Method and system for performance directed optimization of limited slip differential gear oil

Abstract

The present application relates to the technical field of intelligent oil injection control, in particular to a limited slip differential gear oil performance directional optimization method and system, comprising the following steps: calculating the gear shaft elongation through temperature sensor acquisition of temperature rise rate, generating meshing gap trend vector in combination with differential temperature time sequence analysis, calculating stability correction amount by fusing oil film thickness data, solving injection angle compensation drive servo motor based on vector projection, constructing adhesion factor index by coupling oil injection feedback and oil film difference, and switching oil products meeting tension conditions and evaluating lubrication recovery performance when exceeding limit. In the present application, by real-time acquisition of gear shaft temperature rise and differential temperature gradient data, combined with expansion coefficient analysis, precise prediction of meshing gap is realized, stability judgment is improved by fusing oil film thickness data, injection precision is optimized based on vector projection dynamic compensation of injection angle, adhesion factor index containing phase difference is constructed to realize lubrication attenuation early warning, and oil products are accurately switched by dynamically matching tension and viscosity threshold to reduce friction loss.

Description

Technical Field

[0001] This invention relates to the field of intelligent fuel injection control technology, and in particular to a method and system for targeted optimization of gear oil performance in limited-slip differentials. Background Technology

[0002] The field of intelligent oil injection control technology encompasses techniques for dynamically adjusting the lubricating oil injection process based on a control system. The core of this technology lies in utilizing sensors to perceive the equipment's operating status, including parameters such as oil temperature, speed, and load, and then using an embedded control system to calculate the injection timing, injection quantity, and injection position in real time to adapt to the lubrication needs of different mechanical components. This comprehensive system integrates sensing, control, and communication, and is widely used in industrial gears, vehicle powertrains, and construction machinery to improve lubrication efficiency and mechanical operational stability.

[0003] The targeted optimization method for gear oil performance in limited-slip differentials refers to a lubrication management approach that establishes a data association model encompassing performance parameters such as torque response, differential speed, and friction characteristics. This model allows for targeted adjustments to the gear oil formulation and rheological properties used in limited-slip differentials. Specifically, it addresses the changes in friction state between friction plates within the differential during the torque output phase. Based on collected data on shaft speed differences, load conditions, and gear meshing angular velocities, a target parameter model is established. A multi-component oil combination containing polymer modifiers and friction-regulating additives is used to directionally correct the base oil ratio and additive concentration in the formulation, thereby achieving closed-loop control and optimized selection of gear oil performance parameters. This process primarily relies on experimental test databases and operational condition sampling analysis to construct the gear oil parameter model and calculate target deviations, outputting targeted lubricant formulation adjustment schemes.

[0004] Existing intelligent fuel injection control technologies rely on static correlation models built from single-dimensional parameters such as oil temperature and speed. They lack comprehensive analysis of the thermal expansion effect of the gear shaft and the temperature field distribution of the differential housing, resulting in insufficient accuracy in predicting axial clearance changes and a tendency for blind spots in lubrication coverage under high-speed, variable-load conditions. Traditional oil film stability assessments rely solely on absolute oil film thickness for threshold judgments, neglecting the nonlinear influence of temperature gradients on the oil film decay rate. This leads to misjudgments of lubrication status during severe operating condition fluctuations, resulting in over-injection or under-lubrication. Existing injection angle control strategies are based on calculations using fixed geometric parameters, failing to incorporate real-time offset data of the gear shaft's motion trajectory. This results in deviations between the injection angle and the dynamic meshing position, affecting the uniformity of the lubricating oil film. Oil switching mechanisms rely on offline analysis of experimental databases, lacking dynamic monitoring of real-time performance indicators such as adhesion factors. This prevents timely responses to the gradual degradation of oil film adhesion performance, leading to abnormal gear wear due to oil film failure under extreme conditions. Existing lubricant formulation optimization methods primarily target frictional characteristics and fail to establish a dynamic correlation model between viscosity decay threshold and surface tension parameters, resulting in oil switching timing lagging behind actual changes in lubrication needs. Summary of the Invention

[0005] To address the shortcomings of existing intelligent fuel injection control technologies, which rely on static correlation models built from single-dimensional parameters such as oil temperature and speed, and lack comprehensive analysis of the thermal expansion effect of the gear shaft and the temperature field distribution of the differential housing, resulting in insufficient accuracy in predicting axial clearance changes and the potential for blind spots in lubrication coverage under high-speed, variable-load conditions, traditional oil film stability assessments rely solely on absolute oil film thickness for threshold judgments, neglecting the nonlinear influence of temperature gradients on the oil film decay rate. This leads to misjudgments of lubrication status during severe operating condition fluctuations, resulting in over-injection or under-lubrication. Existing injection angle control strategies are based on calculations using fixed geometric parameters, failing to incorporate real-time offset data of the gear shaft motion trajectory. This results in deviations between the injection angle and the dynamic meshing position, affecting the uniformity of the lubricating oil film. Furthermore, oil switching mechanisms rely on offline analysis of experimental databases, lacking dynamic monitoring of real-time performance indicators such as adhesion factors. This prevents timely responses to the gradual degradation of oil film adhesion performance, potentially leading to abnormal gear wear due to oil film failure under extreme conditions. Existing lubricant formulation optimization methods primarily target frictional characteristics, failing to establish a dynamic correlation model between viscosity decay threshold and surface tension parameters. This leads to a technical problem where oil switching lags behind actual changes in lubrication requirements. This invention provides a method and system for targeted optimization of limited-slip differential gear oil performance. The technical solution is as follows:

[0006] On the one hand, a method for targeted optimization of the performance of limited-slip differential gear oil is provided, which includes:

[0007] S1: The temperature rise rate is collected by the temperature sensor, and the axial elongation of the gear shaft is calculated by the linear expansion coefficient of the gear shaft. The temperature change rate of equidistant points on the inner wall of the differential housing is collected. The displacement difference direction is analyzed by calling the time offset model. The slope of the elongation change over time is extracted to generate the meshing clearance change trend vector.

[0008] S2: Call the meshing clearance change trend vector, fuse the real-time collected oil film thickness data at the gear meshing point, perform smoothing, remove outliers and calculate the derivative, identify the current oil film thickness change rate, and compare it with the critical interval to generate the oil film stability correction amount.

[0009] S3: Based on the oil film stability correction amount, the injection angle offset is calculated by the vector projection of the nozzle base coordinates and the tooth shaft motion trajectory on the tooth shaft axial reference. The offset is then input into the multi-factor nonlinear correction function to solve the compensation angle and drive the servo motor to perform fine-tuning. The injection pressure fluctuation is collected to obtain the injection feedback factor.

[0010] S4: Couple the fuel injection feedback factor and the difference amplitude of the oil film rate, and construct the adhesion factor index by normalizing the coverage change rate. When the index exceeds the upper limit of the confidence interval of the original operating condition, an adhesion level downgrade instruction is generated.

[0011] As a further aspect of the present invention, the claimed time-series offset model is based on the thermal strain-oil film attenuation coupling model to dynamically analyze the axial displacement trend caused by gear shaft temperature rise and the oil film thickness change, and to construct a nonlinear compensation function between the injection direction and the oil film stability.

[0012] When calculating parameters for the rate of change of oil film thickness and the rate of change of temperature, they are first processed to be dimensionless by normalization transformation.

