Intelligent online monitoring device, system and method for carbon brush of brush holder of hydro-generator

Abstract

The invention discloses an intelligent online monitoring device, system and method for a brush holder carbon brush of a hydro-generator, and belongs to the technical field of online monitoring of carbon brushes. An intelligent online monitoring device for a brush holder and a carbon brush of a hydro-generator comprises an electric brush cover, a collecting ring is installed in the electric brush cover, a brush holder assembly is connected to the collecting ring, a brush holder is installed on the brush holder assembly, and the carbon brush and monitoring equipment are installed on the brush holder. The problems that in the actual use process in the prior art, due to poor contact between the electric brushes and the collecting rings, the contact resistance between the collecting rings and the electric brushes is increased, the surfaces of the collecting rings are electrically corroded, the electric brushes are abraded too fast, and the maintenance frequency is increased are solved. According to the invention, the operation of the carbon brush is monitored, an alarm is given according to the operation condition of the carbon brush, the conditions that the surface of the collector ring is electrically corroded and the brush is worn too quickly are avoided, the maintenance frequency is increased, the operation is simple and consistent, a strong error data judgment mechanism is provided, and the working intensity of operators is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon brush online monitoring, and in particular to an intelligent online monitoring device, system and method for carbon brushes on a brush holder of a hydro-turbine generator. Background Art

[0002] The hydro-turbine generator set is a large-scale electromechanical energy conversion device, and its reliability is crucial to the safe and stable operation of the hydropower plant. The brush is an important component of the hydro-turbine generator.

[0003] A Chinese patent with publication number CN215297593U discloses a carbon brush structure for a hydro-turbine generator, including a mounting ring and a quick-installation mechanism; the mounting ring: six slides distributed at equal angles are fixedly connected to its upper end, mounting frames are slidably connected inside the slides, carbon brushes are plugged into the inside of the mounting frames, two wires are arranged at the end of the carbon brush away from the center of the mounting ring, avoidance openings corresponding to the positions are provided on the slide and the mounting frame, sliding holes provided inside the end of the slide away from the center of the mounting ring are slidably connected with sliding columns, and the sliding columns are fixedly connected to the mounting frame in the same slide; the quick-installation mechanism: they are respectively arranged inside the slide, the quick-installation mechanism is installed in cooperation with the carbon brush, and two springs are respectively arranged between the end of the mounting frame away from the center of the mounting ring and the slide. The carbon brush structure for the hydro-turbine generator can quickly replace the carbon brushes for the hydro-turbine generator, so that personnel can save time and effort in the process of replacing the carbon brushes.

[0004] In the actual use of the above patent, due to the poor contact between the brush and the collector ring during the operation of the generator, it is a common fault during the operation of the generator, which will cause the contact resistance between the collector ring and the brush to increase, resulting in heating, power generation, sparking and other phenomena, causing electrical corrosion on the surface of the collector ring. Poor contact between the collector ring and the carbon brush will also cause the brush to wear too quickly, increase the maintenance frequency, and cause the brush and collector ring to overheat due to friction, and even cause fire in severe cases; therefore, it does not meet the existing needs, and we have proposed an intelligent online monitoring device, system and method for the brush holder carbon brush of a hydro-turbine generator. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent online monitoring device, system and method for the carbon brush of a brush holder of a hydro-turbine generator, so as to realize the monitoring of the operation of the carbon brush and issue an alarm according to the operation status of the carbon brush to avoid poor contact between the carbon brush and the collector ring, which not only causes the collector ring surface to be electro-corroded, but also causes the brush to wear too quickly and increases the maintenance frequency. The operation is simple and consistent, and it has a powerful error data judgment mechanism, which reduces the workload of operators and solves the problems raised in the above-mentioned background technology.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent online monitoring device for a brush holder and carbon brush of a hydro-turbine generator, comprising a brush cover and a collector ring, wherein a collector ring is installed in the brush cover, a brush holder assembly is connected to the collector ring, a brush holder assembly is installed on the brush holder assembly, and a carbon brush and a monitoring device are installed on the brush holder.

[0007] Preferably, the monitoring equipment includes a temperature measuring device, a displacement sensor, a current sensor, a bolt fixing position, a brush handle fixing position, a main machine room and a communication gateway. The temperature measuring device, the displacement sensor and the current sensor are all arranged on the outside of the main machine room. A brush handle fixing position is provided at the bottom of the main machine room. A bolt fixing position is provided at the bottom of the main machine room below the temperature measuring device. The main machine room is connected to the communication gateway.

[0008] Preferably, the temperature measuring device includes a main panel, an auxiliary laser locator, a temperature measuring sensor and a power supply interface. The auxiliary laser locator and the temperature measuring sensor are provided on the main panel, and the power supply interface is provided on the side of the main panel.

