are as follows:

[0048] Step 1: Collect raw acoustic and vibration data of the synchronous condenser equipment in different states;

[0049] Step 2: Accurate mapping of acoustic signature and vibration one-dimensional time domain signals is achieved based on high-resolution phase space mapping technology, and two-dimensional spectrum matrix elements of acoustic signature and vibration signals are constructed based on a dynamic time warping method with strong stability.

[0050] Step 3: Use the Laplace pyramid technique to fuse the two-dimensional acoustic and vibration spectra;

[0051] Step 4: Effectively identify different operating states of the camera adjustment device based on convolutional neural networks.

[0052] Phase space mapping technology is used to reconstruct the phase space of the one-dimensional time-domain acoustic signature signal x and the one-dimensional time-domain vibration signal y as follows:

[0053]

[0054] In the formula, X i and Y i Let m and τ represent the reconstructed acoustic and vibration signal vectors, respectively, where m represents the embedding dimension, τ represents the delay time, and N represents the number of discrete points in the time domain signal.

[0055] Two-dimensional spectral matrix elements of acoustic and vibration signals are constructed using a highly stable dynamic time warping method, and the calculation formula is as follows:

[0056]

[0057] In the formula, x i x j Representing the i-th and j-th voiceprint signal data respectively, y i y j These represent the i-th and j-th vibration signal data, respectively.

[0058] The fusion of acoustic signature and vibration two-dimensional spectra is achieved using the Laplace pyramid technique, and the fusion calculation formula is as follows:

[0059] LF N (i,j)=w A LPA N +w B LPB N (3)

[0060] In the formula, LPA N The top layer of the N-layer Laplacian pyramid of the image in channel A; LPB N For the top layer of the N-layer Laplacian pyramid of the channel B image; LF N To merge the top layer of the Laplacian pyramid of the image; w A w B These are the weights for channels A and B, respectively.

[0061] This paper utilizes a LeNet-5 convolutional neural network to achieve effective recognition under different operating conditions of various camera adjustment devices. The LeNet-5 convolutional neural network is an intelligent self-learning model, incorporating ensemble learning, backpropagation, and selection optimization. The calculation formulas for its convolutional and pooling layers are as follows:

[0062]

[0063] In the formula, f() represents the activation function, X i,j Let Y represent the input element in the i-th row and j-th column, σ represent the convolution kernel element, m and n represent the width and height of the convolution kernel respectively, δ represent the error offset, D() represent downsampling, and Y i,jM represents the element in the pooling region, and M represents the coordinate position of the convolution kernel on the input feature map, that is, the spatial position of the filter during the convolution process.

[0064] Secondly, an online monitoring system for the status of a synchronous condenser based on acoustic vibration is provided, comprising a data acquisition unit, a data processing unit, an acoustic vibration spectrum fusion unit, and a result display unit; the system device framework is shown in the figure. Figure 2 .

[0065] The data acquisition unit is used to acquire acoustic and vibration signals from the synchronous condenser equipment.

[0066] The data processing unit is used to process raw acoustic and vibration signal data, specifically including collecting acoustic and vibration signals obtained by the data acquisition unit and amplifying the signals, ADC acquisition and signal storage, and transmitting the data to the acoustic-vibration spectrum fusion unit.

[0067] The aforementioned acoustic-vibration spectrum fusion unit is used to process and calculate acoustic patterns and vibration data to achieve the fusion of two-dimensional acoustic-vibration spectra.

[0068] The specific processing procedure of the acoustic vibration spectrum fusion unit is as follows:

[0069] The phase space mapping technique is used to achieve accurate mapping of one-dimensional time-domain signals of acoustic text and vibration. The dynamic time warping method is used to construct two-dimensional spectrum matrix elements of acoustic text and vibration. The Laplace pyramid technique is used to achieve the fusion of two-dimensional acoustic and vibration images to obtain a fused image that can fully represent the current state of the camera condenser. Finally, the convolutional neural network is used to achieve intelligent identification of the camera condenser's equipment state.

[0070] The result display unit is used to display the acoustic-vibration fusion spectrum of the phase-shifting device under different states, as well as the output of state recognition / diagnosis results.

[0071] Thirdly, the present invention also provides a computer monitoring terminal, specifically comprising: one or more processors; a memory coupled to the processors for storing one or more programs; and when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the online monitoring method for the status of a synchronous condenser device as described in any of the preceding claims.

[0072] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the online monitoring method for the status of a synchronous condenser device as described in any of the preceding claims.

[0073] The simplified diagrams for the third and fourth aspects are shown below. Figure 3 .

[0074] Application examples

[0075] To highlight the superiority of the method of this invention, a set of measured data was used, which included both normal and abnormal states of the synchronous condenser equipment. The online monitoring method for the state of the synchronous condenser equipment of this invention was applied for calculation, and the resulting fused two-dimensional image under the two typical states is shown below. Figure 4 and 5 As shown.

[0076] from Figure 4 and Figure 5 As can be seen, the two-dimensional fused images obtained using the method of this invention under normal and abnormal conditions have obvious discriminative power. When the camera condenser is in normal operation, the extreme values ​​of the fused image exhibit a "linear" distribution; while the fused image under abnormal conditions shows a significant point-like diffusion pattern, with a more widespread distribution of extreme values.

[0077] Its performance curve during training is as follows Figure 6 As shown in the figure, the loss function value tends to stabilize after 50 iterations, indicating that the model has good classification performance.