[0013] The meshing clearance change trend vector specifically includes the displacement difference direction, temperature gradient, and time series slope. The oil film stability correction includes the critical rate threshold, lubricating film attenuation coefficient, and error tolerance range. The injection feedback factor includes the pressure fluctuation spectrum, injection angle deviation, and servo response delay. The adhesion factor index specifically refers to the normalized amplitude, phase difference parameter, and confidence interval weight. The adhesion level downgrade command includes the surface tension critical value, viscosity attenuation threshold, switching delay time, and oil film reconstruction time threshold.

[0014] Temperature gradient refers to the temperature change per unit length along the tooth shaft axis, which is calculated by collecting temperature field data from a temperature sensor.

[0015] The lubricating film attenuation coefficient is calculated from the difference between the rate of change of oil film thickness and the critical interval.

[0016] As a further aspect of the present invention, the specific steps of S1 include:

[0017] S101: The temperature rise rate is collected by a temperature sensor, the linear expansion coefficient of the tooth shaft is extracted, and the axial elongation of the tooth shaft is calculated by combining the relationship between the temperature rise rate and the expansion coefficient.

[0018] S102: The temperature change rate at equidistant points on the inner wall of the differential housing is synchronously acquired through a temperature sensor. The temperature change rate data is input into a time-series offset model to calculate the phase offset. The model is used to perform phase difference analysis on the temperature change rate at the differential measuring points of the housing, identify the mapping relationship between the temperature gradient and the housing deformation, and generate a displacement difference direction vector.

[0019] S103: Based on the time series data of the axial elongation of the gear shaft, calculate the slope of the elongation change per unit time, and combine it with the direction parameter of the displacement difference direction vector to perform vector synthesis operation on the slope value and the direction parameter to generate the meshing clearance change trend vector.

[0020] As a further aspect of the present invention, the timing offset model calculates the phase offset using the formula:

[0021]

[0022] Where Φ(t) represents the phase offset at time t, with units of ℃·s⁻¹. This represents the rate of temperature change at the k-th temperature measuring point, expressed in °C / s, T. k The temperature function at the k-th measuring point is represented by the function. T represents k The amount of change over a small time interval It is a very small time increment, λ represents the temperature signal attenuation coefficient, and its value ranges from 0.25 to 0.35, τ k η represents the offset of the k-th measuring point relative to the reference time, and η represents the weighting coefficient of the spatial variation of the temperature field, with a value ranging from 0.1 to 0.3. The Laplace operator represents the temperature field of the housing, p represents the total number of equidistant measuring points on the inner wall of the differential housing, e represents the natural logarithm, t represents the current time variable, k represents the measuring point number variable, ranging from 1 to p, and α represents the axial stress gradient modulation coefficient. This represents the axial stress gradient along the tooth axis at the k-th temperature measuring point, in MPa / m, σ k This represents the internal normal stress distribution function of the gear shaft material in the axial direction caused by temperature at the k-th measuring point, with units of MPa.

[0023] As a further aspect of the present invention, the specific steps of S2 include:

[0024] S201: Call the meshing clearance change trend vector to obtain the original data sequence of oil film thickness at the gear meshing point, use the moving average method to smooth the oil film thickness data, and remove outliers that deviate from the mean based on the statistical distribution criteria to obtain stable oil film data.

[0025] The statistical distribution criteria selected include the 3σ rule, the box plot interquartile range method, or a distribution fitting and elimination strategy based on skewness correction.

[0026] S202: Based on stable oil film data, extract the difference in oil film thickness between adjacent sampling points, combine it with a fixed time interval, and calculate the change per unit time using a splitting algorithm to obtain the dynamic change rate of the oil film;

[0027] S203: Call the dynamic change rate of the oil film, compare the dynamic change rate of the oil film with the boundary value of the interval point by point according to the critical interval parameter, assign state markers according to the comparison results, count the distribution ratio of different markers in the continuous sampling period, calculate the difference in the proportion of positive markers and negative markers, and obtain the oil film stability correction amount.

[0028] The critical interval is determined based on the statistical analysis of the oil film thickness change rate under the original stable operating conditions, and the threshold range is determined using a 95% confidence interval.

[0029] The oil film stability correction is the deviation trend of the oil film change rate outside the critical range. A positive value indicates the tendency of the oil film to thicken, while a negative value reflects the tendency of the oil film to thin. The absolute value represents the degree of instability.

[0030] As a further aspect of the present invention, the specific steps of S3 include:

[0031] S301: Call the oil film stability correction amount, combine the three-dimensional coordinates of the nozzle base and the motion trajectory parameters of the gear shaft, extract the axial reference unit vector of the gear shaft, calculate the projection component of the injection angle reference direction in the axial direction, adjust the direction of the projection component according to the correction amount sign type, and generate the injection angle offset amount.

[0032] The injection angle offset is combined with the abnormal trend of oil film stability to dynamically adjust the injection direction;

[0033] S302: Based on the injection angle offset, a multi-factor matrix including nozzle pressure coefficient, oil viscosity parameter and ambient temperature weight is constructed. The scalar product operation of the matrix and the offset is performed. The calculation result is input into the hyperbolic tangent function for nonlinear mapping to generate the compensation angle correction coefficient.

[0034] The compensation angle correction coefficient, combined with multi-factor adjustment, generates a controllable angle control quantity;

[0035] S303: Based on the compensation angle correction coefficient, the nozzle angle is adjusted in fixed steps, pressure pulsation peak data is collected synchronously, the pressure fluctuation variance value of adjacent sampling cycles is calculated, and when the variance exceeds the set threshold, an anomaly mark is recorded. The frequency of anomaly marks within the sampling cycle is counted, and an injection feedback factor is generated.

[0036] As a further aspect of the present invention, the specific steps of S4 include:

[0037] S401: Based on the difference between the injection feedback factor and the oil film rate, the real-time change sequence of the injection feedback factor and the instantaneous fluctuation sequence of the oil film rate are obtained respectively. The difference amplitude between the two in multiple time periods is calculated to obtain the response difference amplitude between injection and oil film.

[0038] S402: Call the response difference amplitude and coverage change rate sequence between fuel injection and oil film, perform linear normalization on them in the same time period, and combine the weighted values ​​of the two normalization results to construct the adhesion factor numerical sequence.

[0039] S403: Based on the adhesion factor time series value, compare it with the upper limit of the adhesion factor confidence interval for the corresponding time period in the original working condition, identify the time point that exceeds the upper limit of the confidence interval, extract the adhesion factor value at the time point, and generate an adhesion level downgrade instruction according to the abnormal triggering situation.

[0040] Abnormal triggering conditions are three consecutive cycles exceeding the limit, triggering a level 1 downgrade, and five consecutive cycles exceeding the limit, triggering a level 2 downgrade.

[0041] As a further aspect of the present invention, the method further includes step S5:

[0042] S5: Based on the adhesion grade downgrade instruction, select oils in the oil pool that meet the interfacial tension and viscosity stability conditions, perform injection medium replacement through a dual-channel switching valve, and initiate a dynamic monitoring process of the oil film dielectric constant to evaluate the lubrication recovery performance under the new medium.

[0043] Lubrication recovery performance includes the rate of oil film rebuilding under new media, the continuous stability of lubrication effect, and the compatibility with the injection medium.

[0044] As a further aspect of the present invention, the specific steps of S5 include:

[0045] S501: Based on the adhesion grade downgrade instruction, analyze the viscosity compensation requirement range and interfacial tension correction range defined in the instruction, screen oil products that meet the requirements of viscosity compensation coefficient being in the middle range of the requirement range and interfacial tension correction value not being lower than the compensation lower limit, and generate a candidate oil product set.

