Method and system for monitoring oil mist of vertical water-turbine generator set

Abstract

According to the vertical water-turbine generator set oil mist monitoring method and system, multiple types of sensors are arranged in key areas such as a thrust bearing oil groove, the interior of a wind tunnel and a connecting channel, concentration, temperature, flow velocity and pressure data of oil mist are collected, and data quality is improved through wavelet denoising, Kalman filtering and normalization processing. Furthermore, a neural network fusion algorithm is combined with an oil-gas two-phase flow model to evaluate the oil mist state, abnormity judgment and positioning are achieved, and a positioning algorithm is based on the time difference of a sensor array and the triangulation positioning principle. The early warning information comprises abnormal types, degrees and positions, is sent out in various forms such as sound and light, short messages and the like, and is transmitted to a monitoring center through a 5G or WiFi network, so that remote operation and maintenance management is realized. The method can be widely applied to industrial scenes such as hydroelectric power stations, and the accuracy, the real-time performance and the intelligent level of oil mist monitoring are improved.

Description

Technical Field

[0001] This invention relates to the field of oil mist monitoring, and more particularly to a method and system for monitoring oil mist in a vertical hydro-generator unit. Background Technology

[0002] Vertical turbine generators are common power generation equipment in hydropower projects. Their main shaft is arranged vertically and driven by a turbine to generate electricity. They are mainly composed of a stator, rotor, bearings, and frame. The rotor magnetic poles use permanent magnets or electric excitation. The stator windings cut magnetic field lines to generate electrical energy. They are suitable for high-head hydropower stations, with single-unit capacities ranging from tens of kilowatts to millions of kilowatts. They are characterized by compact structure and stable operation. They convert water flow energy into mechanical energy through a turbine, and then into electrical energy through a generator. They are core equipment for the development of clean and renewable energy and are widely used in the field of hydropower generation.

[0003] However, the problem of oil mist in the bearings of hydro-generator units has not been fundamentally solved for a long time, resulting in a series of hazards: the overflowing oil mist combines with the dust from the wear of the carbon brush slip rings, causing a decrease in equipment insulation and triggering accidents such as excitation system failure; the oil mist may enter the air and river water, causing pollution. At present, the treatment of oil mist exhaust gas generally adopts a single centrifugal fan to extract the oil mist exhaust gas outside the production environment, but this will also cause pollution.

[0004] Existing generator oil tank anti-oil mist systems, such as the generator oil tank anti-oil mist system with application number 201910753195.X, have the following technical solution: An internal sealing structure is installed in the oil tank, dividing the oil tank space into an upper space and a lower space. The internal sealing structure completely seals the oil mist within the lower space of the oil tank.

[0005] During this process, the lower space of the oil tank is connected to the oil mist condensation device. After the oil mist condenses into liquid, most of it flows back to the oil tank, and a small portion enters the cold oil collection device. An external pump is installed on the cold oil collection device to pump the cold oil back to the oil tank. In this way, the oil mist concentration in the upper space of the oil tank is greatly reduced. Then, the oil mist in the upper space is sucked out by the oil mist suction and discharge device.

[0006] However, during the operation of vertical hydro-generator units, the lubricating oil in the bearing oil sump generates oil mist due to factors such as temperature rise and agitation of rotating parts. The escape of this oil mist not only pollutes the internal environment of the unit and reduces the insulation performance of the equipment, but may also cause equipment failures, affecting the safe and stable operation of the unit. Currently, the monitoring of oil mist in hydro-generator units faces the following problems: the monitoring points are not rationally arranged, failing to comprehensively reflect the oil mist distribution; the monitoring parameters are limited, only monitoring oil mist concentration and failing to obtain key information such as the movement state and temperature of the oil mist; and the monitoring system has a low level of intelligence, unable to provide timely warnings and locate abnormal oil mist areas. Therefore, a comprehensive, accurate, and intelligent method for monitoring oil mist in vertical hydro-generator units is urgently needed. SUMMARY

[0007] The present application aims to solve at least one of the problems in the related art.

[0008] The present application proposes a method for oil mist monitoring of vertical hydroelectric generating units to solve the problems of unreasonable arrangement of monitoring points, single monitoring parameters and low intelligent degree of monitoring systems in the background art.

[0009] Another object of the present application is to propose a system for oil mist monitoring of vertical hydroelectric generating units.