[0009] An intelligent online monitoring system for carbon brushes of a brush holder of a hydro-turbine generator, comprising:

[0010] The monitoring module is used to monitor the operating temperature of each carbon brush, the moving distance of the carbon brush during operation, and the current of the carbon brush in real time, and upload the monitoring data to the brush monitoring platform;

[0011] The 3D display module is used to receive the data monitored by the monitoring module and display the real-time status of the carbon brush operation in 3D form, as well as visually display the distribution point layout of the carbon brush in the collector ring;

[0012] System management module, used to create, edit and delete menu items, manage permissions for each function item, and manage roles, organizations and personnel;

[0013] The statistical analysis module is used to display the temperature monitoring curve, displacement monitoring curve and current monitoring curve, and to analyze the health index of the hydro-generator set according to the temperature, displacement and current information;

[0014] The monitoring and alarm module is used to trigger corresponding alarms according to the analysis results, and register the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, it selects the given handling method and describes the specific handling content, and records the handler and handling time.

[0015] Preferably, the monitoring module comprises:

[0016] The temperature measurement module is used to accurately collect the surface temperature of the exposed area of ​​the carbon brush using a temperature measurement device;

[0017] The distance measuring module is used to monitor the moving distance of the carbon brush when the carbon brush is running by using a displacement sensor and laser distance measuring, and determine the wear amount of the carbon brush according to the monitored moving distance;

[0018] The current measurement module is used to collect excitation current using a current sensor and Hall contactless sensing.

[0019] Preferably, the system management module includes:

[0020] Menu management module, used to create, edit and delete menu items, as well as manage the permissions of each function item;

[0021] The role management module is used to bind role functions according to the needs of the power plant and complete role function comparison. After specific personnel are assigned roles, they can manage users in their jurisdiction.

[0022] The organization management module is used to display the organization tree, fold, expand, add sub-organizations, edit sub-organizations, and delete sub-organizations;

[0023] The personnel management module is used to manage the information of system administrators, plant administrators and department unit administrators, bind personnel to role information, and bind them to the organization information sheet number.

[0024] Preferably, the monitoring and alarm module specifically includes:

[0025] Temperature alarm: Alarm for over-high carbon brush temperature, set low temperature lower limit, high temperature upper limit, and provide temperature rising status prompt based on the current collected temperature and the last two collected temperatures;

[0026] Current alarm: Set the normal current range to issue high current alarm, excessive current difference alarm and excessive current range alarm;

[0027] Displacement alarm: Set the initial displacement value. After clicking the initial button, the current displacement value is stored. The carbon brush displacement is monitored during the operation of the machine. When the displacement exceeds the limit mm, an alarm is triggered;

[0028] MCU settings: set the reporting cycle, alarm interval, and whether the alarm, voice, SMS and other interfaces need to be configured within a certain period of time after the alarm is triggered;

[0029] Alarm handling: register the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, select the given handling method and describe the specific handling content, and record the person who handled it and the handling time.

[0030] Preferably, the monitoring module is also used to monitor the vibration data and audio data of the carbon brush;

[0031] The statistical analysis module specifically includes:

[0032] A data preprocessing submodule, used for denoising and normalizing the real-time monitoring data received from the monitoring module, wherein the real-time monitoring data includes the operating temperature, moving distance, current, vibration data and audio data of the carbon brush;

[0033] The data fusion submodule is used to use data fusion technology to fuse the operating temperature, moving distance, current, vibration data and audio data of the carbon brush to evaluate and obtain the status information of the carbon brush;

[0034] The visualization output submodule is used to construct a three-dimensional physical model of the carbon brush, associate the status information of the carbon brush with the three-dimensional physical model, and display the real-time status of the carbon brush using the three-dimensional physical model of the carbon brush through the 3D display module;

[0035] The optimization submodule is used to optimize the real-time status of the carbon brush before display. After the optimization is completed, it is provided to the 3D display module for simulation and dynamic display using the carbon brush three-dimensional physical model.

[0036] Preferably, the data fusion submodule performs fusion processing on the real-time monitoring data in the following specific ways:

[0037] Extract statistical characteristic parameters of the operating temperature, moving distance and current data of the carbon brush, the statistical characteristic parameters including mean, variance, maximum value and minimum value;

[0038] Extracting frequency domain features of vibration data and audio data of the carbon brush, that is, extracting vibration frequency and audio frequency respectively by performing Fourier transform on the vibration data and audio data;

[0039] Extracting the time domain features of the vibration data and the audio data of the carbon brush, that is, respectively extracting the vibration time domain waveform features and the audio time domain waveform features by analyzing the signal time domain waveforms of the vibration data and the audio data;

[0040] Perform vector transformation on the statistical feature parameters, frequency domain features and time domain features, and connect the transformed feature vectors to form a fused feature vector;

[0041] Based on the conditional probability (decision) and prior probability (marginal probability) of each sensor that collects real-time monitoring data, the following formula is used to calculate the posterior probability of an event:

[0042]

[0043] Among them, P(A i |B) means that A is known to have occurred when event B occurs i The posterior probability of an event occurring, where A1…A i …A n is an event group; P(A i) means A i Prior probability of an event occurring; P(B|A i ) indicates that A is known i The conditional probability of event B occurring when event A occurs; P(A j ) means A j Prior probability of an event occurring; P(B|A j ) indicates that A is known j The conditional probability of event B occurring when event B occurs, j = 1, 2, ..., n;

[0044] A deep learning model based on recurrent neural network (RNN) is constructed. The deep learning model is used to process the fused feature vector and the calculated posterior probability to identify the status information of the carbon brush, which includes the operating status and potential faults.