[0046] S502: Call the candidate oil set, extract the high temperature shear resistance coefficient and oxidation stability index of the oil, establish a priority evaluation model based on adhesion recovery rate, calculate the score of the oil in the dimensions of high temperature stability and oxidation resistance, select the oil with high score as the switching target, and generate a medium switching scheme.

[0047] S503: Executes the medium switching scheme, controls the dual-channel switching valve to complete the oil circuit medium replacement according to the preset pressure gradient, synchronously activates the oil film inductance sensor array, collects the dielectric constant change rate and thickness growth rate during the oil film reconstruction process, calculates the oil film formation stabilization time and uniformity index, and evaluates the lubrication recovery performance under the new medium.

[0048] On the other hand, a limited-slip differential gear oil performance orientation optimization system is provided. This system is used to execute the aforementioned limited-slip differential gear oil performance orientation optimization method. The system includes:

[0049] The elongation analysis module is used to obtain the rate of temperature change through a temperature sensor, combine it with the linear expansion coefficient of the tooth axis to input the thermal strain calculation model to calculate the elongation, and calculate the axial elongation slope of the tooth axis based on the temperature output and axial reference length, and then transmit it to the clearance trend module.

[0050] The clearance trend module is used to receive the axial elongation slope of the gear shaft and the temperature rise rate at equidistant points on the inner wall of the housing. It identifies the temperature phase difference through the time-series offset model, determines the gradient of meshing clearance change, constructs the meshing clearance change trend vector, and transmits it to the oil film correction module.

[0051] The oil film correction module receives the meshing clearance change trend vector and the oil film thickness change slope, and inputs them into the stability identification model. By comparing with the critical rate boundary value, it extracts the oil film stability correction amount and transmits it to the spray angle compensation module.

[0052] The spray angle compensation module is used to project the oil film correction amount and the gear shaft trajectory onto the nonlinear correction function in the nozzle coordinate system with the center of the differential housing as the origin, collect pressure fluctuation data to calculate the feedback factor, construct the adhesion factor index, and generate an adhesion level downgrade command based on the comparison result with the original confidence interval, which is then transmitted to the media screening module.

[0053] The media replacement module is used to select suitable oil series by calling media interfacial tension and kinematic viscosity data according to the adhesion grade downgrade instruction, perform injection media replacement, and start the dynamic monitoring process of oil film dielectric constant to evaluate the lubrication recovery performance under the new media.

[0054] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0055] By real-time acquisition of gear shaft temperature rise rate and differential housing inner wall temperature gradient data, combined with linear expansion coefficient and displacement difference direction analysis, accurate prediction of dynamic changes in meshing clearance is achieved, effectively solving the problem of axial expansion estimation deviation caused by reliance on static temperature parameters in traditional methods. Integrating real-time oil film thickness data with multi-dimensional features of trend vectors, smoothing processing and critical rate threshold comparison are employed to improve the robustness of oil film stability judgment and avoid misjudgments caused by fluctuations in a single oil film thickness index. A dynamic compensation mechanism for injection angle based on vector projection and multi-factor nonlinear correction optimizes injection trajectory accuracy through closed-loop control of pressure fluctuation spectrum feedback and servo response delay, reducing the risk of uneven lubrication coverage caused by gear shaft motion trajectory deviation. When constructing the normalized adhesion factor index, phase difference parameters and confidence interval weights are introduced, combined with historical operating data to achieve early warning of adhesion performance degradation. Compared with traditional oil switching strategies based on fixed thresholds, this can identify lubrication performance degradation trends in advance and shorten the oil film reconstruction cycle. Through dynamic matching of surface tension critical value and viscosity decay threshold, precise triggering of lubrication medium switching is achieved, reducing frictional losses caused by oil performance mismatch. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the steps of the present invention;

[0057] Figure 2 This is a system flowchart of the present invention. Detailed Implementation

[0058] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0059] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0060] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0061] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0062] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0063] Please see Figure 1 This invention provides a method for targeted optimization of the performance of gear oil in a limited-slip differential. The processing flow of this method may include the following steps:

[0064] S1: The temperature rise rate is collected by the temperature sensor, and the axial elongation of the gear shaft is calculated by the linear expansion coefficient of the gear shaft. The temperature change rate of equidistant points on the inner wall of the differential housing is collected. The displacement difference direction is analyzed by calling the time offset model. The slope of the elongation change over time is extracted to generate the meshing clearance change trend vector.

[0065] S2: Call the meshing clearance change trend vector, fuse the real-time collected oil film thickness data at the gear meshing point, perform smoothing, remove outliers and calculate the derivative, identify the current oil film thickness change rate, and compare it with the critical interval to generate the oil film stability correction amount.

[0066] S3: Based on the oil film stability correction amount, the injection angle offset is calculated by the vector projection of the nozzle base coordinates and the tooth shaft motion trajectory on the tooth shaft axial reference. The offset is then input into the multi-factor nonlinear correction function to solve the compensation angle and drive the servo motor to perform fine-tuning. The injection pressure fluctuation is collected to obtain the injection feedback factor.

[0067] S4: Couple the injection feedback factor and the difference amplitude of the oil film rate, and construct the adhesion factor index by normalizing the coverage change rate. When the index exceeds the upper limit of the confidence interval of historical operating conditions, an adhesion level downgrade instruction is generated.

[0068] S5: Based on the adhesion grade downgrade instruction, select oils in the oil pool that meet the interfacial tension and viscosity stability conditions, perform injection medium replacement through a dual-channel switching valve, and initiate a dynamic monitoring process of the oil film dielectric constant to evaluate the lubrication recovery performance under the new medium.

[0069] The meshing clearance change trend vector specifically includes the displacement difference direction, temperature gradient, and time series slope. The oil film stability correction includes the critical rate threshold, lubricating film attenuation coefficient, and error tolerance range. The injection feedback factor covers the pressure fluctuation spectrum, injection angle deviation, and servo response delay. The adhesion factor index specifically refers to the normalized amplitude, phase difference parameter, and confidence interval weight. The adhesion level downgrade command includes the surface tension critical value, viscosity attenuation threshold, switching delay time, and oil film reconstruction time threshold.

[0070] Temperature gradient refers to the temperature change per unit length along the tooth shaft axis, which is calculated by collecting temperature field data from sensors.

[0071] The slope of the time series was obtained by linear regression analysis of the trend of elongation over time.

[0072] The lubricating film attenuation coefficient is calculated from the difference between the rate of change of oil film thickness and the critical interval.

[0073] The error tolerance range is determined based on historical data statistics;

[0074] The phase difference parameter reflects the temporal matching degree between the adhesion factor and the oil film coverage change, and is quantified through correlation analysis.

[0075] Specifically, the steps in S1 are as follows:

[0076] S101: The temperature rise rate is collected by a temperature sensor, the linear expansion coefficient of the tooth shaft is extracted, and the axial elongation of the tooth shaft is calculated by combining the relationship between the temperature rise rate and the expansion coefficient.

[0077] During gearbox operation, three temperature monitoring points were set up, located at the bearing seats of the input shaft, intermediate shaft, and output shaft, respectively. PT100 temperature sensors were used to collect temperature data at a frequency of 1Hz. Monitoring point 1 recorded temperature values ​​of [45.3, 45.8, 46.2, 46.7, 47.1]℃ over 5 consecutive seconds, and its temperature rise rate ΔT(t) was calculated as [0.5, 0.4, 0.5, 0.4]℃ / s. The linear expansion coefficient α of 38CrMoAlA material was used, which is 11.2 × 10⁻⁶. -6 ℃ -1 The temperature change rate correction factor β was set to 0.2. (Comparative experimental data showed that when ΔT > 0.3℃ / s, a correction factor of 0.15-0.25 can improve accuracy).

