Oil quality evaluation method based on oil multi-parameter online monitoring data
By obtaining and processing multi-parameter data of lubricant oil in real time and calculating the oil quality index using European distances, the problem of inability to monitor the lubricant oil quality in the existing technology is solved, and the timely replacement of lubricant oil for heavy-duty trains is achieved, ensuring that Safe operation of the train.
Patent Information
- Application Number
- CN202411205862.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-12-06
AI Technical Summary
The prior art cannot effectively monitor and evaluate the multi-parameters of the lubricant oil of the locomotive gearbox in real time, resulting in a decrease in lubricating effect, increasing friction on the gear meshing surface, and intensifying wear, affecting the safe operation of heavy-duty trains.
The evaluation method based on the multi-parameter online monitoring data of oil is adopted, and the viscosity, moisture value, mechanical impurities and other data of lubricating oil are obtained in real time. The viscosity is converted using the Arrhenius formula, the data is normalized, and the oil quality is calculated using the European distance. Evaluate index to achieve real-time quality evaluation.
Real-time, objective and accurate evaluation of the quality of lubricant, timely replacement of lubricant oil, and ensure the driving safety of heavy-duty trains.
Smart Images

Figure CN119086883A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to an oil quality evaluation method, and in particular relates to an oil quality evaluation method based on oil multi-parameter online monitoring data. Background Art
[0002] Heavy-loaded trains are towed by locomotives. The locomotive's running gear gearbox, as a key transmission hub, directly affects the traction efficiency and operational safety of heavy-loaded trains. As heavy-loaded trains are subjected to long-term high-intensity, long-distance continuous operations, the working environment inside the gearbox has become particularly complex and harsh. The gears in the gearbox frequently mesh and rub, causing the metal particles, dust and other impurities generated by wear to increase continuously. At the same time, it also causes the temperature of the lubricating oil to rise, causing the lubricating oil to deteriorate continuously, causing the viscosity of the lubricating oil to decrease and the water content to increase, which in turn affects the lubrication effect, causing the lubricating film between the gear meshing surfaces to rupture, causing the gear surfaces in direct contact to bear greater friction and wear. In the long run, the meshing accuracy of the gears is impaired, the wear is aggravated, and it may eventually cause gearbox failure, posing a serious threat to the safe operation of heavy-loaded trains.
[0003] At present, the condition of the lubricating oil in the gearbox of the locomotive running gear is only observed manually through a glass transparent observation window or an oil dipstick to observe the lubricating oil level at regular intervals, while the viscosity, moisture content, mechanical impurities and other parameters of the oil can only be measured offline by sampling from the gearbox of the locomotive running gear. Therefore, it is of great significance to build an online monitoring system for multiple parameters of the oil, evaluate the quality of the oil using real-time monitoring data, and grasp the quality of the oil in real time, so as to provide guidance for replacing the lubricating oil in the gearbox and ensure the driving safety of heavy-load trains. Summary of the invention
[0004] In view of the shortcomings in the current online quality evaluation of locomotive gearbox lubricating oil, the present invention discloses an oil quality evaluation method based on oil multi-parameter online monitoring data.