[0078] A confusion matrix is ​​introduced to quantitatively evaluate the classification effect of the online monitoring method for the status of the synchronous condenser equipment, such as... Figure 7 The figure shows the result of calculating the confusion matrix using the method described in this invention. The diagonals of the confusion matrix represent the correctly identified samples for each state using the method described in this invention. Figure 7 As can be seen, only 2 out of the 111 test cases were misclassified, resulting in a classification accuracy of 98.20% for the test sample set. Table 1 shows that the state recognition precision, recall, and F1 score of the method described in this invention are 97.67%, 97.67%, and 97.67%, respectively. This result demonstrates that the method described in this invention possesses excellent classification performance, proving the effectiveness of the invention.

[0079] Table 1 Evaluation Parameters

[0080] Parameter / label 1 2 Mean Precision 98.78％ 96.55％ 97.67％ Recall 98.78％ 96.55％ 97.67％ F1 value 98.78％ 96.55％ 97.67％

[0081] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for online monitoring of the status of a camera device, characterized in that, include: First, raw acoustic and vibration data of the synchronous condenser equipment under different states are collected; Secondly, based on high-resolution phase space mapping technology, accurate mapping of one-dimensional time-domain signals of acoustic text and vibration is achieved, and based on a dynamic time warping method with strong stability, two-dimensional spectrum matrix elements of acoustic text and vibration signals are constructed. Then, the fusion of two-dimensional acoustic and vibration spectra is achieved using the Laplace pyramid technique; Finally, a convolutional neural network was used to achieve effective identification of different operating states of the camera adjustment device; Phase space mapping technology is used to reconstruct the phase space of the one-dimensional time-domain acoustic signature signal x and the one-dimensional time-domain vibration signal y as follows: In the formula, X i and Y i Let m and τ represent the reconstructed acoustic and vibration signal vectors, respectively, where m represents the embedding dimension, τ represents the delay time, and N represents the number of discrete points in the time-domain signal. Two-dimensional spectral matrix elements of acoustic and vibration signals are constructed using a highly stable dynamic time warping method, and the calculation formula is as follows: In the formula, x i x j Representing the i-th and j-th voiceprint signal data respectively, y i y j These represent the i-th and j-th vibration signal data, respectively.

2. The online monitoring method for the status of a synchronous condenser according to claim 1, characterized in that, The fusion of acoustic signature and vibration two-dimensional spectra is achieved using the Laplace pyramid technique, and the fusion calculation formula is as follows: LF N (i,j)=w A LPA N +w B LPB N (3) In the formula, LPA N The top layer of the N-layer Laplacian pyramid of the image in channel A; LPB N For the top layer of the N-layer Laplacian pyramid of the channel B image; LF N To merge the top layer of the Laplacian pyramid of the image; w A w B These are the weights for channels A and B, respectively.

3. The online monitoring method for the status of a synchronous condenser according to claim 1, characterized in that, This paper utilizes a LeNet-5 convolutional neural network to achieve effective recognition under different operating conditions of various camera adjustment devices. The LeNet-5 convolutional neural network is an intelligent self-learning model, incorporating ensemble learning, backpropagation, and selection optimization. The calculation formulas for its convolutional and pooling layers are as follows: In the formula, f() represents the activation function, X i,j Let Y represent the input element in the i-th row and j-th column, σ represent the convolution kernel element, m and n represent the width and height of the convolution kernel respectively, δ represent the error offset, D() represent downsampling, and Y i,j The pooling region represents the element, and M represents the coordinate position of the convolution kernel on the input feature map.

4. An online monitoring system for the status of a synchronous condenser, used to implement the online monitoring method for the status of a synchronous condenser as described in any one of claims 1-3, characterized in that, It includes a data acquisition unit, a data processing unit, and an acoustic-vibration spectrum fusion unit; The data acquisition unit is used to acquire the acoustic and vibration signals of the synchronous condenser equipment; The data processing unit is used to process the raw acoustic signature and vibration signal data and transmit the data to the acoustic-vibration spectrum fusion unit; The aforementioned acoustic-vibration spectrum fusion unit is used to process and calculate acoustic signatures and vibration data to achieve the fusion of two-dimensional acoustic-vibration spectra; The specific processing procedure of the acoustic vibration spectrum fusion unit is as follows: The phase space mapping technique is used to achieve accurate mapping of one-dimensional time-domain signals of acoustic text and vibration. The dynamic time warping method is used to construct two-dimensional spectrum matrix elements of acoustic text and vibration. The Laplace pyramid technique is used to achieve the fusion of two-dimensional acoustic and vibration images to obtain a fused image that can fully represent the current state of the camera condenser. Finally, the convolutional neural network is used to achieve intelligent identification of the camera condenser's equipment state.

5. The online monitoring system for the status of a synchronous condenser according to claim 4, characterized in that, It also includes a result display unit, which is used to display the acoustic-vibration fusion spectrum of different states of the phase-shifting device and the output of state recognition / diagnosis results.

6. The online monitoring system for the status of synchronous condenser equipment according to claim 4, characterized in that, The data processing unit includes: collecting the acoustic and vibration signals obtained by the data acquisition unit and amplifying the signals, performing ADC acquisition, and storing the signals.

7. A computer monitoring terminal, characterized in that, One or more processors; A memory, coupled to the processor, is used to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the online monitoring method for the status of a camera shifter as described in any one of claims 1-3.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the online monitoring method for the status of the camera adjustment device as described in any one of claims 1-3.

Images