[0078] The axial elongation of the gear shaft is calculated using the following formula:

[0079]

[0080] Where ΔL represents the axial elongation of the gear shaft within the time interval [t0, t1], in meters, and α represents the linear expansion coefficient of the gear shaft material, in degrees Celsius. -1Its value is determined by the material properties. ΔT(t) represents the rate of temperature change collected by the temperature sensor at time t, in °C / s. β represents the temperature change rate correction factor (dimensionless). When ΔT(t) > 0.3 °C / s, the value range is 0.15-0.25, which is used to suppress the nonlinear expansion error caused by high-rate temperature rise. t0 represents the integration start time in seconds, t1 represents the integration end time in seconds, and dt represents the time differential variable in seconds. In discrete calculations, the sampling interval is 1 second.

[0081]

[0082] Taking Δt = 1s, performing discrete integration, the integral value for the time interval is calculated as [5.23 × 10]. -6 4.12×10 -6 5.11×10 -6 4.05×10 -6 The total elongation is 1.85 × 10⁻¹⁰ m. -5 m.

[0083] Table 1: Temperature Monitoring Point Parameter Table

[0084]

[0085]

[0086] As shown in Table 1, the three monitoring points form a temperature monitoring network. After measuring the temperature rise data of the input shaft, the dynamic expansion parameter calculation results are correlated with the material expansion coefficient to finally obtain the axial elongation of the gear shaft.

[0087] S102: The temperature change rate at equidistant points on the inner wall of the differential housing is synchronously acquired through a temperature sensor. The temperature change rate data is input into a time-series offset model to calculate the phase offset. The model is used to perform phase difference analysis on the temperature change rate at the differential measuring points of the housing, identify the mapping relationship between the temperature gradient and the housing deformation, and generate a displacement difference direction vector.

[0088] Five temperature measuring points (numbered T1-T5) were evenly distributed along the circumference of the differential housing. The temperature change rate at a certain moment was collected as [0.47, 0.52, 0.49, 0.51, 0.48]℃ / s. The attenuation coefficient λ = 0.3 (the optimal value range of 0.25-0.35 was determined through thermal conductivity experiments of the housing material), the reference time offset τ = [0, 0.2, 0.4, 0.6, 0.8]s, and the weighting coefficient η = 0.2 (the spatial weight was determined based on the housing thickness h = 6mm). The values ​​were 1.2, 1.5, 1.3, 1.4, and 1.25 MPa / m, respectively, with α set to 0.1. Based on empirical fitting and finite element simulation average calibration, the modulation factors were calculated to be 1.12, 1.15, 1.13, 1.14, and 1.125, respectively.

[0089] The timing offset model calculates the phase offset using the following formula:

[0090]

[0091] Where Φ(t) represents the phase offset at time t, with units of ℃·s⁻¹. This represents the rate of temperature change at the k-th temperature measuring point, expressed in °C / s, T. k The temperature function at the k-th measuring point is represented by the function. T represents k The amount of change over a small time interval It is a very small time increment, λ represents the temperature signal attenuation coefficient, and its value ranges from 0.25 to 0.35, τ k η represents the offset of the k-th measuring point relative to the reference time, and η represents the weighting coefficient of the spatial variation of the temperature field, with a value ranging from 0.1 to 0.3. The Laplace operator represents the temperature field of the housing, p represents the total number of equidistant measuring points on the inner wall of the differential housing, e represents the natural logarithm base, with a value of 2.71828, t represents the current time variable, k represents the measuring point index variable, with a value ranging from 1 to p, and α represents the axial stress gradient modulation coefficient. This represents the axial stress gradient along the tooth axis at the k-th temperature measuring point, in MPa / m, σ k This represents the internal normal stress distribution function of the gear shaft material in the axial direction caused by temperature at the k-th measuring point, with units of MPa.

[0092] When t = 2s, calculate the components at the measuring point:

[0093] [0.47·e -0.3·|2-0 |=0.47·0.5488=0.2579,0.52·e -0.3·|2-0.2 |=0.52·

[0094] 0.2805 = 0.1458, ...];

[0095] After correction, we obtain [0.2888, 0.3475, 0.3427, 0.3810, 0.3768], which sums to get 0.2888 + 0.3475 + 0.3427 + 0.3810 + 0.3768 = 1.7368. Then, we calculate the Laplace operator for the temperature field.

[0096] Finally, Φ(2) = 1.7368 + 0.2 × (-0.13) = 1.7108. The displacement direction is determined as the orientation of the measuring point T2 based on the maximum component of the phase offset.

[0097] S103: Based on the time series data of the axial elongation of the gear shaft, calculate the slope of the elongation change per unit time, and combine the direction parameter of the displacement difference direction vector to perform vector synthesis operation on the slope value and the direction parameter to generate the meshing clearance change trend vector.

[0098] Collect axial elongation data of the gear shaft for 10 consecutive sampling periods:

[0099] 1.85, 2.12, 2.37, 2.64, 2.89, 3.15, 3.42, 3.68, 3.95, 4.21]×10 -5 m, calculate the slope of change displacement difference direction vector magnitude (Assuming two-dimensional plane coordinates), set the trend composition ratio coefficient γ = 0.1 (the optimal value is determined through historical data regression analysis). Calculate the vector composition term:

[0100] The projection of this vector onto the xy plane indicates that the meshing clearance increases by approximately 16.45 mm per hour, with an offset direction angle of 10.7°.

[0101] Specifically, the steps of S2 are as follows:

[0102] S201: Call the meshing clearance change trend vector to obtain the original data sequence of oil film thickness at the gear meshing point, use the moving average method to smooth the oil film thickness data, and remove outliers that deviate from the mean based on the statistical distribution criteria to obtain stable oil film data.

[0103] The statistical distribution criteria selected include the 3σ rule, the box plot interquartile range method, or a distribution fitting and elimination strategy based on skewness correction.

[0104] Call the meshing clearance change trend vector to obtain the original data sequence of oil film thickness at the gear meshing point: 12.3, 12.8, 13.1, 11.9, 12.5, 14.2, 12.7, 12.4, 13.0, 12.6 μm;

[0105] Smoothing is performed using a moving average method with a window width of 3, and the smoothed value for the second data point is calculated. Obtain the smoothed sequence:

[0106] [12.73, 12.93, 12.50, 12.87, 13.13, 12.77, 12.70, 12.67], calculate the mean μ = 12.75 μm and the standard deviation σ = 0.68 μm. Based on engineering experience, k = 2.5 is used to generate the outlier threshold [11.39, 14.11] μm. The 6th original data point 14.2 is removed (because 14.2 > 14.11), generating stable oil film data:

[0107] 12.3, 12.8, 13.1, 11.9, 12.5, 12.7, 12.4, 13.0, 12.6]μm;

[0108] Table 2: Key Data Parameters Table

[0109] Maximum value of original data 14.2 Smoothing data mean 12.75 Outlier threshold upper limit 14.11

[0110] As shown in Table 2, data validity is verified by quantizing the threshold parameter. When outliers appear in three consecutive sampling periods, the k value is automatically reduced (e.g., k is adjusted from 2.5 to 2.3).

