Bevel gear pair self-adaptive adjustment method and system combined with installation distance deviation

Abstract

The invention discloses a bevel gear pair self-adaptive adjustment method and system combined with installation distance deviation, and relates to the related field of intelligent assembling.The method comprises the steps that real-time installation characteristic data are obtained, installation distance deviation adjustment analysis is conducted according to an installation design scheme, and an initial installation adjustment scheme is obtained; performing multi-dimensional implicit installation error trend prediction according to the basic characteristic data and the real-time installation characteristic data, and constructing an implicit installation error trend graph; performing hidden installation error suppression compensation on the initial scheme according to the map to obtain an installation adjustment space; constructing a real-time installation twin body and bevel gear pair installation evaluation model; and carrying out installation evaluation iterative optimization on the installation adjustment space to obtain an installation adjustment optimization strategy. The technical problems that existing installation distance adjustment cannot adapt to dynamic error evolution, and the adjustment precision and stability are difficult to guarantee are solved, and the technical effects of dynamically adapting to the error evolution, improving the installation adjustment precision and guaranteeing the long-term operation stability are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent assembly, and in particular to a bevel gear pair adaptive adjustment method and system combined with installation distance deviation. BACKGROUND

[0002] The installation distance precision of the bevel gear pair directly determines the transmission stability, noise level and service life, and is the core control point of the assembly link of high-end transmission equipment. At present, the industry mainly guides manual assembly through theoretical design values, realizes installation distance correction combined with offline detection and trial-and-error gasket adjustment, and completes meshing state matching relying on experience and single measurement. Such adjustment method relying on static design parameters and manual experience does not consider the dynamic error evolution caused by load, temperature and structural deformation, and is prone to problems such as mismatch between adjustment amount and actual working condition, installation precision attenuation with running, and difficulty in long-term stability of meshing state.

[0003] In the related art, the bevel gear pair installation distance adjustment has the technical problems of being unable to adapt to dynamic error evolution and being difficult to guarantee adjustment precision and stability. SUMMARY

[0004] The present application provides a bevel gear pair adaptive adjustment method and system combined with installation distance deviation, which obtains real-time installation characteristic data of the bevel gear pair, analyzes installation distance deviation adjustment combined with installation design scheme, obtains an installation adjustment initial scheme, predicts the trend of multi-dimensional implicit installation error based on the basic characteristics and real-time installation characteristics of the bevel gear pair, constructs an implicit installation error trend map, compensates for error suppression based on the initial adjustment scheme according to the trend map, forms a feasible installation adjustment space, constructs a real-time installation twin body synchronized with the physical installation state, trains a bevel gear pair installation evaluation model, uses the real-time installation twin body and the installation evaluation model to perform digital twin simulation and iterative optimization on the installation adjustment space, obtains an optimal installation adjustment optimization strategy, and completes adaptive installation adjustment of the bevel gear pair according to the strategy. The technical problems of being unable to adapt to dynamic error evolution and being difficult to guarantee adjustment precision and stability existing in the prior art bevel gear pair installation distance adjustment are solved, and the technical effects of dynamically adapting to error evolution, improving installation adjustment precision and guaranteeing long-term running stability are achieved.

[0005] The application provides a bevel gear pair adaptive adjustment method combined with installation distance deviation, including: acquiring real-time installation characteristic data of a bevel gear pair, and performing installation distance deviation adjustment analysis according to an installation design scheme and the real-time installation characteristic data to acquire an installation adjustment initial scheme; performing multi-dimensional implicit installation error trend prediction according to basic characteristic data of the bevel gear pair and the real-time installation characteristic data to construct an implicit installation error trend map; performing implicit installation error suppression compensation on the installation adjustment initial scheme according to the implicit installation error trend map to acquire an installation adjustment space; constructing a real-time installation twin and a bevel gear pair installation evaluation model; performing installation evaluation iteration optimization under digital twinning according to the real-time installation twin and the bevel gear pair installation evaluation model to acquire an installation adjustment optimization strategy, and performing bevel gear pair adaptive installation adjustment according to the installation adjustment optimization strategy.

