Generator set carbon brush intelligent holographic monitoring method and system

Abstract

The invention discloses an intelligent holographic monitoring method and system for a generator set carbon brush, and relates to the technical field of intelligent monitoring and fault diagnosis, and the method comprises the steps: carrying out the multi-angle temperature measurement of the generator set carbon brush and a collector ring through a high-precision infrared thermal imaging technology, and obtaining three-dimensional temperature distribution data; shaft vibration data are processed through an adaptive filtering algorithm, online measurement is carried out on the abrasion loss of the carbon brush through a deep learning technology, abrasion data are obtained in combination with a morphological processing algorithm, multi-modal data fusion processing is carried out on all the obtained data, and a machine learning technology based on a neural network is constructed; and processing the fused data by using a machine learning technology, and carrying out intelligent early warning on the state of the carbon brush through feature selection and model optimization. The invention provides a set of omnibearing monitoring system, which not only improves the safety and reliability of the operation of the generator set, but also improves the maintenance efficiency, reduces the labor cost, and realizes the double improvement of the technical universality and practicability.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring and fault diagnosis, and in particular to a method and system for intelligent holographic monitoring of carbon brushes of a generator set. Background Art

[0002] Large synchronous generator sets usually use carbon brushes to introduce excitation current into the rotor, so that the generator rotor establishes a constant magnetic field. Only by passing a large current of several hundred or even thousands of amperes between the carbon brushes and the collector rings can the motor provide sufficient power to the outside. Therefore, carbon brushes are key components for the generator to generate electricity. At present, large synchronous generators mostly use sliding mechanical contact guide rings between carbon brushes and collector rings. Due to long-term mechanical wear and equipment loss, jitter, heat and other problems, carbon brushes are used as wear parts. Not only do they require regular replacement, but they often cause carbon brush ignition and combustion due to various reasons. In serious cases, the collector ring and slip ring chamber may catch fire, and the generator set will be forced to stop, stopping the power production process and affecting the safety and stability of the power industry. Large hydro-turbine generators have large excitation currents and frequent starts and stops. Once the carbon brush current is unevenly distributed, the above problems are more likely to occur.

[0003] With the development of large-scale hydroelectric generating units in China, there is a new opportunity for refined management of generating units, making the research and development of auxiliary supporting technologies for excitation systems a primary condition, which puts higher requirements on the stable operation of carbon brushes and slip rings. Whether from the perspective of safe operation of hydroelectric generating units or reducing downtime and improving power generation efficiency, the research on intelligent holographic monitoring system for carbon brushes of generating units is of great significance. Summary of the invention

[0004] In view of the above existing problems, the present invention provides a method and system for intelligent holographic monitoring of carbon brushes of a generator set, so as to solve the problems of low efficiency of manual inspection and inability to achieve real-time monitoring in the prior art.

[0005] In order to solve the above technical problems, a method for intelligent holographic monitoring of carbon brushes of generator sets is proposed, including:

[0006] High-precision infrared thermal imaging technology is used to perform multi-angle temperature measurement on the carbon brushes and collector rings of the generator set to obtain three-dimensional temperature distribution data. Sensors are used to measure the swing of the carbon brush position of the generator set, and the shaft vibration data is processed through an adaptive filtering algorithm. The wear of the carbon brush is measured online through deep learning-assisted computer vision measurement technology, and the wear data is obtained in combination with a morphological processing algorithm. All the obtained data is then subjected to multimodal data fusion processing. A machine learning technology based on a neural network is constructed, and the fused data is processed using machine learning technology, and intelligent early warning of the carbon brush status is provided through feature selection and model optimization.

[0007] As a preferred solution of the method for intelligent holographic monitoring of carbon brushes of a generator set described in the present invention, the acquisition of three-dimensional temperature distribution data includes using high-precision infrared thermal imaging technology to perform multi-angle temperature measurement on the carbon brushes and collector rings of the generator set to obtain three-dimensional temperature distribution data.

[0008] The multi-angle temperature measurement includes performing a safety inspection on the generator set, selecting an infrared thermal imager lens and a measuring distance according to the specific structure and position of the generator set's carbon brushes and collector rings, and calibrating the infrared thermal imager, planning the measurement angle and path according to the specific layout of the generator set, and performing multi-angle temperature measurement on the generator set's carbon brushes and collector rings using an infrared thermal imager according to the planned measurement angles and paths, and collecting and storing image data at each angle in real time.