[0005] The present invention adopts the following technical solution: A method for evaluating oil quality based on oil multi-parameter online monitoring data comprises the following steps:
[0006] Step 1: Real-time data acquisition
[0007] According to the lubricating oil change standard, obtain the viscosity at 40°C or (%) and moisture value V m (%), mechanical impurities I M (%) data; through the constructed oil multi-parameter online monitoring system, the oil level when adding new oil and when working stably is obtained l (mm), oil temperature T (℃), moisture valueV m (%), viscosity or (cst), mechanical impurities I M (%) Real-time data of 5 oil parameters;
[0008] Step 2: Data conversion and normalized evaluation vector determination
[0009] (1) Oil viscosity conversion
[0010] According to the Arrhenius formula, the oil temperature when new oil is added is T Viscosity at 0 or 0 is converted to oil temperature T Viscosity or T The specific conversion formula is: (1) (2)
[0011] In formula (1)-(2), α is the viscosity-pressure coefficient, P For oil pressure, r is the oil density, g is the acceleration due to gravity, h is the oil level, β is the viscosity-temperature coefficient;
[0012] But for lubricating oil, at low pressure ( P< 2MPa), without considering the effect of oil pressure on viscosity, the conversion formula is simplified to (3)
[0013] Then calculate the oil temperature T The specific calculation formula is: (4)
[0014] (2) Determination of normalized evaluation vector
[0015] The oil temperature at 40°C when the oil change standard is reached is obtained from the oil change standard data. , , Parameter data, real-time data of oil parameters obtained through the oil multi-parameter online monitoring system. Since the gearbox lubricating oil is approximately greater than 1 atmospheric pressure, 1 atmospheric pressure is 0.101325MPa, the relevant data is substituted into formula (3) and converted to the viscosity at 40°C, so that the oil when adding new oil can be obtained. , , Parameter data, oil during stable operation , , Parameter data;
[0016] In order to eliminate the dimensional differences between different parameters of the oil and make them comparable, the data are normalized to [0, 1]. The specific normalization calculation formula is: (5)
[0017] In the formula, X The oil parameters have reached the standard data for oil change. x The oil parameters reach the oil change standard data or actual measured data;
[0018] Will , , The parameter data are substituted into formula (5) for normalization, and the multi-parameter normalized vector of the oil at 40°C is obtained when the oil change standard is reached: A =(1, 1, 1); and , and , and Substitute the parameter data into formula (4) to obtain the normalized vector of multiple parameters of the oil when adding new oil: ;Will and , and , and Substitute the parameter data into formula (4) to obtain the normalized vector of multiple parameters of the oil under stable operation: ;
[0019] Step 3: Evaluation threshold calculation
[0020] The Euclidean distance is used to measure the similarity between two vectors. The smaller the distance, the higher the similarity. The Euclidean distance calculation formula is: (6)
[0021] In the formula, the vector X 1=( x 11 , x 12 , …, x 1n ),vector X 2=( x 21 , x 22 , …,x 2n );
[0022] Will X 1= B , X 2= A Substituting into formula (6), we can get the Euclidean distance between the two vectors when new oil is added and when the oil change standard is reached: D max , since the oil multi-parameter vectors differ the most when new oil is added and when the oil change standard is reached, D max As a threshold for evaluating oil quality;
[0023] Step 4: Oil quality evaluation
[0024] Will X 1= B , X 2= C Substituting into formula (6), we can obtain the Euclidean distance between the multi-parameter vector of the oil when adding new oil and when working stably: D W , and the oil quality evaluation index is calculated according to formula (7): (7)
[0025] when l The smaller the oil, the better the oil quality. l =0%, indicating that the oil is new oil. l =100%, indicating that the oil quality is the worst, and the oil moisture value, viscosity, and mechanical impurities have reached the oil change standard; when 0%≤ l When ≤30%, the oil quality is good; when 30%< l When ≤60%, the oil quality is good; when 60%< l When ≤80%, the oil quality is average; when 80%< l When ≤100%, the oil quality is poor and should be paid special attention to. Some parameters of the oil have reached the oil change standard and oil change is required in time.
[0026] The beneficial effects of the present invention are as follows: the oil level, oil temperature, moisture content, viscosity, and mechanical impurity data of the lubricating oil are obtained in real time through the oil multi-parameter online monitoring system, and the oil quality is evaluated by using the nearest neighbor learning classification method using the real-time monitoring data. The evaluation is real-time, objective and accurate, and the process is simple. The oil quality status can be grasped in real time, the gearbox lubricating oil can be replaced in time, and the driving safety of heavy-loaded trains can be ensured. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 This is a flow chart of oil quality evaluation based on multi-parameter online monitoring data of oil in the present invention.
[0028] Figure 2 This is a topological block diagram of the oil multi-parameter online monitoring system of the present invention.
[0029] Figure 3 The figure is a principle block diagram of the oil composite sensor of the present invention. DETAILED DESCRIPTION
[0030] The scheme of the present invention is further described in detail below in conjunction with the drawings and embodiments.