[0010] To achieve the above object, the present application proposes a method for oil mist monitoring of vertical hydroelectric generating units, comprising:

[0011] In the vertical hydroelectric generating unit, multiple types of sensors are arranged in key areas in response to the oil mist monitoring requirements; wherein the key areas include the thrust bearing oil tank, the inside of the wind tunnel, the connecting channel of the oil tank and the wind tunnel;

[0012] The multiple types of sensors are used to collect multiple parameter data in real time and perform preprocessing; wherein the multiple parameter data include the concentration, temperature, flow rate and pressure data of the oil mist;

[0013] A multi-parameter fusion algorithm is used to analyze the preprocessed data to establish an oil mist state evaluation model;

[0014] Based on the output results of the oil mist state evaluation model, it is determined whether the oil mist is abnormal, and if it is determined to be abnormal, a warning signal is triggered;

[0015] In response to the warning signal, a positioning algorithm is used to determine the position of the oil mist abnormal area, the positioning algorithm is based on the time difference of the abnormal signal received by the sensor array, and the triangular positioning principle is used to realize the positioning of the abnormal area;

[0016] The monitoring data and the warning information are transmitted to the monitoring center through a wireless communication network, and the warning information includes the type, degree and position information of the oil mist abnormality.

[0017] The method for oil mist monitoring of vertical hydroelectric generating units of the present application embodiment can further have the following additional technical features:

[0018] Preferably, the positions of the vertical hydroelectric generating unit prone to oil mist include the thrust bearing oil tank, the lower guide bearing oil tank, the inside of the wind tunnel, the connecting channel of the oil tank and the wind tunnel and the oil mist collecting device.

[0019] The technical scheme is adopted, sensors are arranged at key positions such as a thrust bearing oil groove and a lower guide bearing oil groove, the oil mist prone area can be comprehensively covered, the monitoring data can truly reflect the oil mist distribution, the abnormal oil mist leakage caused by the missing monitoring point is avoided, and the monitoring coverage is improved.

[0020] Preferably, the oil mist concentration sensor adopts a laser scattering principle, the temperature sensor is a thermocouple sensor, the flow rate sensor is an ultrasonic flow rate sensor, and the pressure sensor is a piezoresistive pressure sensor.

[0021] The technical scheme is adopted, the oil mist concentration sensor adopting the laser scattering principle has improved detection accuracy, the thermocouple sensor has reduced temperature measurement error, the ultrasonic flow rate sensor has reduced speed measurement error, and the multiple types of high-precision sensors ensure the accuracy of data acquisition and provide a reliable basis for oil mist state evaluation.

[0022] Preferably, in the preprocessing process, wavelet transform denoising algorithm is used for denoising, Kalman filter algorithm is used for filtering, and minimum-to-maximum normalization method is used for normalization processing.

[0023] The technical scheme is adopted, the wavelet transform denoising algorithm can effectively remove high-frequency interference, the Kalman filter algorithm can reduce data fluctuation, the minimum-to-maximum normalization processing can unify the data dimension, the error of the preprocessed data is reduced, and the stability of the subsequent analysis model is improved.

[0024] Preferably, the multi-parameter fusion algorithm adopts a fusion model based on a neural network, the neural network includes an input layer, a hidden layer and an output layer, the input layer is oil mist concentration, temperature, flow rate and pressure data, the output layer is an oil mist state evaluation result, the oil mist state evaluation includes analysis and calculation of an oil mist flow field, and the mass conservation equation used for the analysis and calculation of the oil mist flow field is:

[0025]

[0026] wherein, p is the density of the oil mist, t is time, Sm is the mass added to the continuous phase, and the evaporation of the oil mist becomes a source term.

[0027] The technical scheme is adopted, the multi-parameter fusion algorithm based on the neural network analyzes the oil mist flow field in combination with the mass conservation equation, can comprehensively consider concentration, temperature, flow rate and pressure parameters, accurately evaluates the oil mist state, improves the accuracy of single parameter monitoring, and can predict the abnormal trend of the oil mist in advance.

[0028] Preferably, the process of establishing the oil mist state evaluation model comprises: collecting historical monitoring data, labeling the data, dividing the labeled data into a training set and a test set, training the neural network using the training set, verifying the trained neural network using the test set, adjusting the parameters of the neural network until the accuracy requirement is met, the calculation condition of the mist state evaluation model is oil and gas two-phase flow, and the density calculation formula of the oil and gas two-phase flow is:

[0029] ρ = αgρg+ αvρv

[0030] The viscosity calculation formula of the oil and gas two-phase flow is:

[0031] μ = αgμg+ αvμv

[0032] Wherein, α is the volume fraction, ρ is the density, the unit is kg / m3, μ is the viscosity, the unit is Pa·S, g is the air phase, and v is the oil phase.