[0045] An intelligent online monitoring method for a brush holder carbon brush of a hydro-turbine generator comprises the following steps:

[0046] Step 1: Initialize the unit. The system administrator and plant administrator log in to the system, add and edit the list of turbine generator units, and set the effective number of layers and the number of carbon brushes per layer.

[0047] Step 2: During the operation of the carbon brush, the surface temperature, moving distance and excitation current of the exposed area of ​​the carbon brush are collected in real time through the monitoring equipment, and the wear amount of the carbon brush is determined according to the moving distance. The real-time collected information is uploaded to the 3D display module through the communication gateway;

[0048] Step 3: The 3D display module receives the data monitored by the monitoring module, and displays the real-time status of the carbon brush operation in D format, as well as the distribution point layout of the carbon brush in the collector ring for visual display;

[0049] Step 4: Plot the temperature monitoring data, displacement monitoring data and current monitoring data into temperature monitoring curves, displacement monitoring curves and current monitoring curves through the statistical analysis module, and perform health index analysis on the hydro-generator set according to the temperature, displacement and current information;

[0050] Step 5: The monitoring alarm module triggers the corresponding alarm according to the analysis results, and registers the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, the given handling method is selected and the specific handling content is described, and the handling person and handling time are recorded;

[0051] Step 6: When the monitoring equipment needs maintenance, initialize the carbon brushes of the hydro-turbine generator set that needs maintenance, bind the number of layers of carbon brushes, the number of carbon brushes on each layer, and the position information of each carbon brush, and perform basic information maintenance on the temperature sensor, displacement sensor, and current sensor.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The present invention monitors the operation of the carbon brush and issues an alarm based on the operation status of the carbon brush, thereby avoiding poor contact between the carbon brush and the collector ring, which leads to increased contact resistance between the collector ring and the carbon brush, resulting in heating, power generation, sparking and the like, which not only causes electrical corrosion on the collector ring surface, but also causes excessive wear of the brush, increases the maintenance frequency, and even causes fire in severe cases. The present invention adopts a three-layer identity verification mechanism of operating system, database management, and application program, and utilizes role-based permission control to prevent illegal viewing, theft and tampering of data, thereby ensuring that customers obtain greater benefits with relatively small investments. The operation is simple and consistent, and it has a powerful error data judgment mechanism to reduce the workload of operators. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic diagram of the intelligent online monitoring device for carbon brushes of a brush holder of a hydro-turbine generator of the present invention;

[0055] Figure 2 It is a schematic diagram of the installation of the intelligent online monitoring device for the brush holder carbon brush of the hydro-generator of the present invention;

[0056] Figure 3 It is a schematic diagram of the assembly of the carbon brush and the monitoring device of the present invention;

[0057] Figure 4 It is a schematic diagram of the structure of the monitoring device of the present invention;

[0058] Figure 5 It is a schematic diagram of the structure of the temperature measuring device of the present invention;

[0059] Figure 6 It is a topological diagram of the intelligent online monitoring system for the brush holder and carbon brush of the hydro-generator of the present invention;

[0060] Figure 7 It is the large screen picture of the monitoring center of the present invention;

[0061] Figure 8 This is an interface diagram of the intelligent online monitoring system for the brush holder and carbon brush of the hydro-generator of the present invention;

[0062] Fig. 9 The present invention is a flow chart of the working method of the intelligent online monitoring system for carbon brushes of a brush holder of a hydro-turbine generator.

[0063] In the figure: 1. Brush cover; 2. Collector ring; 3. Monitoring equipment; 31. Temperature measuring device; 311. Main panel; 312. Auxiliary laser locator; 313. Temperature sensor; 314. Power interface; 32. Displacement sensor; 33. Current sensor; 34. Bolt fixing position; 35. Brush handle fixing position; 36. Main machine position and cabin; 4. Carbon brush; 5. Brush holder assembly; 6. Brush holder. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] In order to solve the problem that the existing technology has poor contact between the brush and the collector ring during actual use, which will lead to increased contact resistance between the collector ring and the brush, resulting in heating, power generation, sparking, etc., which will not only cause the collector ring surface to be corroded by electricity, but also cause the brush to wear too quickly, increase the maintenance frequency, and cause the brush and collector ring to overheat due to friction, and even cause fire in serious cases. Please refer to Figure 1-Figure 9 , this embodiment provides the following technical solutions:

[0066] An intelligent online monitoring device for brush holder carbon brushes of a hydro-turbine generator comprises a brush cover 1 and a collector ring 2. The collector ring 2 is installed in the brush cover 1, the collector ring 2 is connected to a brush holder assembly 5, a brush holder 6 is installed on the brush holder assembly 5, a carbon brush 4 and a monitoring device 3 are installed on the brush holder 6, and the monitoring device 3 is integrated and bound with the carbon brush 4 to form a wireless transmission set based on a 4G module to the cloud, and information is reported to a brush monitoring platform through real-time monitoring.