[0009] The three-dimensional temperature includes importing the collected multi-angle temperature images into thermal image analysis software, performing image processing and three-dimensional reconstruction, analyzing and verifying the reconstructed three-dimensional temperature distribution data, and compiling a temperature measurement report based on the analysis results.

[0010] The image processing includes removing noise, enhancing contrast and correcting lens distortion; the analysis of three-dimensional temperature distribution includes analyzing uniformity of temperature distribution, hot spot identification and temperature gradient.

[0011] As a preferred solution of the method for intelligent holographic monitoring of carbon brushes of a generator set described in the present invention, the measurement of the swing of the carbon brush position of the generator set includes calibrating the sensor, collecting the swing data of the carbon brush position of the generator set, and preprocessing the collected data.

[0012] The adaptive filtering algorithm includes using the least mean square algorithm according to the characteristics of the shaft vibration data, implementing the algorithm through programming, filtering the pre-processed data, analyzing the filtering effect, extracting the characteristic parameters of the shaft vibration, and evaluating the operating status of the generator set carbon brush position.

[0013] The least mean square algorithm formula is expressed as:

[0014] w(n+1)=w(n)+μe(n)x(n)

[0015] e(n)=d(n)-y(n)

[0016] Where w(n) is the filter weight vector at time n, μ is the step size factor, e(n) is the error signal at time n, d(n) is the desired signal, y(n) is the filter output, x(n) is the input signal vector at time n, and n is the discrete time index.

[0017] As a preferred solution of the method for intelligent holographic monitoring of carbon brushes of a generator set described in the present invention, the online measurement of the wear of the carbon brushes includes: using a high-definition camera to monitor the carbon brushes of the generator set in real time, adjusting the position and focal length of the camera, and graying the collected image, using Gaussian filtering to remove image noise, using an edge detection algorithm to extract edge information of the carbon brush, and using a deep learning framework to build a convolutional neural network model, train the CNN model, and use the trained CNN model to infer the preprocessed image, identify the wear area, perform morphological processing and calculate the wear amount.

[0018] The morphological processing formula is:

[0019]

[0020] Among them, A is the original image, B is the structural element, For corrosion operations, For expansion operation.

[0021] The wear calculation formula is:

[0022]

[0023] Where S is the area of ​​the worn area, R is the pixel set in the worn area, and (x, y) is the pixel coordinate in the worn area.

[0024] As a preferred solution of the method for intelligent holographic monitoring of carbon brushes of a generator set described in the present invention, the multimodal data fusion includes performing feature-level fusion and decision-level fusion processing on the obtained three-dimensional temperature distribution data, shaft vibration data and high-precision wear data.

[0025] The feature-level fusion includes using linear discriminant analysis to reduce the dimension of the features of the data source, and merging the reduced features to form a comprehensive feature vector; the decision-level fusion includes using a support vector machine model to train the comprehensive feature vector and generate a prediction result, and using a weighted summation mechanism to merge the output of the model to obtain a final decision.

[0026] As a preferred solution of the method for intelligent holographic monitoring of carbon brushes of generator sets described in the present invention, the construction of a neural network-based machine learning technology includes dividing the fused data set into a training set, a validation set and a test set, designing a neural network architecture, determining the number of neurons and activation functions in each layer, constructing a machine learning model, and performing model training.

[0027] The model training includes using a back propagation algorithm and a gradient descent method to perform model training, and using a validation set to adjust model parameters.

[0028] As a preferred solution of the method for intelligent holographic monitoring of carbon brushes of generator sets described in the present invention, the intelligent early warning of the carbon brush status includes building a machine learning technology based on neural networks, and using machine learning technology to process the fused data, and calculating the risk coefficient according to business needs and model performance, setting a warning threshold and dividing the prediction results into different areas according to the warning threshold, and providing an intelligent early warning of the carbon brush status through feature selection and model optimization.