[0031] See attached figure, Figure 1 The present invention is a flow chart of oil quality evaluation based on oil multi-parameter online monitoring data, an oil quality evaluation method based on oil multi-parameter online monitoring data, comprising the following four steps: (1) data acquisition, (2) data processing, (3) evaluation threshold calculation, and (4) oil quality evaluation.
[0032] Step 1: Real-time data acquisition
[0033] The gearbox lubricant of HXD3 electric locomotive was selected for oil quality evaluation. HXD3 has 1 locomotive and 2 bogies. Each bogie has 3 axles, and each axle is equipped with a gearbox, for a total of 6 gearboxes. Mobil Lobo GX lubricant is used, model 80W-90, the new oil viscosity at 40°C is 135cst, and the density is 881.0kg / m³.
[0034] Construct a multi-parameter online monitoring system for oil, such as Figure 2 The oil multi-parameter online monitoring system includes n The oil compound sensor based on CAN bus, CAN bus and monitoring host are three parts. Since the HXD3 electric locomotive has 6 gearboxes, n The value is 6. The 6 oil composite sensors are connected to the monitoring host through the CAN bus to realize data transmission between the oil composite sensors and the monitoring host.
[0035] See attached figure, Figure 3The oil composite sensor of the present invention is a principle block diagram, the oil composite sensor comprises a main control unit, an oil level front-end measuring element, an isolation chip, an amplifier chip, an oil temperature front-end measuring element, a moisture value and mechanical impurity front-end measuring element, a gain buffer chip, a capacitance measuring chip, a viscosity front-end measuring element, an impedance measuring chip, a CAN bus transceiver module, and a voltage-stabilized power supply module; an STM32F103C8T6 single-chip microcomputer is selected as the main control chip; a thin film differential pressure sensor FSR406, an operational amplifier LM324AN, and an operational amplifier MCP6001 are selected as the oil level front-end measuring element, the isolation chip, and the amplifier chip, respectively; the thin film differential pressure sensor FSR406 is connected to the operational amplifier LM324AN, The operational amplifier LM324AN is connected to the operational amplifier MCP6001, which is connected to the main control unit. The thin film differential pressure sensor FSR406 converts the pressure signal of the lubricating oil on it into a voltage signal, which is isolated and buffered by the LM324AN operational amplifier to obtain a more stable voltage signal. The input voltage signal is then amplified by the MCP6001 operational amplifier, and finally the amplified signal is sent to the main control unit for data processing to obtain the oil level. The SMD digital temperature sensor DS18B20 is selected as the front-end measurement element of the oil temperature. The DS18B20 is connected to the main control unit. The sensor is a single-bus digital temperature sensor, and its control commands and data are in digital format. The main control unit reads the temperature data by sending a control command to DS18B20; bipolar plate capacitors, operational amplifier AD8515, and capacitance digital converter AD7745 are selected as the moisture value and mechanical impurity front-end measurement elements, gain buffer chips, and capacitance measurement chips, respectively. The two ends of the bipolar plate capacitors are connected to the operational amplifier AD8515 and the capacitance digital converter AD7745, respectively. The capacitance digital converter AD7745 is connected to the main control unit, and a low-impedance driving signal source is formed after gain buffering by the operational amplifier AD8515 to ensure that the bipolar plate capacitors are fully charged when AD7745 starts sampling, and the bipolar plate capacitors are measured by the capacitance digital converter AD7745 chip. The data is sent to the main control unit for data processing to obtain the moisture value and mechanical impurity content of the oil; a quartz tuning fork and an impedance converter AD5933 are selected as the viscosity front-end measurement element and the impedance measurement chip respectively. The quartz tuning fork is