[0033] By using the above technical scheme, the oil mist state evaluation model established by calculating the density and viscosity of the oil and gas two-phase flow can accurately simulate the motion characteristics of the oil mist under different working conditions, the prediction error of the model is reduced, and a scientific and quantitative standard for oil mist abnormality judgment is provided.

[0034] Preferably, the process of judging whether the oil mist is abnormal comprises: comparing the output result of the oil mist state evaluation model with a preset threshold value, and if the output result exceeds the threshold value, judging that it is abnormal.

[0035] By using the above technical scheme, the evaluation result is compared with the preset threshold value, whether the oil mist is abnormal can be quickly judged, the response speed is high, the abnormal situation can be found in time, and equipment damage caused by delayed early warning is avoided.

[0036] Preferably, the positioning algorithm adopts a positioning method based on a sensor array, the position of the oil mist abnormal area is determined by calculating the time difference of the abnormal signals received by each sensor and using the triangular positioning principle.

[0037] By using the above technical scheme, based on the triangular positioning principle of the sensor array, the positioning accuracy is improved, the oil mist abnormal area can be accurately locked, the staff can quickly investigate and handle, and the fault investigation time is reduced.

[0038] Preferably, the warning signal comprises sound warning, light warning and short message warning, and the warning signal contains the type, degree and position information of the oil mist abnormality.

[0039] By using the above technical scheme, the sound, light and short message multi-form warning contains the type, degree and position information of the abnormality, and can comprehensively remind the staff, avoid information omission, and improve the conveying efficiency of the warning information.

[0040] Preferably, the monitoring data and the early warning information are transmitted to the monitoring center through a wireless communication network, which is a 5G network or a WiFi network.

[0041] By using the above technical solution, data is transmitted through a 5G or WiFi network, the transmission rate is improved, the monitoring data and the early warning information are ensured to be synchronized to the monitoring center in real time, remote monitoring and management are realized, and the operation and maintenance efficiency is improved.

[0042] To achieve the above purpose, another aspect of the present application provides a vertical hydroelectric generating set oil mist monitoring system, comprising:

[0043] A data acquisition module is arranged for arranging multiple types of sensors in the thrust bearing oil tank, the wind tunnel inside, and the connecting channel of the oil tank and the wind tunnel of the vertical hydroelectric generating set, and acquiring multiple parameter data in real time through the multiple types of sensors, including concentration, temperature, flow rate, and pressure data of the oil mist;

[0044] A data preprocessing module is arranged for preprocessing the acquired multiple parameter data;

[0045] A state evaluation module is arranged for analyzing the preprocessed data using a multi-parameter fusion algorithm, establishing an oil mist state evaluation model, and analyzing and calculating the oil mist flow field in combination with an oil-gas two-phase flow model;

[0046] An abnormality judgment module is arranged for judging whether the oil mist is abnormal based on the output result of the oil mist state evaluation model, and triggering an early warning signal if it is judged to be abnormal;

[0047] A positioning module is arranged for determining the position of the oil mist abnormal area using a sensor array-based time difference and a triangular positioning principle in response to the early warning signal;

[0048] An early warning module is arranged for generating an early warning signal containing oil mist abnormal type, degree, and position information, and performing early warning;

[0049] A communication module is arranged for transmitting the monitoring data and the early warning information to the monitoring center through a wireless communication network, which is a 5G network or a WiFi network.

[0050] The vertical hydroelectric generating set oil mist monitoring method and system of the embodiments of the present application realize comprehensive perception, intelligent evaluation, and efficient management of the oil mist state through multiple parameter acquisition, data preprocessing, multi-parameter fusion analysis, physical model combination, abnormality judgment and positioning, multi-form early warning, and remote transmission, thereby improving the accuracy, response speed, and operation and maintenance efficiency of oil mist monitoring.

[0051] Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following and / or can be learned by practice of the application. Attached Figure Description

[0052] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0053] Figure 1 This is a flowchart of a method for monitoring oil mist in a vertical hydro-generator unit according to an embodiment of the present invention;

[0054] Figure 2 This is a data architecture diagram of a method for monitoring oil mist in a vertical hydro-generator unit according to an embodiment of the present invention;

[0055] Figure 3 This is a system structure diagram of oil mist monitoring for a vertical hydro-generator unit according to an embodiment of the present invention. Detailed Implementation

[0056] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0057] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0058] The method and system for monitoring oil mist in a vertical hydro-generator set according to an embodiment of the present invention are described below with reference to the accompanying drawings.