[0029] The setting of the early warning threshold includes setting the safety threshold to 0.1, setting the warning threshold to 0.5, and setting the danger threshold to 1; when F(θ,X)<0.1, it indicates that the carbon brush is in normal condition, with no signs of wear, and routine monitoring is maintained; when 0.1≤F(θ,X)<0.5, it indicates that the carbon brush is in abnormal condition, and attention should be paid to monitoring, regular inspections should be performed, the monitoring frequency should be increased, and a maintenance plan should be prepared; when 0.5≤F(θ,X)<1, it indicates that the carbon brush is in worn condition, and a detailed inspection should be carried out immediately, preventive maintenance measures should be implemented, and operating parameters should be adjusted; when F(θ,X)≥1, it indicates that the carbon brush is in dangerous condition and has caused damage to the generator set, and the generator set should be shut down immediately for a comprehensive inspection, the carbon brushes should be replaced, and a comprehensive diagnosis of the system should be carried out.

[0030] Another object of the present invention is to provide a system for intelligent holographic monitoring of carbon brushes of generator sets. The present invention optimizes inspection efficiency and realizes real-time monitoring of carbon brush wear. The system of the present invention integrates three-dimensional temperature distribution, shaft vibration and high-precision wear data through multimodal data fusion processing, thereby realizing a comprehensive assessment of the carbon brush status. The machine learning technology of neural networks is used to provide early warning for carbon brush failures and avoid potential damage. By setting early warning thresholds, clear maintenance guidance is provided to the maintenance team and the maintenance process is optimized. The safety and reliability of generator set operation are improved, downtime and maintenance costs are reduced, and the continuity and economy of power supply are ensured.

[0031] As a preferred solution of the system for intelligent holographic monitoring of carbon brushes of a generator set described in the present invention, it is characterized by including a three-dimensional temperature distribution data acquisition module, a swing and shaft vibration data processing module, an online measurement module for carbon brush wear, a multimodal data fusion module and an intelligent early warning module.

[0032] The three-dimensional temperature distribution data acquisition module is used to perform multi-angle temperature measurement on the carbon brushes and collector rings of the generator set using high-precision infrared thermal imaging technology to obtain three-dimensional temperature distribution data.

[0033] The swing and shaft vibration data processing module is used to measure the swing of the carbon brush position of the generator set through a sensor, and use an adaptive filtering algorithm to process the shaft vibration data, extract characteristic parameters of the shaft vibration and evaluate the operating status of the carbon brush position.

[0034] The carbon brush wear online measurement module is used to measure the carbon brush wear online using computer vision and deep learning technology, calculate the wear amount through a morphological processing algorithm, and monitor the wear state of the carbon brush.

[0035] The multimodal data fusion module is used to perform feature-level and decision-level fusion processing on the three-dimensional temperature distribution data, shaft vibration data and high-precision wear data to improve the accuracy and reliability of monitoring.

[0036] The intelligent early warning module is used to build a machine learning model based on a neural network, process the fused data, and provide intelligent early warning of the carbon brush status through feature selection and model optimization.

[0037] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of a method described in the intelligent holographic monitoring of carbon brushes of a generator set are implemented.

[0038] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of a method described in the intelligent holographic monitoring of carbon brushes of a generator set are implemented.

[0039] Beneficial effects of the present invention: The present invention uses high-precision infrared thermal imaging technology to perform multi-angle temperature measurement, combined with the swing data collected by the sensor and the shaft vibration data processed by the adaptive filtering algorithm, to achieve comprehensive monitoring of the carbon brush status of the generator set, effectively prevent equipment damage caused by overheating, and improve the early warning capability of the safe operation of the equipment. At the same time, the carbon brush wear is measured online using computer vision technology assisted by deep learning, and the wear data is accurately calculated through the morphological processing algorithm, which provides a quantitative basis for maintenance decisions, extends the service life of the equipment, and reduces maintenance costs. Multimodal data fusion processing further improves the accuracy and reliability of the monitoring system, making data complementarity and comprehensive analysis possible. The intelligent early warning system constructed based on neural network machine learning technology improves the intelligence and accuracy of the early warning through feature selection and model optimization, optimizes the maintenance process, and enhances the emergency response capability. Overall, the present invention provides a comprehensive monitoring system, which not only improves the safety and reliability of the operation of the generator set, but also improves the maintenance efficiency and reduces the labor cost through computer equipment that is easy to deploy and operate, achieving a dual improvement in technology popularization and practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0041] Figure 1 An overall flow chart of a method for intelligent holographic monitoring of carbon brushes of a generator set provided by one embodiment of the present invention.