[0067] The monitoring device 3 includes a temperature measuring device 31, a displacement sensor 32, a current sensor 33, a bolt fixing position 34, a brush handle fixing position 35, a main machine compartment 36 and a communication gateway. The temperature measuring device 31, the displacement sensor 32 and the current sensor 33 are all arranged on the outside of the main machine compartment 36. The bottom of the main machine compartment 36 is provided with a brush handle fixing position 35. The main machine compartment 36 is located below the temperature measuring device 31 and is provided with a bolt fixing position 34. The main machine compartment 36 is connected to the communication gateway. The current sensor 33 adopts Hall non-contact sensing to perform excitation current. The displacement sensor 32 uses a laser ranging module to monitor the wear of the carbon brush 4 with an accuracy of 0.1mm. The temperature measuring device 31 uses infrared temperature measurement to accurately collect the surface temperature of the exposed area of ​​the carbon brush 4 for temperature compensation. The aperture of the bolt fixing position 34 is 6mm. The monitoring device 3 is fixed to the brush holder 6 with an M6 bolt. The brush handle fixing position 35 adopts an upper and lower clamping force structure to fix the handle shaft of the brush holder 6 and enhance the fixation of the fuselage. The main machine compartment 36 is used to fix the mainboard and battery and connect with the terminal wires.

[0068] The temperature measuring device 31 includes a main panel 311, an auxiliary laser locator 312, a temperature sensor 313 and a power interface 314. The main panel 311 is provided with an auxiliary laser locator 312 and a temperature sensor 313. The side of the main panel 311 is provided with a power interface 314. The main panel 311 is an integrated panel, which is fixed by four 2.5mm bolt holes. The auxiliary laser locator 312 uses visible light to align the auxiliary temperature measurement point with the monitoring area of ​​the collector ring 2. The temperature sensor 313 uses infrared temperature measurement to accurately collect the surface temperature of the exposed area of ​​the collector ring 2 for temperature compensation. The power interface 314 is a 5.5x2.1mm charging port.

[0069] An intelligent online monitoring system for carbon brushes of a brush holder of a hydro-turbine generator, comprising:

[0070] The monitoring module is used to monitor the operating temperature of each carbon brush 4, the moving distance of the carbon brush 4 during operation, and the current of the carbon brush 4 in real time, and upload the monitored data to the brush monitoring platform;

[0071] The 3D display module is used to receive the data monitored by the monitoring module, and display the real-time status of the carbon brush 4 in 3D form, as well as the distribution point deployment of the carbon brush 4 in the collector ring 2. It also displays the list of monitoring devices 3, and each device displays the globally unique device number, device IMEI, device status normal, offline, etc., monitoring temperature, monitoring current, and monitoring displacement. The title is the name of the generator set, and the upper right part displays the current system time and setting button;

[0072] System management module, used to create, edit and delete menu items, manage permissions for each function item, and manage roles, organizations and personnel;

[0073] The statistical analysis module is used to display the temperature monitoring curve, displacement monitoring curve and current monitoring curve, and to analyze the health index of the hydro-generator set according to the temperature, displacement and current information;

[0074] Specifically, check the temperature, displacement, and current monitoring details and trend charts by day, week, and month, monitor the temperature, displacement, and current data, and analyze the health index of the entire unit based on the temperature, displacement, current, and other information. It is necessary to obtain various parameters and calculation methods of the unit health indicators. For example, in a unit, if more than three carbon brushes 4 have over-temperature alarms, the unit health will be displayed in yellow, and more than five will be displayed in red;

[0075] The monitoring and alarm module is used to trigger corresponding alarms according to the analysis results, and register the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, it selects the given handling method and describes the specific handling content, and records the handler and handling time.

[0076] Monitoring modules include:

[0077] A temperature measurement module, used to accurately collect the surface temperature of the exposed area of ​​the carbon brush 4 using a temperature measurement device 31;

[0078] The distance measuring module is used to monitor the moving distance of the carbon brush 4 when the carbon brush 4 is running by using the displacement sensor 32 to measure the distance with laser, and determine the wear amount of the carbon brush 4 according to the monitored moving distance, with an accuracy of 0.1 mm;

[0079] The current measurement module is used to collect the excitation current using the current sensor 33 and Hall contactless sensing, with an accuracy of 1%.

[0080] System management module, including:

[0081] Menu management module, used to create, edit and delete menu items, as well as manage the permissions of each function item;

[0082] The menu management module includes the following functions:

[0083] Menu list display - can display all menu items and submenu items at the second level for users to manage and operate.

[0084] Add New Menu - Allows the user to add new menu items and submenu items.

[0085] Edit Menu - Allows users to edit existing menu items and submenu items, including changing menu names, link addresses, etc.

[0086] Delete Menu - Allows the user to delete menu items and submenu items that are no longer needed.

[0087] Menu tree list acquisition - used to obtain the hierarchical structure of the menu so as to display it on the front-end page. Through these functions, administrators can easily manage the menus of the Web system, thereby controlling and adjusting its functions. For example, administrators can adjust the order of menus, add or delete menu items, and adjust the display method of submenus as needed.