[0059] Figure 1 This is a flowchart of a method for monitoring oil mist in a vertical hydro-generator unit according to an embodiment of the present invention, as shown below. Figure 1 As shown, it includes:

[0060] S1, in the vertical hydro turbine generator set, in response to the oil mist monitoring requirements, multiple types of sensors are deployed in key areas; wherein, the key areas include the thrust bearing oil tank, the inside of the wind tunnel, and the connection channel between the oil tank and the wind tunnel;

[0061] S2, real-time acquisition of multi-parameter data through the aforementioned multi-type sensors, followed by preprocessing; wherein,

[0062] The multi-parameter data includes oil mist concentration, temperature, flow rate, and pressure data;

[0063] S3, a multi-parameter fusion algorithm is used to analyze the preprocessed data to establish an oil mist state evaluation model;

[0064] S4, based on the output result of the oil mist state evaluation model, it is judged whether the oil mist is abnormal, if it is judged to be abnormal, a warning signal is triggered;

[0065] S5, in response to the warning signal, the position of the oil mist abnormal area is determined by using a positioning algorithm, the positioning algorithm is based on the time difference of the abnormal signal received by the sensor array, and the positioning of the abnormal area is realized by using the triangular positioning principle;

[0066] S6, the monitoring data and the warning information are transmitted to the monitoring center through the wireless communication network, and the warning information includes the type, degree and position information of the oil mist abnormality.

[0067] Specifically, as Figure 2 The method architecture diagram of the oil mist monitoring of the vertical hydroelectric generator set of the application is shown, and the specific implementation process includes the following steps:

[0068] A plurality of types of sensors are arranged at positions prone to oil mist in the vertical hydroelectric generator set, the sensors include oil mist concentration sensors, temperature sensors, flow rate sensors and pressure sensors; the concentration, temperature, flow rate and pressure data of the oil mist are collected in real time by using the sensors; the collected data are preprocessed, including denoising, filtering and normalization processing; a multi-parameter fusion algorithm is used to analyze the preprocessed data to establish an oil mist state evaluation model; whether the oil mist is abnormal is judged according to the oil mist state evaluation model, and a warning signal is issued if it is abnormal; the position of the oil mist abnormal area is determined by using a positioning algorithm; the monitoring data and the warning information are transmitted to the monitoring center, so that the staff can handle it in time.

[0069] It can be understood that the positions prone to oil mist in the vertical hydroelectric generator set include the thrust bearing oil tank, the lower guide bearing oil tank, the inside of the wind tunnel, the connecting channel of the oil tank and the wind tunnel and the oil mist collecting device, the oil mist concentration sensor adopts the laser scattering principle, the temperature sensor is a thermocouple sensor, the flow rate sensor is an ultrasonic flow rate sensor, and the pressure sensor is a piezoresistive pressure sensor; the sensors are arranged at key positions such as the thrust bearing oil tank and the lower guide bearing oil tank, which can comprehensively cover the areas prone to oil mist, ensure that the monitoring data can truly reflect the oil mist distribution, avoid missing detection of oil mist abnormality due to missing monitoring points, improve the monitoring coverage rate, the oil mist concentration sensor adopting the laser scattering principle improves the detection accuracy, the thermocouple sensor reduces the temperature measurement error, the ultrasonic flow rate sensor reduces the speed measurement error, and the plurality of types of high-precision sensors ensure the accuracy of data acquisition, providing a reliable basis for oil mist state evaluation.

[0070] It can be understood that in the preprocessing process, the wavelet transform denoising algorithm is used for denoising, the Kalman filter algorithm is used for filtering, and the minimum-to-maximum normalization method is used for normalization processing. The wavelet transform denoising algorithm can effectively remove high-frequency interference, the Kalman filter algorithm reduces data fluctuation, and the minimum-to-maximum normalization processing unifies data dimension. The error of the preprocessed data is reduced, and the stability of the subsequent analysis model is improved.