[0042] Figure 2 A technical roadmap for developing a method for intelligent holographic monitoring of carbon brushes of a generator set provided in accordance with an embodiment of the present invention.

[0043] Figure 3 A system solution flow chart of a system for intelligent holographic monitoring of carbon brushes of a generator set provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0045] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0046] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is mutually exclusive with other embodiments, either individually or selectively.

[0047] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0048] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0049] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0050] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a method for intelligent holographic monitoring of carbon brushes of a generator set, comprising:

[0051] S1: Use high-precision infrared thermal imaging technology to perform multi-angle temperature measurement on the carbon brushes and collector rings of the generator set to obtain three-dimensional temperature distribution data, use sensors to measure the swing of the carbon brush position of the generator set, and process the shaft vibration data through an adaptive filtering algorithm.

[0052] The obtaining of three-dimensional temperature distribution data includes using high-precision infrared thermal imaging technology to perform multi-angle temperature measurement on the carbon brushes and collector rings of the generator set to obtain the three-dimensional temperature distribution data.

[0053] Furthermore, the multi-angle temperature measurement includes performing a safety inspection on the generator set, selecting the infrared thermal imager lens and measurement distance according to the specific structure and position of the generator set's carbon brushes and collector rings, and calibrating the infrared thermal imager, planning the measurement angle and path according to the specific layout of the generator set, and performing multi-angle temperature measurement on the generator set's carbon brushes and collector rings using the infrared thermal imager according to the planned measurement angles and paths, and collecting and storing image data at each angle in real time.

[0054] The three-dimensional temperature includes importing the collected multi-angle temperature images into thermal image analysis software, performing image processing and three-dimensional reconstruction, analyzing and verifying the reconstructed three-dimensional temperature distribution data, and compiling a temperature measurement report based on the analysis results.

[0055] The image processing includes removing noise, enhancing contrast and correcting lens distortion; the analysis of three-dimensional temperature distribution includes analyzing uniformity of temperature distribution, hot spot identification and temperature gradient.

[0056] Furthermore, the measuring of the swing of the carbon brush position of the generator set includes calibrating the sensor, collecting the swing data of the carbon brush position of the generator set, and preprocessing the collected data.

[0057] The adaptive filtering algorithm includes using the least mean square algorithm according to the characteristics of the shaft vibration data, implementing the algorithm through programming, filtering the pre-processed data, analyzing the filtering effect, extracting the characteristic parameters of the shaft vibration, and evaluating the operating status of the generator set carbon brush position.

[0058] The least mean square algorithm formula is expressed as:

[0059] w(n+1)=w(n)+μe(n)x(n)

[0060] e(n)=d(n)-y(n)

[0061] Where w(n) is the filter weight vector at time n, μ is the step size factor, e(n) is the error signal at time n, d(n) is the desired signal, y(n) is the filter output, x(n) is the input signal vector at time n, and n is the discrete time index.

[0062] S2: The wear of the carbon brush is measured online through deep learning-assisted computer vision measurement technology, and the wear data is obtained by combining the morphological processing algorithm. All the data obtained are then processed by multimodal data fusion.

[0063] Furthermore, the online measurement of the wear of the carbon brush includes: using a high-definition camera to monitor the carbon brush of the generator set in real time, adjusting the position and focal length of the camera, and graying the collected image, using Gaussian filtering to remove image noise, using an edge detection algorithm to extract edge information of the carbon brush, and using a deep learning framework to build a convolutional neural network model, train the CNN model, and use the trained CNN model to infer the preprocessed image, identify the wear area, perform morphological processing and calculate the wear amount.

[0064] The morphological processing formula is:

[0065]

[0066] Among them, A is the original image, B is the structural element, For corrosion operations, For expansion operation.

[0067] The wear calculation formula is:

[0068]

[0069] Where S is the area of ​​the worn area, R is the pixel set in the worn area, and (x, y) is the pixel coordinate in the worn area.