[0088] The role management module is used to bind role functions according to the needs of the power plant and complete role function comparison. After specific personnel are assigned roles, they can manage users in their jurisdiction.

[0089] The organization management module is used to display the organization tree, fold, expand, add sub-organizations, edit sub-organizations, and delete sub-organizations; the organization is a tree structure, the system administrator is the first-level organization, the single number is 01, the plant area number is in the form of 01, 02, and the unit / department is in the form of 01, 02, 03. In the system, the organization can filter the personnel, equipment and other information under the organization.

[0090] The personnel management module is used to manage the information of system administrators, plant administrators and department unit administrators, bind personnel and role information, and bind them with the organization information sheet number, so that specific plant managers can only view the equipment information within their jurisdiction.

[0091] Monitoring and alarm module, including

[0092] Alarm refers to the alarm of three types of sensors, which can consider whether the gateway itself should alarm, such as signal strength, power alarm, etc.;

[0093] Temperature alarm: Alarm for over-high temperature of carbon brush 4, set low temperature lower limit, high temperature upper limit, and prompt temperature rising status according to the current collected temperature and the last two collected temperatures;

[0094] Current alarm: Set the normal current range, generally 0-100A, to issue high current alarm, excessive current difference alarm and over-range alarm. Over-range means the current difference is greater than the rated value such as 10A or 20A.

[0095] Displacement alarm: Set the initial displacement value. After clicking the initial button, the current displacement value is stored. During the operation of the machine, the displacement of carbon brush 4 is monitored. When the displacement exceeds the limit of 50mm, an alarm is triggered.

[0096] MCU settings: set the reporting cycle, alarm interval, and whether the alarm, voice, SMS and other interfaces need to be configured within a certain period of time after the alarm is triggered;

[0097] Alarm handling: register the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, select the given handling method and describe the specific handling content, and record the person who handled it and the handling time.

[0098] Based on the above-mentioned embodiment, the monitoring module is also used to monitor the vibration data and audio data of the carbon brush;

[0099] The statistical analysis module specifically includes:

[0100] A data preprocessing submodule, used for denoising and normalizing the real-time monitoring data received from the monitoring module, wherein the real-time monitoring data includes the operating temperature, moving distance, current, vibration data and audio data of the carbon brush;

[0101] The data fusion submodule is used to use data fusion technology to fuse the operating temperature, moving distance, current, vibration data and audio data of the carbon brush to evaluate and obtain the status information of the carbon brush;

[0102] The visualization output submodule is used to construct a three-dimensional physical model of the carbon brush, associate the status information of the carbon brush with the three-dimensional physical model, and display the real-time status of the carbon brush using the three-dimensional physical model of the carbon brush through the 3D display module;

[0103] The optimization submodule is used to optimize the real-time status of the carbon brush before display. After the optimization is completed, it is provided to the 3D display module for simulation and dynamic display using the carbon brush three-dimensional physical model.

[0104] By adopting the above technical solution, after preprocessing the real-time monitoring data including denoising and normalization, the status information of the carbon brush is evaluated through multi-sensor data fusion, avoiding the defect that a single sensor may not be able to fully capture the operating status of the carbon brush, and can significantly improve the accuracy and reliability of the intelligent online monitoring system of the brush holder carbon brush of the hydro-turbine generator; by integrating multiple sensors, the system can capture the status information of the carbon brush from multiple dimensions during operation, and the status information includes the operating status, thereby providing more accurate health assessment and fault prediction; in addition, the status information is also associated with the pre-constructed three-dimensional physical model of the carbon brush. When the three-dimensional physical model of the carbon brush is used to display the real-time status of the carbon brush, the display of the three-dimensional physical model of the carbon brush is adjusted accordingly according to the change of the status information, and the real-time status of the carbon brush and the predicted fault information can be observed on the display interface, thereby having a simulation dynamic display effect, which is convenient for users to intuitively understand the operation of the hydro-turbine generator; the application of this aspect can not only improve the operating efficiency of the equipment, but also reduce maintenance costs and extend the service life of the equipment.

[0105] Based on the above embodiment, the data fusion submodule performs fusion processing on the real-time monitoring data in the following specific manners:

[0106] Extract statistical characteristic parameters of the operating temperature, moving distance and current data of the carbon brush, the statistical characteristic parameters including mean, variance, maximum value and minimum value;

[0107] Extracting frequency domain features of vibration data and audio data of the carbon brush, that is, extracting vibration frequency and audio frequency respectively by performing Fourier transform on the vibration data and audio data;

[0108] Extracting the time domain features of the vibration data and the audio data of the carbon brush, that is, respectively extracting the vibration time domain waveform features and the audio time domain waveform features by analyzing the signal time domain waveforms of the vibration data and the audio data;

[0109] Perform vector transformation on the statistical feature parameters, frequency domain features and time domain features, and connect the transformed feature vectors to form a fused feature vector;

[0110] Based on the conditional probability (decision) and prior probability (marginal probability) of each sensor that collects real-time monitoring data, the following formula is used to calculate the posterior probability of an event:

[0111]

[0112] Among them, P(A i |B) means that A is known to have occurred when event B occurs i The posterior probability of an event occurring, where A1…A i …A n is an event group; P(A i ) means A i Prior probability of an event occurring; P(B|A i ) indicates that A is known i The conditional probability of event B occurring when event A occurs; P(A j ) means A j Prior probability of an event occurring; P(B|A j ) indicates that A is known j The conditional probability of event B occurring when event B occurs, j = 1, 2, ..., n;

[0113] A deep learning model based on recurrent neural network (RNN) is constructed. The deep learning model is used to process the fused feature vector and the calculated posterior probability to identify the status information of the carbon brush, which includes the operating status and potential faults.