[0111] S202: Based on stable oil film data, extract the difference in oil film thickness between adjacent sampling points, combine it with a fixed time interval, and calculate the change per unit time using a splitting algorithm to obtain the dynamic change rate of the oil film;

[0112] Based on stable oil film data, the thickness difference between adjacent sampling points is extracted. The difference between points 1 and 2 is 12.8 - 12.3 = 0.5 μm, and the difference between points 2 and 3 is 13.1 - 12.8 = 0.3 μm. Combined with a fixed sampling interval Δt = 0.1 s, the instantaneous change rate is calculated as 0.5 / 0.1 = 5.0 μm / s and 0.3 / 0.1 = 3.0 μm / s, constructing a complete change rate sequence [5.0, 3.0, -12.0, 6.0, 2.0, -3.0, 6.0, -4.0] μm / s. When the difference between adjacent points exceeds 10 μm, a data verification mechanism is triggered to generate the dynamic change rate of the oil film.

[0113] S203: Call the dynamic change rate of the oil film, compare the dynamic change rate of the oil film with the boundary value of the interval point by point according to the preset critical interval parameter, assign state markers according to the comparison results, count the distribution ratio of different markers in the continuous sampling period, calculate the difference in the proportion of positive markers and negative markers, and obtain the oil film stability correction amount.

[0114] The oil film stability correction is the deviation trend of the oil film change rate outside the critical range. A positive value indicates the tendency of oil film thickening, while a negative value reflects the tendency of oil film thinning. The absolute value represents the degree of instability.

[0115] The oil film dynamic change rate sequence is invoked, and the critical interval is determined based on the distribution of oil film thickness change rate under stable operating conditions in historical operating data, using a 95% confidence interval to determine the upper and lower limits:

[0116] L = μ - 2σ, U = μ + 2σ;

[0117] Set the critical interval [L, U] = [-11.412, 12.162] μm / s, and generate state markers by comparing points: the rate of change at point 3 is -12.0 < -11.412, marked as -1; the rate of change at point 4 is 6.0 < 12.162, marked as 0, resulting in the marker sequence [0, 0, -1, 0, 0, 0, 0, 0]. Calculate the marker distribution within the statistical window period T = 4 periods, and calculate the correction amount for periods 1-4: the number of positive markers N. + =0, negative label number N - =-1, total number of tags N total =4, calculate correction amount C = 0 + 1 / 4 = 0.25, calculate correction amount for cycles 5-8 as 0, trigger warning when C value exceeds 0.5 for two consecutive cycles, and generate oil film stability correction amount.

[0118] Specifically, the steps of S3 are as follows:

[0119] S301: Call the oil film stability correction amount, combine the three-dimensional coordinates of the nozzle base and the motion trajectory parameters of the gear shaft, extract the axial reference unit vector of the gear shaft, calculate the projection component of the injection angle reference direction in the axial direction, adjust the direction of the projection component according to the correction amount sign type, and generate the injection angle offset amount.

[0120] The injection angle offset is combined with the abnormal trend of oil film stability to dynamically adjust the injection direction;

[0121] After the oil film stability correction is input, the system first extracts a reference unit vector for the gear shaft axis in three-dimensional space based on the collected three-dimensional coordinate information of the nozzle base and the gear shaft motion trajectory parameters. The coordinate value of this unit vector in the axial direction is used as the axial projection baseline. Then, the injection offset trend is determined by comparing the input correction value with the sign value. If the correction is positive, the original axial direction remains unchanged; if the correction is negative, the axial vector direction is adjusted in the opposite direction, and the spatial coordinates of the nozzle base are corrected to obtain the offset injection angle direction vector. This vector is then mapped onto the axial direction to determine its axial projection component. This projected component is the axial offset of the injection angle. For example, in a set of gear shaft coordinates (5.6, 2.2, 3.0) mm, the gear shaft motion vector is (0.5, 0.5, 0.707), which is normalized to a direction vector of (0.5, 0.5, 0.707). If the oil film stability correction is -0.25, then the direction is reversed to (-0.5, -0.5, -0.707), and the projected component is -0.707, indicating that the injection angle needs to be offset and corrected in the opposite direction. This value will be used as an adjustment amount in the subsequent steps to participate in the weight calculation of the multi-factor matrix to generate the injection angle correction angle.

[0122] S302: Based on the injection angle offset, a multi-factor matrix including nozzle pressure coefficient, oil viscosity parameter and ambient temperature weight is constructed. The scalar product operation of the matrix and the offset is performed. The calculation result is input into the hyperbolic tangent function for nonlinear mapping to generate the compensation angle correction coefficient.

[0123] The compensation angle correction coefficient, combined with multi-factor adjustment, generates a controllable angle control quantity;

[0124] After obtaining the above-mentioned injection angle offset, it is passed as a variable into a multi-factor matrix composed of three types of parameters. The multiple parameters of this matrix are the nozzle pressure coefficient, the oil viscosity parameter, and the ambient temperature weight. In practical implementation, the following typical settings can be used: the nozzle pressure coefficient is set to 2.1 MPa, corresponding to a k1 weight coefficient of 0.5; the oil viscosity is 0.35 Pa·s, with a k2 weight of 0.8; and the ambient temperature is 35℃, with a k3 weight of 0.2. These three types of parameters form corresponding term vectors with the injection angle offset. When performing the matrix dot product, if the offset is... 3.2°, then the three terms are 2.1×0.5×3.2, 0.35×0.8×3.2, and 35×0.2×3.2, with corresponding values ​​of 3.36, 0.896, and 22.4. Summing the three terms yields a total scalar product of 26.656, which is then passed to the hyperbolic tangent function tanh(26.656) for mapping. Since this value is much greater than the saturation value of the tanh function, the final mapping result is approximately 1.000, indicating that the maximum compensation degree of the injection angle has been reached. The mapping coefficient value is the final compensation angle correction coefficient used for subsequent angle adjustments.

[0125] S303: Based on the compensation angle correction coefficient, the nozzle angle is adjusted in fixed step size, pressure pulsation peak data is collected synchronously, the pressure fluctuation variance value of adjacent sampling cycles is calculated, and when the variance exceeds the set threshold, an abnormal mark is recorded. The frequency of abnormal marks in the sampling cycle is counted, and the fuel injection feedback factor is generated.

[0126] After obtaining the compensation angle correction coefficient, the nozzle angle is dynamically adjusted according to a set step size using it as an input factor. During the adjustment process, the peak value of the pressure pulsation in the current cycle is recorded simultaneously. For example, if the nozzle angle increases from 35° to 36.2° in one sampling, the corresponding pressure peak sequence is [2.5, 2.6, 2.8, 3.0, 3.3] MPa. Then, the pressure fluctuation variance between adjacent cycles can be calculated to be 0.0925 MPa. 2 If the current threshold is set to 0.08 MPa 2 If the frequency of anomalies is not recorded, it is considered an abnormal fluctuation. An anomaly is recorded once in that period. The number of anomaly markers is accumulated within the sampling period. If the number of anomalies exceeds 3 in 10 consecutive sampling periods, the control module generates an injection feedback factor to adjust the injection strategy. The generation of this feedback factor is based on the frequency of anomalies and the rate of change of pressure variance, and a secondary evaluation is performed in combination with the compensation angle correction coefficient to ensure that the feedback results are representative within the sampling period.

[0127] Table 3: Input Parameter Table for Injection Angle Correction Coefficient

[0128]

[0129] As shown in Table 3, the total scalar product of the three sets of parameters set based on the current operating conditions is 26.656 under the premise of an offset angle of 3.2°. The result after hyperbolic tangent mapping is 1.000, indicating that the current nozzle angle needs to be fully compensated and adjusted.