[0070] Furthermore, the multimodal data fusion includes performing feature-level fusion and decision-level fusion processing on the obtained three-dimensional temperature distribution data, shaft vibration data and high-precision wear data.

[0071] The feature-level fusion includes using linear discriminant analysis to reduce the dimension of the features of the data source, and merging the reduced features to form a comprehensive feature vector; the decision-level fusion includes using a support vector machine model to train the comprehensive feature vector and generate a prediction result, and using a weighted summation mechanism to merge the output of the model to obtain a final decision.

[0072] S3: Build machine learning technology based on neural networks, and use machine learning technology to process the fused data, and provide intelligent early warning of carbon brush status through feature selection and model optimization.

[0073] Furthermore, the construction of a neural network-based machine learning technology includes dividing the fused data set into a training set, a validation set, and a test set, designing a neural network architecture, determining the number of neurons and activation functions in each layer, building a machine learning model, and performing model training.

[0074] The model training includes using a back propagation algorithm and a gradient descent method to perform model training, and using a validation set to adjust model parameters.

[0075] like Figure 2 The intelligent early warning of the carbon brush status includes building a machine learning technology based on a neural network, and using the machine learning technology to process the fused data, and calculating the risk coefficient according to business needs and model performance, setting the early warning threshold and dividing the prediction results into different areas according to the early warning threshold, and providing an intelligent early warning of the carbon brush status through feature selection and model optimization.

[0076] The formula for calculating the risk coefficient is:

[0077]

[0078] Among them, F(θ,X) is the risk coefficient, θ is the warning threshold vector, X is the fused feature vector, m is the number of features, and f i (X) is the output of the i-th feature after being processed by the neural network, g i (X) is the weight function of the i-th feature, h i(X) is the standard deviation function of the i-th feature, and i is the variable index.

[0079] The weight function formula is:

[0080]

[0081] Among them, g i (X) is the weight function of the i-th feature, IG i (X,Y) is the information gain of the i-th feature to the target Y, i and k are variable indices, and m is the number of features.

[0082] The standard deviation function formula is:

[0083]

[0084] Where N is the number of samples, X il is the ith eigenvalue of the lth sample, μ i is the mean of the ith feature, h i (X) is the standard deviation function of the i-th feature, and i and l are variable indices.

[0085] The setting of the early warning threshold includes setting the safety threshold to 0.1, setting the warning threshold to 0.5, and setting the danger threshold to 1; when F(θ,X)<0.1, it indicates that the carbon brush is in normal condition, with no signs of wear, and routine monitoring is maintained; when 0.1≤F(θ,X)<0.5, it indicates that the carbon brush is in abnormal condition, and attention should be paid to monitoring, regular inspections should be performed, the monitoring frequency should be increased, and a maintenance plan should be prepared; when 0.5≤F(θ,X)<1, it indicates that the carbon brush is in worn condition, and a detailed inspection should be carried out immediately, preventive maintenance measures should be implemented, and operating parameters should be adjusted; when F(θ,X)≥1, it indicates that the carbon brush is in dangerous condition and has caused damage to the generator set, and the generator set should be shut down immediately for a comprehensive inspection, the carbon brushes should be replaced, and a comprehensive diagnosis of the system should be carried out.

[0086] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0087] Example 2, reference Figure 3 , which is the second embodiment of the present invention, provides a system for intelligent holographic monitoring of carbon brushes of a generator set, including a three-dimensional temperature distribution data acquisition module 100, a swing and shaft vibration data processing module 200, an online measurement module for carbon brush wear 300, a multimodal data fusion module 400 and an intelligent early warning module 500.

[0088] The three-dimensional temperature distribution data acquisition module 100 is used to perform multi-angle temperature measurement on the carbon brushes and collector rings of the generator set using high-precision infrared thermal imaging technology to obtain three-dimensional temperature distribution data.

[0089] The swing and shaft vibration data processing module 200 is used to measure the swing of the carbon brush position of the generator set through a sensor, and use an adaptive filtering algorithm to process the shaft vibration data, extract the characteristic parameters of the shaft vibration and evaluate the operating status of the carbon brush position.