[0114] The above technical solution is adopted to realize the fusion processing of real-time monitoring data, by extracting corresponding features from each sensor data, quantizing the features, and then combining these feature vectors to form a fused feature vector; the Bayesian formula is used for posterior probability calculation, and the machine learning model is used to process the fused data to identify the operating status of the carbon brush and predict potential faults; the system can process and fuse data in real time and respond quickly to possible faults; in actual use, a feedback mechanism can also be set in the system to adjust the model and algorithm according to the actual feedback results obtained by the feedback mechanism, so as to improve the applicability and robustness of the system; the effectiveness of the data fusion algorithm can be tested in a simulation environment, such as deploying the system in an actual hydro-turbine generator environment, collecting data and verifying the accuracy and reliability of the algorithm. This solution can realize comprehensive monitoring and accurate prediction of the operating status of the carbon brush, and improve the operating efficiency and safety of the hydro-turbine generator.

[0115] An intelligent online monitoring method for a brush holder carbon brush of a hydro-turbine generator comprises the following steps:

[0116] Step 1: Initialize the unit. The system administrator and the plant administrator log in to the system, add and edit the list of turbine generator units, set the effective number of layers of the unit and the number of carbon brushes 4 on each layer. If the plant floor plan is editable and sensitive information is involved, no specific editing is required. The list of turbine generator units can be directly added and edited to provide basic information reference for the equipment management personnel to configure carbon brush 4 in the future. The editor selects the corresponding carbon brush 4 position to realize the binding of the monitoring host and the position of carbon brush 4;

[0117] Step 2: During the operation of the carbon brush 4, the surface temperature, moving distance and excitation current of the exposed area of ​​the carbon brush 4 are collected in real time through the monitoring device 3, and the wear amount of the carbon brush 4 is determined according to the moving distance, and the real-time collected information is uploaded to the 3D display module through the communication gateway;

[0118] Step 3: The 3D display module receives the data monitored by the monitoring module, and displays the real-time operation status of the carbon brush 4 in 3D form, and visually displays the distribution point layout of the carbon brush 4 in the collector ring 2;

[0119] Step 4: Plot the temperature monitoring data, displacement monitoring data and current monitoring data into temperature monitoring curves, displacement monitoring curves and current monitoring curves through the statistical analysis module, and perform health index analysis on the hydro-generator set according to the temperature, displacement and current information;

[0120] Step 5: The monitoring alarm module triggers the corresponding alarm according to the analysis results, and registers the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, the given handling method is selected and the specific handling content is described, and the handling person and handling time are recorded;

[0121] Step six: When the monitoring device 3 needs maintenance, initialize the carbon brush 4 of the hydro-turbine generator set that needs maintenance, bind the layer number of the carbon brush 4, the number of carbon brushes 4 on each layer, and the position information of each carbon brush 4, and perform basic information maintenance on the temperature sensor 313, displacement sensor 32, and current sensor 33, including filling in the sensor device serial number, device number monitoring status, label name / software version, group name, IMEI, collection / reporting cycle, device binding time, and person in charge.

[0122] In summary, the present invention takes carbon brush 4 displacement monitoring, carbon brush 4 temperature monitoring, and collector ring current monitoring as main objectives, and adapts physical displacement, contact temperature measurement, laser displacement, infrared temperature measurement, Hall current sensor and other sensors to match carbon brush 4 monitoring targets, realizes monitoring of carbon brush 4 operation, and issues an alarm according to the operation status of carbon brush 4, so as to avoid poor contact between carbon brush 4 and collector ring 2, which leads to increased contact resistance between collector ring 2 and carbon brush 4, and the occurrence of heating, power generation, sparking and the like, which not only causes electrical corrosion on the surface of collector ring 2, but also causes excessive wear of brush 4, increases maintenance frequency, and even causes fire in severe cases. A three-layer identity verification mechanism of operating system, database management, and application is adopted, and role-based permission control is used to prevent data It can prevent illegal viewing, theft and tampering. It adopts off-site backup and one-way transmission mechanism of internal and external network data, and adopts Md5 and hardware encryption software and hardware encryption technology. It has no high requirements for work terminal computers, which can meet system requirements and be easy to use and maintain. It also comes with a WEB display and application platform, which can better serve the society. It can ensure the sustainable utilization of data for business needs at each stage without special processing, so as to ensure that customers can obtain greater benefits with smaller investment. Each system can complete various tasks by calling and modifying the basic library, avoiding repeated operations, reducing work intensity, shortening work cycle and improving work efficiency. The system opens a standard data interface and can receive and support data such as andro It can communicate with id applications, ios applications, WeChat public accounts, etc., and has a data sharing interface, which can be adjusted and expanded with the business. It adopts a three-tier modular development model to facilitate various upgrades of system business, establish a reliable off-site data backup mechanism, and adopt the most mature and popular development framework. It has monitoring, tracking and remote management functions, and is strictly tested in accordance with the quality standards of system development. It can give error reports and processing methods for possible software and hardware failures. It closely revolves around the internal management and actual situation of the power generation industry, and designs an information system with simple operation, reasonable structure, smooth operation and strong adaptability according to the characteristics of different business processes. The client only needs to access it through a browser, which is safe and convenient. It adopts the middleware mode, can be applied across platforms, can adapt to various large-scale relational databases, and has a standard data interface to meet customer needs. This system is simple and consistent to operate, and has a powerful error data judgment mechanism to reduce the workload of operators.