connected to the impedance converter AD5933, and the impedance converter AD5933 is connected to the main control unit. The frequency signal generated by the quartz tuning fork X3 is used to excite the external complex impedance. The response signal of the external impedance is sampled by the ADC on the AD5933 chip, and then processed by discrete Fourier transform DFT. A real data word and an imaginary data word are returned at each frequency, and the data is sent to the main control unit to calculate the impedance amplitude and relative phase of each scanning frequency point, and then the oil viscosity is obtained by calibration;The high-speed CAN transceiver TJA1050 is selected as the CAN bus transceiver module. The high-speed CAN transceiver TJA1050 is connected to the main control unit. The processed data is transmitted to the monitoring host through the CAN bus transceiver module. The oil quality evaluation method based on the oil multi-parameter online monitoring data is used to analyze the data, evaluate the oil quality, and obtain the oil degradation status. ;
[0036] The oil change standard for HXD3 electric locomotive gearbox lubricating oil is that the viscosity (cst) at 40°C varies by more than ±20% of the new oil, the moisture value (%) is ≥0.5, and the mechanical impurities (%) is ≥0.5. 0.5, For 20, The moisture content of the new oil added to the gearbox is less than 0.03%, and the mechanical impurities are less than 0.01%; the oil temperature of the gearbox is usually in the range of 60℃-80℃ when it is working stably; the oil multi-parameter online monitoring system is constructed to obtain the first k Oil level when adding new oil to the gearbox (mm), oil temperature (℃), moisture value (%), viscosity (cst), mechanical impurities (%), oil level during stable operation (mm), oil temperature (℃), moisture value (%), viscosity (cst), mechanical impurities (%), where 1≤ k ≤6.
[0037] Step 2: Data conversion and normalized evaluation vector determination
[0038] Substituting the relevant data into formula (3) Convert the viscosity at (°C) to the viscosity at 40°C , Substituting the relevant data into formula (3) The viscosity at ℃ is converted to the viscosity at 40℃ ;
[0039] Will , , The parameter data are substituted into formula (4) for normalization, and the multi-parameter normalized vector of the oil at 40°C is obtained when the oil change standard is reached: A =(1, 1, 1); Since the moisture content of the new lubricating oil added to the gearbox is less than 0.03% and the mechanical impurities are less than 0.01%, =0.03%, =0.01%, and from formula (3) we can get =0, and , and , and Substitute the parameter data into formula (4) to obtain k Normalized vector of multiple parameters of oil when new oil is added to a gearbox = (0.06, 0, 0.02); and , and , and Substitute the parameter data into formula (4) to obtain k Normalized vector of multiple parameters of oil when a gearbox is working stably , the six gearbox lubricating oil multi-parameter normalized vectors are C 1=(1,0.8,0.76)、 C 2=(0.5,1,0.9)、 C 3=(0.7,0.8,1)、 C 4=(0.9,0.7,0.7)、 C 5=(0.8,0.85,0.7)、 C 6=(1, 1, 1).
[0040] Step 3: Evaluation threshold calculation
[0041] Will X 1= B k , X 2= A Substituting into formula (6), we get k Gearbox lubricant oil quality evaluation threshold D maxk =1.6864;
[0042] Step 4: Oil quality evaluation
[0043] Will X 1= B k , X 2= C k Substituting into formula (6), we can calculate k The Euclidean distance of the oil multi-parameter vector of a gearbox when new oil is added and when it is working stably D W , 6 gearboxes D WThey are 1.4791, 14.029, 1.4177, 1.2876, 1.3162, and 1.6864, respectively. According to formula (7), the oil quality evaluation indexes are 87.70%, 83.19%, 84.07%, 76.35%, 78.05%, and 100%, respectively. The lubricating oils of the 1st, 2nd, 3rd, and 6th gearboxes are of poor quality. The water content of the lubricating oil of the 1st gearbox, the viscosity of the lubricating oil of the 2nd gearbox, the mechanical impurities of the lubricating oil of the 3rd gearbox, and the three parameters of the lubricating oil of the 4th gearbox all meet the oil change standard and need to be changed. The lubricating oils of the 4th and 5th gearboxes are of average quality and can continue to be used.