[0130] Specifically, the steps of S4 are as follows:

[0131] S401: Based on the difference between the injection feedback factor and the oil film rate, the real-time change sequence of the injection feedback factor and the instantaneous fluctuation sequence of the oil film rate are obtained respectively. The difference amplitude between the two in multiple time periods is calculated to obtain the response difference amplitude between injection and oil film.

[0132] Based on the injection feedback factor and the difference in oil film rate, the injection pulse width correction coefficient in the engine ECU is acquired at a time resolution of 0.1 seconds as the injection feedback factor. Simultaneously, the oil film evaporation rate deviation value monitored by the high-pressure fuel rail pressure sensor is acquired as the oil film rate difference. At t = 1.2 seconds, the recorded value of the injection feedback factor is 1.15 (the standard operating condition baseline value is 1.0), corresponding to an oil film rate difference of -0.02 L / min. The difference amplitude at this moment is calculated as Δ = |1.15 - (-0.02)| = 1.17. When t = 1.3 seconds, the injection feedback factor suddenly increases to 1.28, and the oil film rate difference reaches 0.05 L / min. Δ = |1.28 - 0.05| = 1.2 3. Collect data at 10 consecutive time points to form a sequence. Take the data within the time window [1.2s, 2.1s] to construct response difference features. The maximum difference amplitude occurs at t = 1.7 seconds, Δ = 1.45, and the minimum is at t = 2.0 seconds, Δ = 0.89. Calculate the average difference amplitude for this period as (1.17 + 1.23 + ... + 0.89) / 10 = 1.12. Set the difference amplitude threshold Δ_th = 1.30. When Δ > Δ_th for 3 consecutive sampling points, trigger the feature anomaly label. As shown in Table 4, the Δ values ​​from t = 1.5s to 1.7s are 1.31, 1.38, and 1.45, respectively, reaching the threshold trigger condition, and generating the response difference amplitude feature sequence.

[0133] Table 4: Monitoring Data of Difference Amplitude

[0134]

[0135]

[0136] As shown in Table 4, the difference amplitude is calculated using absolute value operation. When the difference between the injection feedback factor and the oil film rate changes in opposite directions, a large Δ value will be generated. For example, when t = 1.7 seconds, the injection quantity increases but the oil film formation rate decreases. At this time, the Δ value reaches a peak of 1.45. The final output of the response difference amplitude characteristic sequence is [1.17, 1.23, 1.10, 1.31, 1.38, 1.45, 1.28, 1.17, 0.89, 0.87].

[0137] S402: Call the response difference amplitude and coverage change rate sequence between fuel injection and oil film, perform linear normalization on them in the same time period, and combine the weighted values ​​of the two normalization results to construct the adhesion factor numerical sequence.

[0138] The response difference amplitude sequence [1.17, 1.23, ..., 0.87] was linearly normalized, taking the maximum value MAX = 1.45 and the minimum value MIN = 0.87 for this period. The normalization formula is N. i =(Δ i-MIN) / (MAX-MIN), for example, at t=1.2s, N=(1.17-0.87) / (1.45-0.87)=0.517, synchronously process the coverage change rate sequence [0.15, 0.18, ..., 0.22], where MAX=0.25, MIN=0.12, and at t=1.2s, the coverage change rate 0.15 is normalized to (0.15-0.12) / (0.25-0.12)=0.231. Set the fuel injection response weight α=0.6 and the coverage change weight β=0.4, and obtain the adhesion factor by weighted summation. Taking t = 1.2s as an example: M = 0.6 × 0.517 + 0.4 × 0.231 = 0.394. The adhesion factor sequence [0.394, 0.423, 0.381, 0.502, 0.567, 0.642, 0.521, 0.394, 0.301, 0.287] is obtained by continuously calculating 10 time points. The weight coefficient is set according to the fact that the response delay time of the fuel injection system τ = 80ms is less than the oil film formation delay time τ′ = 120ms, so it is given a higher weight. The weight allocation is determined by α / (α+β) = τ′ / (τ+τ′) = 120 / (80+120) = 0.6.

[0139] S403: Based on the adhesion factor time series value, compare it with the upper limit of the adhesion factor confidence interval for the corresponding time period in the original working condition, identify the time point that exceeds the upper limit of the confidence interval, extract the adhesion factor value at the time point, and generate an adhesion level downgrade instruction according to the abnormal triggering situation.

[0140] Abnormal triggering conditions are three consecutive cycles exceeding the limit, triggering a level 1 downgrade, and five consecutive cycles exceeding the limit, triggering a level 2 downgrade.

[0141] Retrieving historical data from the past 30 days under the same operating conditions, the mean adhesion factor for the same time period each day was calculated as μ = 0.412, and the standard deviation σ = 0.098. The upper limit of the confidence interval was set as μ + 3σ = 0.412 + 3 × 0.098 = 0.706. A point-by-point comparison of the current adhesion factor sequence [0.394, 0.423, ..., 0.287] revealed that M = 0.567 < 0.706 at t = 1.6s, M = 0.642 < 0.706 at t = 1.7s, and M = 0.5 at t = 1.8s. 21 < 0.706, no degradation condition was triggered. When the experimental ambient temperature increased by 15℃, the test was repeated. At t = 2.3s, M = 0.735 > 0.706 was measured. The M values ​​of the three consecutive sampling points were 0.735, 0.718 and 0.741, which exceeded the upper limit of the confidence interval. The abnormal time points [2.3s, 2.4s, 2.5s] and their M values ​​[0.735, 0.718, 0.741] were extracted to form the degradation trigger set, and a control command was generated to reduce the adhesion level from level II to level III.

[0142] Specifically, the steps of S5 are as follows:

[0143] S501: Based on the adhesion grade downgrade instruction, analyze the viscosity compensation requirement range and interfacial tension correction range defined in the instruction, screen oil products that meet the requirements of viscosity compensation coefficient being in the middle range of the requirement range and interfacial tension correction value not being lower than the compensation lower limit, and generate a candidate oil product set.

[0144] Based on the adhesion grade downgrade instruction, the viscosity compensation requirement range in the instruction is [0.85, 1.15] (baseline value is 1.0), the interfacial tension correction range is set to [28.5, 32.0] mN / m, and the filtering condition is that the viscosity compensation coefficient is in the median range of [0.95, 1.05] and the interfacial tension correction value is not lower than 29.0 mN / m. The oil product pool database is queried to obtain 12 initial oil product entries, and the parameters are compared item by item. For example, the viscosity compensation of oil product A... The viscosity compensation coefficient of 1.02 and the interfacial tension of 30.2 mN / m meet the requirements. Oil B with a viscosity compensation coefficient of 0.91 (below the lower limit of the median range of 0.95) is excluded. Finally, five oils that meet the dual constraints are selected, with viscosity compensation coefficients of [0.97, 1.03, 1.01, 0.98, 1.04] and corresponding interfacial tension values ​​of [29.5, 30.8, 31.2, 29.7, 30.5] mN / m, generating a candidate oil set.

[0145] Table 5: Candidate Oil Product Parameter Table

[0146] C-1023 0.97 29.5 D-2045 1.03 30.8 E-3017 1.01 31.2 F-4098 0.98 29.7 G-5071 1.04 30.5

[0147] As shown in Table 5, oils with a viscosity compensation coefficient median range that deviate from the benchmark value by more than ±5% are excluded, while oils with an interfacial tension correction value that is more than 1.7% higher than the lower limit of compensation are retained. The final candidate oil set includes 5 oil items that meet the adhesion compensation requirements.