[0071] It can be understood that the multi-parameter fusion algorithm adopts a fusion model based on a neural network. The neural network includes an input layer, a hidden layer and an output layer. The input layer is the oil mist concentration, temperature, flow rate and pressure data. The output layer is the oil mist state evaluation result. The oil mist state evaluation includes analysis and calculation of the oil mist flow field, and the mass conservation equation used for the analysis and calculation of the oil mist flow field is:

[0072]

[0073] wherein p is the density of the oil mist, t is the time, and Sm is the mass added to the continuous phase. The evaporation of the oil mist becomes a source term. The establishment process of the oil mist state evaluation model includes collecting historical monitoring data, labeling the data, dividing the labeled data into a training set and a test set, training the neural network using the training set, verifying the trained neural network using the test set, adjusting the parameters of the neural network until the accuracy requirement is met. The calculation condition of the mist state evaluation model is oil-gas two-phase flow, and the density calculation formula of the oil-gas two-phase flow is:

[0074] p = agpg + avpv

[0075] The viscosity calculation formula of the oil-gas two-phase flow is:

[0076] p = agpg + avpv

[0077] wherein a is the volume fraction, p is the density with the unit of kg / m3, and p is the viscosity with the unit of Pa·S. g is the air phase, and v is the oil phase. The multi-parameter fusion algorithm based on the neural network analyzes the oil mist flow field in combination with the mass conservation equation, can comprehensively evaluate the concentration, temperature, flow rate and pressure parameters, accurately evaluate the oil mist state, improve the accuracy of single parameter monitoring, predict the abnormal trend of the oil mist in advance, calculate the density and viscosity through the oil-gas two-phase flow model, and the established oil mist state evaluation model can accurately simulate the motion characteristics of the oil mist under different working conditions, reduce the model prediction error, and provide a scientific and quantitative standard for oil mist abnormality judgment.

[0078] Further, the process of judging whether the oil mist is abnormal includes: comparing the output result of the oil mist state evaluation model with a preset threshold value, and judging as abnormal if the output result exceeds the threshold value; the positioning algorithm adopts a positioning method based on a sensor array, the position of the oil mist abnormal area is determined by calculating the time difference of each sensor receiving the abnormal signal and using the triangular positioning principle, and the evaluation result is compared with the preset threshold value, so that whether the oil mist is abnormal can be quickly judged, the response speed is high, the abnormal condition can be found in time, equipment damage caused by delayed warning is avoided, the positioning accuracy is improved based on the triangular positioning principle of the sensor array, the oil mist abnormal area can be accurately locked, the staff can quickly troubleshoot and handle, and the troubleshooting time is reduced.

[0079] Further, the warning signal includes sound warning, light warning and short message warning, the type, degree and position information of the oil mist abnormality are contained in the warning signal, the monitoring data and the warning information are transmitted to the monitoring center through a wireless communication network, the wireless communication network is a 5G network or a WiFi network, the sound, light and short message multi-form warning contain the type, degree and position information of the abnormality, the staff is warned from all directions, information omission is avoided, the warning information transmission efficiency is improved, the data is transmitted by using the 5G or WiFi network, the transmission rate is improved, the monitoring data and the warning information are ensured to be transmitted to the monitoring center in real time, remote monitoring and management are realized, and the operation and maintenance efficiency is improved.

[0080] In an embodiment of the present application, the specific equipment is configured as follows:

[0081] Sensor arrangement: 18 monitoring points are arranged in the thrust bearing oil tank (4 points), the lower guide bearing oil tank (4 points), the inside of the wind tunnel (8 points) and the oil tank and wind tunnel connection channel (2 points) of each unit, and each monitoring point is installed with:

[0082] Laser scattering oil mist concentration sensor (model: OPC-3+, range 0-1000mg / m 3 , accuracy ±1%);

[0083] Armored thermocouple temperature sensor (model: K type, range 0-120℃, accuracy ±0.5℃);

[0084] Ultrasonic flow sensor (model: UFM10, range 0-30m / s, accuracy ±0.3m / s);

[0085] Piezoresistive pressure sensor (model: GP250, range -100kPa-100kPa, accuracy ±0.2%FS);

[0086] Data processing system: industrial-grade PLC (model: S7-1500) is used for data acquisition and preprocessing, and the host computer (configuration: i7-10700 / 16G / 512G SSD) runs the neural network fusion model based on TensorFlow;

[0087] Communication network: full station deployment of 5G private network, data transmission rate stable at above 300Mbps, ensuring real-time data transmission.