[0090] The carbon brush wear online measurement module 300 is used to measure the carbon brush wear online using computer vision and deep learning technology, calculate the wear amount through a morphological processing algorithm, and monitor the wear state of the carbon brush.

[0091] The multimodal data fusion module 400 is used to perform feature-level and decision-level fusion processing on the three-dimensional temperature distribution data, shaft vibration data and high-precision wear data to improve the accuracy and reliability of monitoring.

[0092] The intelligent early warning module 500 is used to build a machine learning model based on a neural network, process the fused data, and perform intelligent early warning on the carbon brush status through feature selection and model optimization.

[0093] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0094] Embodiment 3, the third embodiment of the present invention, is different from the first two embodiments in that:

[0095] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0096] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0097] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0098] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

Claims

1. A method for intelligent holographic monitoring of carbon brushes of a generator set, characterized in that: include, Use high-precision infrared thermal imaging technology to measure the temperature of the carbon brushes and collector rings of the generator set from multiple angles to obtain three-dimensional temperature distribution data. Use sensors to measure the swing of the carbon brush position of the generator set, and process the shaft vibration data through an adaptive filtering algorithm. The wear of carbon brushes is measured online through computer vision measurement technology assisted by deep learning, and the wear data is obtained by combining morphological processing algorithms. All the data obtained are then processed through multimodal data fusion. Build machine learning technology based on neural networks, use machine learning technology to process the fused data, and provide intelligent early warning of carbon brush status through feature selection and model optimization.

2. A method for intelligent holographic monitoring of carbon brushes of a generator set as claimed in claim 1, characterized in that: The obtaining of the three-dimensional temperature distribution data includes using high-precision infrared thermal imaging technology to perform multi-angle temperature measurement on the carbon brushes and collector rings of the generator set to obtain the three-dimensional temperature distribution data; The multi-angle temperature measurement includes performing a safety inspection on the generator set, selecting an infrared thermal imager lens and a measuring distance according to the specific structure and position of the carbon brush and the collector ring of the generator set, and calibrating the infrared thermal imager, planning the measuring angle and path according to the specific layout of the generator set, and performing multi-angle temperature measurement on the carbon brush and the collector ring of the generator set using the infrared thermal imager according to the planned measuring angle and path, and collecting and storing image data of each angle in real time; The three-dimensional temperature includes importing the collected multi-angle temperature images into thermal image analysis software, performing image processing and three-dimensional reconstruction, analyzing and verifying the reconstructed three-dimensional temperature distribution data, and compiling a temperature measurement report based on the analysis results; The image processing includes removing noise, enhancing contrast and correcting lens distortion; the analysis of three-dimensional temperature distribution includes analyzing uniformity of temperature distribution, hot spot identification and temperature gradient.

3. A method for intelligent holographic monitoring of carbon brushes of a generator set as claimed in claim 2, characterized in that: The measuring of the swing of the carbon brush position of the generator set includes calibrating the sensor, collecting the swing data of the carbon brush position of the generator set, and preprocessing the collected data; The adaptive filtering algorithm includes using the least mean square algorithm according to the characteristics of the shaft vibration data, implementing the algorithm through programming, filtering the pre-processed data, analyzing the filtering effect, extracting the characteristic parameters of the shaft vibration, and evaluating the operating status of the carbon brush position of the generator set; The least mean square algorithm formula is expressed as: w(n+1)=w(n)+μe(n)x(n) e(n)=d(n)-y(n) Where w(n) is the filter weight vector at time n, μ is the step size factor, e(n) is the error signal at time n, d(n) is the desired signal, y(n) is the filter output, x(n) is the input signal vector at time n, and n is the discrete time index.

4. A method for intelligent holographic monitoring of carbon brushes of a generator set as claimed in claim 3, characterized in that: The online measurement of the wear amount of the carbon brush includes: using a high-definition camera to monitor the carbon brush of the generator set in real time, adjusting the position and focal length of the camera, graying the collected image, removing image noise by Gaussian filtering, extracting edge information of the carbon brush by edge detection algorithm, building a convolutional neural network model by deep learning framework, training the CNN model, and using the trained CNN model to infer the pre-processed image, identify the wear area, perform morphological processing and calculate the wear amount; The morphological processing formula is: Among them, A is the original image, B is the structural element, is the corrosion operation, ⊕ is the expansion operation; The wear calculation formula is: Where S is the area of ​​the worn area, R is the pixel set in the worn area, and (x, y) is the pixel coordinate in the worn area.