[0148] S502: Call the candidate oil set, extract the high temperature shear resistance coefficient and oxidation stability index of the oil, establish a priority evaluation model based on adhesion recovery rate, calculate the score of the oil in the dimensions of high temperature stability and oxidation resistance, select the oil with the highest score as the switching target, and generate a medium switching scheme.

[0149] The candidate oil set was used to extract the high-temperature shear strength coefficient (η≥85%) and oxidation stability index (δ≤0.15mg / cm). 2 The adhesion recovery rate assessment model parameters were set with weights of α = 0.6 and β = 0.4, and the comprehensive score S = αη + β(1 - δ / 0.15) was calculated.

[0150] Where η represents the high-temperature shear resistance coefficient, δ represents the oxidation stability index, and α and β are weighting coefficients.

[0151] Taking oil product C-1023 as an example: η=88, δ=0.12, then S=0.6×88+0.4×(1-0.12 / 0.15)=52.8+0.4×0.2=53.6. Similarly, the scores of the other oil products are calculated as [53.6, 55.1, 56.3, 54.0, 55.2]. The oil product E-3017 corresponding to the highest score of 56.3 is selected to generate the medium switching scheme.

[0152] S503: Executes the medium switching scheme, controls the dual-channel switching valve to complete the oil circuit medium replacement according to the preset pressure gradient, synchronously activates the oil film inductance sensor array, collects the dielectric constant change rate and thickness growth rate during the oil film reconstruction process, calculates the oil film formation stabilization time and uniformity index, and evaluates the lubrication recovery performance under the new medium.

[0153] The media switching scheme was implemented, and the dual-channel switching valve was controlled to replace the oil circuit media at a pressure gradient of 0.5 MPa / s. The switching was completed when the main oil circuit pressure dropped from 3.2 MPa to 2.0 MPa. The inductive sensor array was activated to collect oil film reconstruction data. Within 500 ms after the switching, the dielectric constant change rate was recorded to increase from 1.25 to 2.80, and the thickness growth rate was recorded to increase from 0.05 mm / ms to 0.12 mm / ms. The stabilization time was calculated as 320 ms, which is the time required for the standard deviation of the dielectric constant change rate to be ≤0.05. The uniformity index was the coefficient of variation of the thickness growth rate of 0.08. The lubrication recovery performance under the new media was evaluated.

[0154] like Figure 2 As shown, the limited-slip differential gear oil performance optimization system includes:

[0155] The elongation analysis module is used to obtain the rate of temperature change through a temperature sensor, combine it with the linear expansion coefficient of the tooth axis to input the thermal strain calculation model to calculate the elongation, and calculate the axial elongation slope of the tooth axis based on the temperature output and axial reference length, and then transmit it to the clearance trend module.

[0156] The clearance trend module is used to receive the axial elongation slope of the gear shaft and the temperature rise rate at equidistant points on the inner wall of the housing. It identifies the temperature phase difference through the time-series offset model, determines the gradient of meshing clearance change, constructs the meshing clearance change trend vector, and transmits it to the oil film correction module.

[0157] The oil film correction module receives the meshing clearance change trend vector and the oil film thickness change slope, and inputs them into the stability identification model. By comparing with the critical rate boundary value, it extracts the oil film stability correction amount and transmits it to the spray angle compensation module.

[0158] The spray angle compensation module is used to project the oil film correction amount and the gear shaft trajectory onto the nonlinear correction function in the nozzle coordinate system with the center of the differential housing as the origin, collect pressure fluctuation data to calculate the feedback factor, construct the adhesion factor index, and generate an adhesion level downgrade command based on the comparison result with the original confidence interval, which is then transmitted to the media screening module.

[0159] The media replacement module is used to select suitable oil series by calling media interfacial tension and kinematic viscosity data according to the adhesion grade downgrade instruction, perform injection media replacement, and start the dynamic monitoring process of oil film dielectric constant to evaluate the lubrication recovery performance under the new media.

[0160] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for targeted optimization of gear oil performance in limited-slip differentials, characterized in that, Includes the following steps: S1: The temperature rise rate is collected by the temperature sensor, and the axial elongation of the gear shaft is calculated by the linear expansion coefficient of the gear shaft. The temperature change rate of equidistant points on the inner wall of the differential housing is collected. The displacement difference direction is analyzed by calling the time offset model. The slope of the elongation change over time is extracted to generate the meshing clearance change trend vector. The specific steps of S1 include: S101: The temperature rise rate is collected by a temperature sensor, the linear expansion coefficient of the tooth shaft is extracted, and the axial elongation of the tooth shaft is calculated by combining the relationship between the temperature rise rate and the expansion coefficient. S102: The temperature change rate at equidistant points on the inner wall of the differential housing is synchronously acquired through a temperature sensor. The temperature change rate data is input into a time-series offset model to calculate the phase offset. The model is used to perform phase difference analysis on the temperature change rate at the differential measuring points of the housing, identify the mapping relationship between the temperature gradient and the housing deformation, and generate a displacement difference direction vector. S103: Based on the time series data of the axial elongation of the gear shaft, calculate the slope of the elongation change per unit time, and combine the direction parameter of the displacement difference direction vector to perform vector synthesis operation on the slope value and the direction parameter to generate the meshing clearance change trend vector. S2: Call the meshing clearance change trend vector, fuse the real-time collected oil film thickness data at the gear meshing point, perform smoothing processing, remove outliers and calculate the derivative, identify the current oil film thickness change rate, and compare it with the critical interval to generate an oil film stability correction amount. S3: Based on the oil film stability correction amount, the injection angle offset is calculated by the vector projection of the nozzle base coordinates and the tooth shaft motion trajectory on the tooth shaft axial reference. The offset is then input into the multi-factor nonlinear correction function to solve the compensation angle and drive the servo motor to perform fine-tuning. The injection pressure fluctuation is collected to obtain the injection feedback factor. The specific steps of S3 include: S301: Call the oil film stability correction amount, combine the three-dimensional coordinates of the nozzle base and the motion trajectory parameters of the gear shaft, extract the axial reference unit vector of the gear shaft, calculate the projection component of the injection angle reference direction in the axial direction, adjust the direction of the projection component according to the correction amount sign type, and generate the injection angle offset amount; The injection angle offset is dynamically adjusted in conjunction with the abnormal trend of oil film stability to adjust the injection direction. S302: Based on the injection angle offset, construct a multi-factor matrix including nozzle pressure coefficient, oil viscosity parameter, and ambient temperature weight, perform scalar product operation between the matrix and the offset, input the operation result into the hyperbolic tangent function for nonlinear mapping, and generate compensation angle correction coefficient. The compensation angle correction coefficient, combined with multi-factor adjustment, generates a controllable angle control quantity; S303: Based on the compensation angle correction coefficient, adjust the nozzle angle in fixed steps, synchronously collect pressure pulsation peak data, calculate the pressure fluctuation variance value of adjacent sampling periods, record an anomaly marker when the variance exceeds a set threshold, count the frequency of anomaly marker occurrences within the sampling period, and generate an injection feedback factor. S4: Couple the injection feedback factor with the oil film rate difference amplitude, and construct the adhesion factor index by normalizing the coverage change rate. When the index exceeds the upper limit of the original working condition confidence interval, an adhesion level downgrade instruction is generated.