[0088] Based on the above implementation process and specific equipment, the principles of the present application are as follows:

[0089] Data acquisition and preprocessing: laser scattering oil mist concentration sensors, thermocouple temperature sensors, etc. are arranged in key areas such as thrust bearing oil tank and lower guide bearing oil tank, real-time acquisition of oil mist multi-parameter data, high-frequency interference is removed by wavelet transform denoising algorithm, data fluctuation is smoothed by Kalman filter, minimum to maximum normalization unified data format, ensuring the reliability of input data;

[0090] Multi-parameter fusion and state evaluation: three-layer neural network (input layer receives concentration, temperature, flow rate, pressure data, hidden layer processes feature extraction, output layer generates evaluation results) is used, combined with oil-gas two-phase flow model (ρ = αgρg + αvρv, μ = αgμg + αvμv) to calculate oil mist flow field characteristics. The model is trained by historical data (training set accounts for 70%, test set accounts for 30%), dynamically adjusts the weight, realizes the accurate evaluation of oil mist state;

[0091] Abnormal judgment and positioning early warning: compare the evaluation results with the preset threshold (such as concentration > 50mg / m 3 , temperature > 70℃), trigger early warning when exceeding the threshold. At the same time, based on the time difference (time difference measurement accuracy ≤0.1ms) of the abnormal signal received by the sensor array, the abnormal area coordinates are calculated by using the triangular positioning principle, and the early warning information (including abnormal type, degree, position) is transmitted to the monitoring center in real time through the 5G network, realizing sound and light and short message multi-form reminders.

[0092] Therefore, the key point of the present application is:

[0093] Data acquisition and preprocessing: laser scattering oil mist concentration sensors, thermocouple temperature sensors, etc. are arranged in key areas such as thrust bearing oil tank and lower guide bearing oil tank, real-time acquisition of oil mist multi-parameter data, high-frequency interference is removed by wavelet transform denoising algorithm, data fluctuation is smoothed by Kalman filter, minimum to maximum normalization unified data format, ensuring the reliability of input data;

[0094] Multi-parameter fusion and state evaluation: a three-layer neural network is used, the input layer receives concentration, temperature, flow rate and pressure data, the hidden layer processes feature extraction, and the output layer generates evaluation results, combined with the oil-gas two-phase flow model (p = alpha g p g + alpha v p v, mu = alpha g mu g + alpha v mu v) to calculate the oil mist flow field characteristics. The model is trained by historical data (70% training set, 30% test set), dynamically adjusts the weight, and realizes accurate evaluation of the oil mist state;

[0095] Abnormality judgment and positioning early warning: compare the evaluation results with the preset threshold, trigger the early warning when the threshold is exceeded, at the same time, based on the time difference of the abnormal signal received by the sensor array, calculate the abnormal area coordinates by using the triangular positioning principle, transmit the early warning information to the monitoring center in real time through the 5G network, realize the sound and light and short message multi-form reminding.

[0096] The method for monitoring oil mist of vertical hydroelectric generating set according to the embodiment of the application, through the technical means of multi-sensor cooperative arrangement, multi-parameter fusion analysis, physical model combined with neural network modeling, abnormality judgment and positioning, multi-form early warning and remote transmission, a set of efficient, intelligent and reliable vertical hydroelectric generating set oil mist monitoring system is constructed, effectively solves the problems of incomplete monitoring, inaccurate judgment and slow response in the prior art, and has significant technical progress and practical application value.

[0097] The beneficial effects of the application are as follows:

[0098] 1. Improve monitoring coverage: by arranging multiple types of sensors in key areas such as thrust bearing oil tank, lower guide bearing oil tank, wind tunnel interior and connecting channel, comprehensive coverage of oil mist generation and diffusion path is realized, effectively avoiding abnormality missed detection caused by missing monitoring points;

[0099] 2. Improve abnormality identification accuracy: laser scattering oil mist concentration sensor, thermocouple temperature sensor, ultrasonic flow rate sensor and piezoresistive pressure sensor are used, combined with wavelet transform denoising, Kalman filtering and minimum to maximum normalization processing, the accuracy of data acquisition and processing is significantly improved, providing a reliable basis for oil mist state evaluation;

[0100] 3. Enhance state evaluation capability: based on the multi-parameter fusion algorithm of neural network combined with the oil-gas two-phase flow model, the concentration, temperature, flow rate and pressure of oil mist are comprehensively analyzed, dynamic modeling of oil mist flow field is realized, so that the oil mist state is more scientifically and accurately evaluated, and the abnormality prediction capability is improved;

[0101] 4. Realize rapid abnormality judgment and response: by comparing the model output results with the preset threshold, it is quickly judged whether the oil mist is abnormal or not, the response time is significantly shortened, which is helpful for taking measures in time to prevent equipment damage;