[0044] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. In an oil quality evaluation method based on oil multi-parameter online monitoring data, it is characterized in that: The oil multi-parameter online monitoring system is used to obtain the lubricating oil level, oil temperature, moisture value, viscosity, and mechanical impurity data in real time, and the oil quality evaluation index is obtained by using the nearest neighbor learning classification method. λ The evaluation results are obtained in four steps: Step 1: real-time data acquisition; Step 2: data conversion and normalized evaluation vector determination; Step 3: evaluation threshold calculation; Step 4: oil quality evaluation; Step 1: Real-time data acquisition According to the lubricating oil change standard, obtain the viscosity at 40°C η (%) and moisture value V m (%), mechanical impurities I M (%) data; through the constructed oil multi-parameter online monitoring system, the oil level when adding new oil and when working stably is obtained l (mm), oil temperature T (℃), moisture value V m (%), viscosity η (cst), mechanical impurities I M (%) Real-time data of 5 oil parameters; Step 2: Data conversion and normalized evaluation vector determination (1) Oil viscosity conversion According to the Arrhenius formula, the oil temperature when new oil is added is T Viscosity at 0 η 0 is converted to oil temperature T Viscosity η T The specific conversion formula is: (1) (2) In formula (1)-(2), α is the viscosity-pressure coefficient, P For oil pressure, ρ is the oil density, g is the acceleration due to gravity, h is the oil level, β is the viscosity-temperature coefficient; But for lubricating oil, at low pressure ( P< 2MPa), without considering the effect of oil pressure on viscosity, the conversion formula is simplified to (3) Then calculate the oil temperature T The specific calculation formula is: (4) (2) Determination of normalized evaluation vector The oil temperature at 40°C when the oil change standard is reached is obtained from the oil change standard data. , , Parameter data, real-time data of oil parameters obtained through the oil multi-parameter online monitoring system. Since the gearbox lubricating oil pressure is about higher than 1 atmosphere, 1 atmosphere is 0.101325MPa, the relevant data is substituted into formula (3) and converted to the viscosity at 40°C, so as to obtain the oil viscosity when adding new oil. , , Parameter data, oil during stable operation , , Parameter data, In order to eliminate the dimensional differences between different parameters of the oil and make them comparable, the data are normalized to [0, 1]. The specific normalization calculation formula is: (5) In the formula, X The oil parameters have reached the standard data for oil change. x The oil parameters reach the oil change standard data or actual measured data; Will , , The parameter data are substituted into formula (5) for normalization, and the multi-parameter normalized vector of the oil at 40°C when the oil change standard is reached is obtained: A =(1, 1, 1); and , and , and Substitute the parameter data into formula (4) to obtain the normalized vector of multiple parameters of the oil when adding new oil: ;Will and , and , and Substitute the parameter data into formula (4) to obtain the normalized vector of multiple parameters of the oil under stable operation: ; Step 3: Evaluation threshold calculation The Euclidean distance is used to measure the similarity between two vectors. The smaller the distance, the higher the similarity. The Euclidean distance calculation formula is: (6) In the formula, the vector X 1=( x 11 , x 12 , …, x 1n ),vector X 2=( x 21 , x 22 , …, x 2n ); Will X 1= B , X 2= A Substituting into formula (6), we can get the Euclidean distance between the two vectors when new oil is added and when the oil change standard is reached: D max , since the oil multi-parameter vectors differ the most when new oil is added and when the oil change standard is reached, D max As a threshold for evaluating oil quality; Step 4: Oil quality evaluation Will X 1= B , X 2= C Substituting into formula (6), we can obtain the Euclidean distance between the multi-parameter vector of the oil when adding new oil and when working stably: D W , and the oil quality evaluation index is calculated according to formula (7): (7) when λ The smaller the oil, the better the oil quality. λ =0%, indicating that the oil is new oil. λ =100%, indicating that the oil quality is the worst, and the oil moisture value, viscosity, and mechanical impurities have reached the oil change standard; when 0%≤ λ When ≤30%, the oil quality is good; when 30%< λ When ≤60%, the oil quality is good; when 60%< λ When ≤80%, the oil quality is average; when 80%< λ When ≤100%, the oil quality is poor and should be paid special attention to. Some parameters of the oil have reached the oil change standard and oil change is required in time.