[0123] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0124] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. An intelligent online monitoring device for brush holder carbon brushes of a hydro-turbine generator, comprising a brush cover (1) and a collector ring (2), characterized in that: A collector ring (2) is installed in the brush cover (1), a brush holder assembly (5) is connected to the collector ring (2), a brush holder (6) is installed on the brush holder assembly (5), and a carbon brush (4) and a monitoring device (3) are installed on the brush holder (6).

2. According to claim 1, the intelligent online monitoring device for carbon brushes of a hydraulic generator brush holder is characterized in that: The monitoring device (3) comprises a temperature measuring device (31), a displacement sensor (32), a current sensor (33), a bolt fixing position (34), a brush handle fixing position (35), a main machine position cabin (36) and a communication gateway. The temperature measuring device (31), the displacement sensor (32) and the current sensor (33) are all arranged on the outside of the main machine position cabin (36). A brush handle fixing position (35) is arranged at the bottom of the main machine position cabin (36). A bolt fixing position (34) is arranged in the main machine position cabin (36) below the temperature measuring device (31). The main machine position cabin (36) is connected to the communication gateway.

3. The intelligent online monitoring device for carbon brushes of a brush holder of a hydro-generator according to claim 2 is characterized in that: The temperature measuring device (31) comprises a main panel (311), an auxiliary laser locator (312), a temperature measuring sensor (313) and a power supply interface (314); the main panel (311) is provided with the auxiliary laser locator (312) and the temperature measuring sensor (313); and the side of the main panel (311) is provided with the power supply interface (314).

4. An intelligent online monitoring system for brush holder carbon brushes of a hydro-turbine generator, applied to the intelligent online monitoring device for brush holder carbon brushes of a hydro-turbine generator as claimed in claim 3, characterized in that: include: A monitoring module, used to monitor the operating temperature of each carbon brush (4), the moving distance of the carbon brush (4) during operation, and the current of the carbon brush (4) in real time, and upload the monitored data to a brush monitoring platform; A 3D display module is used to receive data monitored by the monitoring module, and to display the real-time operating status of the carbon brush (4) in a 3D format using the monitoring data, and to visually display the distribution point layout of the carbon brush (4) in the collector ring (2); System management module, used to create, edit and delete menu items, manage permissions for each function item, and manage roles, organizations and personnel; The statistical analysis module is used to display the temperature monitoring curve, displacement monitoring curve and current monitoring curve, and to analyze the health index of the hydro-generator set according to the temperature, displacement and current information; The monitoring and alarm module is used to trigger corresponding alarms according to the analysis results, and register the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, it selects the given handling method and describes the specific handling content, and records the handler and handling time.

5. The intelligent online monitoring system for carbon brushes of a hydraulic generator brush holder according to claim 4 is characterized in that: The monitoring module comprises: A temperature measurement module, used to accurately collect the surface temperature of the exposed area of ​​the carbon brush (4) using a temperature measurement device (31); A distance measurement module, used to monitor the movement distance of the carbon brush (4) when the carbon brush (4) is running by using a displacement sensor (32) and laser distance measurement, and to determine the wear amount of the carbon brush (4) according to the monitored movement distance; The current measurement module is used to collect excitation current using a current sensor (33) and Hall contactless sensing.

6. The intelligent online monitoring system for carbon brushes of a hydraulic generator brush holder according to claim 4 is characterized by: The system management module comprises: Menu management module, used to create, edit and delete menu items, as well as manage the permissions of each function item; The role management module is used to bind role functions according to the needs of the power plant and complete the role function comparison. After the specific personnel are assigned roles, they can manage the users in their jurisdiction; The organization management module is used to display the organization tree, fold, expand, add sub-organizations, edit sub-organizations, and delete sub-organizations; The personnel management module is used to manage the information of system administrators, plant administrators and department unit administrators, bind personnel to role information, and bind them to the organization information sheet number.