[0102] 5. Improve abnormal positioning accuracy: adopt time difference triangle positioning algorithm based on sensor array, can accurately locate oil mist abnormal area, positioning error control in 0.5 meters, facilitate operation and maintenance personnel to quickly investigate and handle;

[0103] 6. Realize multi-form early warning and remote transmission: the early warning signal contains the type, degree and position information of oil mist anomaly, and reminds through sound, light and short message and other forms, ensures the comprehensiveness and timeliness of information transmission; At the same time, through 5G or WiFi network, the monitoring data and early warning information are transmitted to the monitoring center in real time, realizing remote monitoring and management, and improving the operation and maintenance efficiency;

[0104] 7. Reduce operation and maintenance cost and risk: through intelligent monitoring and early warning mechanism, reduce the frequency of artificial inspection, improve the fault response speed, reduce the equipment downtime risk caused by oil mist anomaly, and prolong the service life of equipment.

[0105] In order to realize the above embodiment, as Figure 3 shown, the vertical hydroelectric generating set oil mist monitoring system 10 is also provided in the embodiment, comprising:

[0106] The data acquisition module 100 is used for arranging multiple types of sensors in the thrust bearing oil tank, the wind tunnel inside, and the connecting channel of the oil tank and the wind tunnel of the vertical hydroelectric generating set, and acquiring multiple parameter data in real time through the multiple types of sensors, including oil mist concentration, temperature, flow rate and pressure data;

[0107] The data preprocessing module 200 is used for preprocessing the collected multiple parameter data;

[0108] The state evaluation module 300 is used for analyzing the preprocessed data by using a multi-parameter fusion algorithm, establishing an oil mist state evaluation model, and combining an oil-gas two-phase flow model to analyze and calculate the oil mist flow field;

[0109] The abnormality judgment module 400 is used for judging whether the oil mist is abnormal based on the output result of the oil mist state evaluation model, and triggering an early warning signal if it is judged to be abnormal;

[0110] The positioning module 500 is used for determining the position of the oil mist abnormal area by using the time difference and triangle positioning principle based on the sensor array in response to the early warning signal;

[0111] The early warning module 600 is used for generating an early warning signal containing the type, degree and position information of oil mist anomaly, and warning;

[0112] The communication module 700 is used for transmitting the monitoring data and early warning information to the monitoring center through a wireless communication network, and the wireless communication network is a 5G network or a WiFi network.

[0113] Further, the data acquisition module 100 comprises a laser scattering oil mist concentration sensor, a thermocouple temperature sensor, an ultrasonic flow rate sensor and a piezoresistive pressure sensor, each of which is used to acquire the concentration, temperature, flow rate and pressure parameters of the oil mist.

[0114] Further, the data preprocessing module 200 adopts a wavelet transform denoising algorithm, a Kalman filtering algorithm and a minimum-to-maximum normalization method to process the acquired data.

[0115] Further, the state evaluation module 300 adopts a three-layer neural network structure, the input layer of which receives the concentration, temperature, flow rate and pressure data of the oil mist, the hidden layer of which is used for feature extraction, and the output layer of which outputs the oil mist state evaluation result, the oil mist state evaluation model being combined with an oil-gas two-phase flow model to analyze and calculate the oil mist flow field.

[0116] The vertical hydro-generating unit oil mist monitoring system according to the embodiment of the application, through the technical means of multi-sensor cooperative arrangement, multi-parameter fusion analysis, physical model combined with neural network modeling, abnormality judgment and positioning, multi-form early warning and remote transmission, constructs an efficient, intelligent and reliable vertical hydro-generating unit oil mist monitoring system, effectively solves the problems of incomplete monitoring, inaccurate judgment and untimely response in the prior art, and has significant technical progress and practical application value.

[0117] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0118] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is at least two, for example, two, three, etc., unless otherwise specifically limited.

Claims

1. A method for oil mist monitoring of a vertical hydroelectric generator unit, characterized by, The method comprises the following steps: In a vertical hydroelectric generator set, multiple types of sensors are arranged in key areas in response to oil mist monitoring requirements; wherein the key areas include a thrust bearing oil tank, an internal wind tunnel, and a connecting channel between the oil tank and the wind tunnel; Real-time multi-parameter data is collected by the multiple types of sensors and preprocessed; wherein the multi-parameter data includes oil mist concentration, temperature, flow rate, and pressure data; A multi-parameter fusion algorithm is used to analyze the preprocessed data to establish an oil mist state evaluation model; Based on the output results of the oil mist state evaluation model, it is determined whether the oil mist is abnormal, and if it is determined to be abnormal, a warning signal is triggered; In response to the warning signal, a positioning algorithm is used to determine the location of the oil mist abnormal area based on the time difference of the abnormal signal received by the sensor array, and the positioning algorithm uses the triangular positioning principle to realize the positioning of the abnormal area; The monitoring data and warning information are transmitted to the monitoring center through a wireless communication network, and the warning information includes the type, degree, and location information of the oil mist abnormality.