5. A method for intelligent holographic monitoring of carbon brushes of a generator set as claimed in claim 4, characterized in that: The multimodal data fusion includes performing feature-level fusion and decision-level fusion processing on the obtained three-dimensional temperature distribution data, shaft vibration data and high-precision wear data; The feature-level fusion includes using linear discriminant analysis to reduce the dimension of the features of the data source, and merging the reduced features to form a comprehensive feature vector; the decision-level fusion includes using a support vector machine model to train the comprehensive feature vector and generate a prediction result, and using a weighted summation mechanism to merge the output of the model to obtain a final decision.

6. A method for intelligent holographic monitoring of carbon brushes of a generator set as claimed in claim 5, characterized in that: The construction of the neural network-based machine learning technology includes dividing the fused data set into a training set, a validation set and a test set, designing a neural network architecture, determining the number of neurons and activation functions of each layer, building a machine learning model, and performing model training; The model training includes using a back propagation algorithm and a gradient descent method to perform model training, and using a validation set to adjust model parameters.

7. A method for intelligent holographic monitoring of carbon brushes of a generator set as claimed in claim 6, characterized in that: The intelligent early warning of the carbon brush status includes building a machine learning technology based on a neural network, and using the machine learning technology to process the fused data, and calculating the risk coefficient according to business needs and model performance, setting the early warning threshold and dividing the prediction results into different areas according to the early warning threshold, and providing an intelligent early warning of the carbon brush status through feature selection and model optimization; The setting of the early warning threshold includes setting the safety threshold to 0.1, setting the warning threshold to 0.5, and setting the danger threshold to 1; when F(θ,X)<0.1, it indicates that the carbon brush is in normal condition, with no signs of wear, and routine monitoring is maintained; when 0.1≤F(θ,X)<0.5, it indicates that the carbon brush is in abnormal condition, and attention should be paid to monitoring, regular inspections should be performed, the monitoring frequency should be increased, and a maintenance plan should be prepared; when 0.5≤F(θ,X)<1, it indicates that the carbon brush is in worn condition, and a detailed inspection should be carried out immediately, preventive maintenance measures should be implemented, and operating parameters should be adjusted; when F(θ,X)≥1, it indicates that the carbon brush is in dangerous condition and has caused damage to the generator set, and the generator set should be shut down immediately for a comprehensive inspection, the carbon brushes should be replaced, and a comprehensive diagnosis of the system should be carried out.

8. A system using a method for intelligent holographic monitoring of carbon brushes of a generator set as claimed in any one of claims 1 to 7, characterized in that: It includes three-dimensional temperature distribution data acquisition module, swing and shaft vibration data processing module, carbon brush wear online measurement module, multi-modal data fusion module and intelligent early warning module; The three-dimensional temperature distribution data acquisition module is used to perform multi-angle temperature measurement on the carbon brushes and collector rings of the generator set using high-precision infrared thermal imaging technology to obtain three-dimensional temperature distribution data; The swing and shaft vibration data processing module is used to measure the swing of the carbon brush position of the generator set through a sensor, and use an adaptive filtering algorithm to process the shaft vibration data, extract the characteristic parameters of the shaft vibration and evaluate the operating status of the carbon brush position; The carbon brush wear online measurement module is used to measure the carbon brush wear online using computer vision and deep learning technology, calculate the wear amount through a morphological processing algorithm, and monitor the wear state of the carbon brush; The multimodal data fusion module is used to perform feature-level and decision-level fusion processing on the three-dimensional temperature distribution data, shaft vibration data and high-precision wear data to improve the accuracy and reliability of monitoring; The intelligent early warning module is used to build a machine learning model based on a neural network, process the fused data, and provide intelligent early warning of the carbon brush status through feature selection and model optimization.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for intelligent holographic monitoring of carbon brushes of a generator set described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for intelligent holographic monitoring of carbon brushes of a generator set according to any one of claims 1 to 7 are implemented.

Images