7. The intelligent online monitoring system for carbon brushes of a hydraulic generator brush holder according to claim 4 is characterized by: The monitoring and alarm module specifically includes: Temperature alarm: alarm for over-high temperature of carbon brush (4), set low temperature lower limit, high temperature upper limit, and prompt temperature rising status according to the current collected temperature and the last two collected temperatures; Current alarm: Set the normal current range to issue high current alarm, excessive current difference alarm and excessive current range alarm; Displacement alarm: Set the initial displacement value. After clicking the initial button, the current displacement value is stored. During the operation of the machine, the displacement of the carbon brush (4) is monitored. When the displacement exceeds the limit of 50 mm, an alarm is triggered. MCU settings: set the reporting cycle, alarm interval, and whether the alarm, voice, SMS and other interfaces need to be configured within a certain period of time after the alarm is triggered; Alarm handling: register the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, select the given handling method and describe the specific handling content, and record the person who handled it and the handling time.

8. The intelligent online monitoring system for carbon brushes of a hydraulic generator brush holder according to claim 5 is characterized by: The monitoring module is also used to monitor the vibration data and audio data of the carbon brush; The statistical analysis module specifically includes: A data preprocessing submodule, used for denoising and normalizing the real-time monitoring data received from the monitoring module, wherein the real-time monitoring data includes the operating temperature, moving distance, current, vibration data and audio data of the carbon brush; The data fusion submodule is used to use data fusion technology to fuse the operating temperature, moving distance, current, vibration data and audio data of the carbon brush to evaluate and obtain the status information of the carbon brush; The visualization output submodule is used to construct a three-dimensional physical model of the carbon brush, associate the status information of the carbon brush with the three-dimensional physical model, and display the real-time status of the carbon brush using the three-dimensional physical model of the carbon brush through the 3D display module; The optimization submodule is used to optimize the real-time status of the carbon brush before display. After the optimization is completed, it is provided to the 3D display module for simulation and dynamic display using the carbon brush three-dimensional physical model.

9. The intelligent online monitoring system for carbon brushes of a hydraulic generator brush holder according to claim 8 is characterized by: The data fusion submodule performs fusion processing on the real-time monitoring data in the following specific ways: Extract statistical characteristic parameters of the operating temperature, moving distance and current data of the carbon brush, the statistical characteristic parameters including mean, variance, maximum value and minimum value; Extracting frequency domain features of vibration data and audio data of the carbon brush, that is, extracting vibration frequency and audio frequency respectively by performing Fourier transform on the vibration data and audio data; Extracting the time domain features of the vibration data and the audio data of the carbon brush, that is, respectively extracting the vibration time domain waveform features and the audio time domain waveform features by analyzing the signal time domain waveforms of the vibration data and the audio data; Perform vector transformation on the statistical feature parameters, frequency domain features and time domain features, and connect the transformed feature vectors to form a fused feature vector; Based on the conditional probability and prior probability of each sensor that collects real-time monitoring data, the posterior probability of the event is calculated using the following formula: Among them, P(A i |B) means that A is known to have occurred when event B occurs i The posterior probability of an event occurring, where A1…A i …A n is an event group; P(A i ) means A i Prior probability of an event occurring; P(B|A i ) indicates that A is known i The conditional probability of event B occurring when event A occurs; P(A j ) means A j Prior probability of an event occurring; P(B|A j ) indicates that A is known j The conditional probability of event B occurring when event B occurs, j = 1, 2, ..., n; A deep learning model based on recurrent neural network is constructed. The deep learning model is used to process the fused feature vector and the calculated posterior probability to identify the status information of the carbon brush, which includes the operating status and potential faults.

10. An intelligent online monitoring method for brush holder carbon brushes of a hydro-turbine generator, implemented based on the intelligent online monitoring system for brush holder carbon brushes of a hydro-turbine generator according to claim 7, characterized in that: The following steps are involved: Step 1: Initialize the unit. The system administrator and plant administrator log in to the system, add and edit the list of turbine generator units, and set the effective number of layers and the number of carbon brushes (4) per layer. Step 2: During the operation of the carbon brush (4), the surface temperature of the exposed area of ​​the carbon brush (4), the moving distance and the excitation current are collected in real time by the monitoring device (3), and the wear amount of the carbon brush (4) is determined according to the moving distance, and the real-time collected information is uploaded to the 3D display module through the communication gateway; Step 3: The 3D display module receives the data monitored by the monitoring module, and displays the real-time operation status of the carbon brush (4) in 3D format, and visually displays the distribution point layout of the carbon brush (4) in the collector ring (2); Step 4: Plot the temperature monitoring data, displacement monitoring data and current monitoring data into temperature monitoring curves, displacement monitoring curves and current monitoring curves through the statistical analysis module, and perform health index analysis on the hydro-generator set according to the temperature, displacement and current information; Step 5: The monitoring alarm module triggers the corresponding alarm according to the analysis results, and registers the alarm information serial number, equipment number, identification name, alarm type, alarm content, and alarm time. During the handling process, the given handling method is selected and the specific handling content is described, and the handling person and handling time are recorded; Step 6: When the monitoring device (3) needs maintenance, the carbon brush (4) of the hydro-generator set that needs maintenance is initialized, the number of layers where the carbon brush (4) is located, the number of carbon brushes (4) on each layer, and the position information of each carbon brush (4) are bound, and basic information maintenance is performed on the temperature sensor (313), the displacement sensor (32) and the current sensor (33).

Images