2. The method for targeted optimization of gear oil performance in a limited-slip differential according to claim 1, characterized in that, The time-series offset model is based on the thermal strain-oil film attenuation coupling model to dynamically analyze the axial displacement trend caused by gear shaft temperature rise and the oil film thickness change, and to construct a nonlinear compensation function between the injection direction and the oil film stability. When calculating the parameters of the oil film thickness change rate and temperature change rate, they are first processed to be dimensionless by normalization transformation. The meshing clearance change trend vector specifically includes the displacement difference direction, temperature gradient, and time series slope. The oil film stability correction includes the critical rate threshold, lubricating film attenuation coefficient, and error tolerance range. The injection feedback factor includes the pressure fluctuation spectrum, injection angle deviation, and servo response delay. The adhesion factor index specifically refers to the normalized amplitude, phase difference parameter, and confidence interval weight. The adhesion level degradation command includes the surface tension critical value, viscosity attenuation threshold, switching delay time, and oil film reconstruction time threshold. The temperature gradient refers to the temperature change per unit length along the tooth shaft axis, which is calculated by collecting temperature field data from a temperature sensor. The lubricating film attenuation coefficient is calculated from the difference between the oil film thickness change rate and the critical interval.

3. The method for targeted optimization of gear oil performance in a limited-slip differential according to claim 1, characterized in that, The phase offset is calculated using the following formula: ； in, represent Phase shift at time, in °C·s -1 , Representing the The rate of temperature change at each temperature measuring point, in °C / s. Indicates the first Temperature function at each measuring point express The amount of change over a small time interval It is a very small time increment. Represents the temperature signal attenuation coefficient. Representing the The offset of the measuring point from the reference time. Represents the weighting coefficient of spatial variation in the temperature field. The Laplace operator representing the shell temperature field, This represents the total number of equidistant measuring points on the inner wall of the differential housing. Represents the cardinality of the natural logarithm. Represents the time variable at the current moment. Represents the variable of the measurement point number. Represents the axial stress gradient modulation coefficient. Representing the The axial stress gradient along the tooth axis at each temperature measuring point, in MPa / m. Representing the The internal normal stress distribution function of the gear shaft material in the axial direction caused by temperature at each measuring point is expressed in MPa.

4. The method for targeted optimization of gear oil performance in a limited-slip differential according to claim 1, characterized in that, The specific steps of S2 include: S201: Call the meshing clearance change trend vector to obtain the original data sequence of oil film thickness at the gear meshing point, use the moving average method to smooth the oil film thickness data, and remove outliers that deviate from the mean based on the statistical distribution criteria to obtain stable oil film data. The statistical distribution criteria selected are the 3σ rule, the box plot quartile method, or a distribution fitting and elimination strategy based on skewness correction. S202: Based on the stable oil film data, extract the oil film thickness difference between adjacent sampling points, and combine it with a fixed time interval to calculate the change per unit time using a splitting algorithm to obtain the dynamic change rate of the oil film; S203: Call the oil film dynamic change rate, compare the oil film dynamic change rate with the interval boundary value point by point according to the critical interval parameter, assign state markers according to the comparison results, count the distribution ratio of different markers in the continuous sampling period, calculate the ratio difference between positive and negative markers, and obtain the oil film stability correction amount. The critical interval is determined based on the statistical analysis of the oil film thickness change rate under the original stable operating conditions, and the threshold range is determined using a 95% confidence interval. The oil film stability correction is the deviation trend of the oil film change rate outside the critical range. A positive value indicates the tendency of the oil film to thicken, while a negative value reflects the tendency of the oil film to thin. The absolute value represents the degree of instability.

5. The method for targeted optimization of gear oil performance in a limited-slip differential according to claim 1, characterized in that, The specific steps of S4 include: S401: Based on the difference between the injection feedback factor and the oil film rate, obtain the real-time change sequence of the injection feedback factor and the instantaneous fluctuation sequence of the oil film rate, calculate the amplitude of the difference between the two in multiple time periods, and obtain the amplitude of the response difference between injection and oil film. S402: Call the response difference amplitude and coverage change rate sequence between the oil injection and the oil film, perform linear normalization on them in the same time period, and combine the weighted values ​​of the two normalization results to construct the adhesion factor numerical sequence. S403: Based on the adhesion factor time series value, compare it with the upper limit of the adhesion factor confidence interval for the corresponding time period in the original working condition, identify the time point that exceeds the upper limit of the confidence interval, extract the adhesion factor value at the time point, and generate an adhesion level downgrade instruction according to the abnormal triggering situation. The abnormal triggering conditions are: exceeding the limit for three consecutive cycles triggering a first-level downgrade, and exceeding the limit for five consecutive cycles triggering a second-level downgrade.

6. The method for targeted optimization of gear oil performance in a limited-slip differential according to claim 1, characterized in that, The method also includes step S5: S5: According to the adhesion level downgrade instruction, select oils in the oil pool that meet the interfacial tension and viscosity stability conditions, perform injection medium replacement through a dual-channel switching valve, and start the dynamic monitoring process of oil film dielectric constant to evaluate the lubrication recovery performance under the new medium. The lubrication recovery performance includes the oil film rebuilding rate under new medium, the continuous stability of lubrication effect, and the compatibility with the injection medium.

7. The method for targeted optimization of gear oil performance in a limited-slip differential according to claim 6, characterized in that, The specific steps of S5 include: S501: Based on the adhesion level downgrade instruction, analyze the viscosity compensation requirement range and interfacial tension correction range defined in the instruction, screen oil products that meet the requirements of viscosity compensation coefficient being in the middle range of the requirement range and interfacial tension correction value not being lower than the compensation lower limit, and generate a candidate oil product set. S502: Call the candidate oil set, extract the high temperature shear resistance coefficient and oxidation stability index of the oil, establish a priority evaluation model based on adhesion recovery rate, calculate the score of the oil in the dimensions of high temperature stability and oxidation resistance, select the oil with high score as the switching target, and generate a medium switching scheme. S503: Execute the medium switching scheme, control the dual-channel switching valve to complete the oil circuit medium replacement according to the preset pressure gradient, simultaneously activate the oil film inductance sensor array, collect the dielectric constant change rate and thickness growth rate during the oil film reconstruction process, calculate the oil film formation stabilization time and uniformity index, and evaluate the lubrication recovery performance under the new medium.

8. A gear oil performance optimization system for limited-slip differentials, characterized in that, The system is used to implement the gear oil performance optimization method for limited-slip differentials according to any one of claims 1-7, the system comprising: The elongation analysis module is used to obtain the rate of temperature change through a temperature sensor, combine it with the linear expansion coefficient of the tooth axis to input the thermal strain calculation model to calculate the elongation, and calculate the axial elongation slope of the tooth axis based on the temperature output and axial reference length, and then transmit it to the clearance trend module. The clearance trend module is used to receive the axial elongation slope of the gear shaft and the temperature rise rate at equidistant points on the inner wall of the housing, identify the temperature phase difference through the time-series offset model, determine the meshing clearance change gradient, construct the meshing clearance change trend vector, and transmit it to the oil film correction module. The oil film correction module is used to receive the meshing clearance change trend vector and the oil film thickness change slope, and jointly input the stability identification model. By comparing with the critical rate boundary value, the oil film stability correction amount is extracted and transmitted to the spray angle compensation module. The spray angle compensation module is used to project the oil film stability correction amount and the gear shaft trajectory onto the nonlinear correction function in the nozzle coordinate system with the center of the differential housing as the origin, collect pressure fluctuation data to calculate the feedback factor, construct the adhesion factor index, and generate an adhesion level downgrade command based on the comparison result with the original confidence interval, and transmit it to the media screening module. The media replacement module is used to select suitable oil series by calling media interfacial tension and kinematic viscosity data according to the adhesion level downgrade instruction, perform injection media replacement, and initiate a dynamic monitoring process of oil film dielectric constant to evaluate lubrication recovery performance under the new media.

Images