2. The method of claim 1, wherein: The multiple types of sensors include an oil mist concentration sensor based on laser scattering principle, a thermocouple temperature sensor, an ultrasonic flow rate sensor, and a piezoresistive pressure sensor, each of which is used to collect the concentration, temperature, flow rate, and pressure parameters of the oil mist.

3. The method of claim 2, wherein: The preprocessing process includes using a wavelet transform denoising algorithm to remove high-frequency interference, using a Kalman filter algorithm to smooth data fluctuations, and using a minimum-to-maximum normalization method to standardize the data.

4. The method of claim 3, wherein: The multi-parameter fusion algorithm uses a fusion model based on a neural network, which includes an input layer, a hidden layer, and an output layer, the input layer receives the concentration, temperature, flow rate, and pressure data of the oil mist, the output layer outputs the oil mist state evaluation results, and the oil mist state evaluation model combines an oil-gas two-phase flow model to analyze and calculate the oil mist flow field.

5. The method of claim 4, wherein: The mass conservation equation used in the analysis and calculation of the oil mist flow field is: wherein p is the density of the oil mist, t is the time, Sm is the mass added to the continuous phase, and the evaporation of the oil mist becomes a source term.

6. The method of claim 5, wherein: Establishing the oil mist state evaluation model includes: Collecting historical monitoring data and labeling the data, dividing the labeled data into a training set and a test set; Training the neural network using the training set, verifying the trained neural network using the test set, adjusting the parameters of the neural network to train the oil mist state evaluation model; the calculation condition of the oil mist state evaluation model is oil-gas two-phase flow, and the density calculation formula of the oil-gas two-phase flow is: p = agpg + avpv The viscosity calculation formula of the oil-gas two-phase flow is: μ = agμg + avμv Wherein a is the volume fraction, p is the density, the unit is kg / m3, μ is the viscosity, the unit is Pa·S, g is the air phase, and v is the oil phase.

7. A vertical hydroelectric generator oil mist monitoring system, characterized by, It comprises: A data acquisition module is used to arrange multiple types of sensors in the thrust bearing oil tank, the internal wind tunnel, and the connecting channel between the oil tank and the wind tunnel of the vertical hydroelectric generator set, and to collect real-time multi-parameter data including oil mist concentration, temperature, flow rate, and pressure data through the multiple types of sensors; A data preprocessing module is used to preprocess the collected multi-parameter data; The state evaluation module is configured to analyze the preprocessed data by using a multi-parameter fusion algorithm, and to establish an oil mist state evaluation model, which is combined with an oil-gas two-phase flow model to analyze and calculate an oil mist flow field. The abnormality judgment module is configured to judge whether the oil mist is abnormal based on an output result of the oil mist state evaluation model, and to trigger a warning signal if the oil mist is judged to be abnormal. The positioning module is configured to determine a position of an abnormal area of the oil mist by using a time difference based on a sensor array and a triangular positioning principle in response to the warning signal. The warning module is configured to generate a warning signal containing information about an abnormal type, degree and position of the oil mist, and to perform a warning. The communication module is configured to transmit monitoring data and warning information to a monitoring center through a wireless communication network, which is a 5G network or a WiFi network.

8. The system of claim 7, wherein: The data acquisition module includes a laser scattering oil mist concentration sensor, a thermocouple temperature sensor, an ultrasonic flow rate sensor and a piezoresistive pressure sensor, each of which is configured to acquire concentration, temperature, flow rate and pressure parameters of the oil mist.

9. The system of claim 7, wherein: The data preprocessing module is configured to preprocess the acquired data by using a wavelet transform denoising algorithm, a Kalman filtering algorithm and a minimum-to-maximum normalization method.

10. The system of claim 7, wherein: The state evaluation module is configured to adopt a three-layer neural network structure, an input layer of which receives concentration, temperature, flow rate and pressure data of the oil mist, a hidden layer of which is configured to extract features, and an output layer of which outputs an oil mist state evaluation result, which is combined with an oil-gas two-phase flow model to analyze and calculate an oil mist flow field.

Images