[0006] In a possible implementation, according to the installation design scheme and the real-time installation characteristic data, the installation distance deviation adjustment analysis is performed to acquire the installation adjustment initial scheme, and the following processing is performed: the installation design scheme is analyzed to determine a target installation distance parameter and an allowed assembly stable interval; the real-time installation characteristic data is identified according to the target installation distance parameter to acquire an installation distance deviation vector; the installation distance deviation vector is tolerant optimized according to the allowed assembly stable interval to acquire a tolerant optimization deviation vector; and the bevel gear pair is subjected to installation distance deviation adjustment decision according to the tolerant optimization deviation vector to generate the installation adjustment initial scheme.

[0007] In a possible implementation, according to the basic characteristic data of the bevel gear pair and the real-time installation characteristic data, the multi-dimensional implicit installation error trend prediction is performed to construct the implicit installation error trend map, and the following processing is performed: the basic characteristic data and the real-time installation characteristic data are used to perform housing deformation installation error trend prediction to acquire a housing deformation installation error path; the basic characteristic data and the real-time installation characteristic data are used to perform thermal deformation installation error trend prediction to acquire a thermal deformation installation error path; and the basic characteristic data and the real-time installation characteristic data are used to perform bearing pre-tightening attenuation installation error trend prediction to acquire a bearing pre-tightening attenuation installation error path; the housing deformation installation error path, the thermal deformation installation error path and the bearing pre-tightening attenuation installation error path are subjected to time dimension alignment and action direction unification to generate the implicit installation error trend map.

[0008] In a possible implementation, based on the basic characteristic data and the real-time installation characteristic data, the shell deformation installation error trend is predicted to obtain the shell deformation installation error path, and the following processing is performed: based on the basic characteristic data and the real-time installation characteristic data, the shell elastic deformation under load is predicted to obtain the shell deformation trend characteristics; based on the shell deformation trend characteristics, the axial movement influence of the gear is identified to obtain the axial movement influence characteristics; based on the shell deformation trend characteristics, the bevel gear pair tilt influence is identified to obtain the tilt influence characteristics; based on the shell deformation trend characteristics, the meshing point offset influence is identified to obtain the meshing point offset influence characteristics; the coupling relationship of the shell deformation trend characteristics, the axial movement influence characteristics, the tilt influence characteristics, and the meshing point offset influence characteristics is sorted out to generate the shell deformation installation error path.

[0009] In a possible implementation, a real-time installation twin and a bevel gear pair installation evaluation model are constructed, and the following processes are performed: loading the installation state history set, the installation stability evaluation history set, and the transmission error evaluation history set of the bevel gear pair; using the installation state history set of the bevel gear pair as input data and the installation stability evaluation history set as output data, the installation stability evaluation model is trained; based on the installation state history set of the bevel gear pair and the transmission error evaluation history set, the transmission error evaluation model is trained; the installation stability evaluation model and the transmission error evaluation model are connected as parallel nodes to generate the bevel gear pair installation evaluation model.

[0010] In a possible implementation, the installation adjustment space is iteratively optimized under a digital twin based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an installation adjustment optimization strategy. The following processes are then performed: The installation adjustment space is optimized using a digital twin installation evaluation based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an optimized installation adjustment space; weights are assigned to the multi-dimensional installation evaluation indicators of the bevel gear pair installation evaluation model to obtain an installation quality calculation model, where the multi-dimensional installation evaluation indicators include installation stability and transmission error; installation quality is analyzed in the optimized installation adjustment space based on the installation quality calculation model to obtain an installation quality distribution; and iterative optimization is performed on the optimized installation adjustment space based on the installation quality distribution to generate the installation adjustment optimization strategy.

[0011] In a possible implementation, the installation adjustment space is optimized using a twin installation evaluation based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an optimized installation adjustment space. The following processes are then performed: The f-th installation adjustment scheme (where f is a positive integer) is extracted from the installation adjustment space; the real-time installation twin is simulated and adjusted according to the f-th installation adjustment scheme to obtain the f-th simulated installation state data; the f-th simulated installation state data is input into the bevel gear pair installation evaluation model to obtain the f-th installation evaluation sequence; if the f-th installation evaluation sequence satisfies the installation evaluation constraints, the f-th installation adjustment scheme is added to the optimized installation adjustment space.

[0012] In a possible implementation, the following processing is performed: the basic characteristic data includes bevel gear pair structure data and bevel gear pair material data.

[0013] In a possible implementation, a real-time installation twin and a bevel gear pair installation evaluation model are constructed, and the following processing is performed: digital twin modeling is performed based on the basic characteristic data and the real-time installation characteristic data to generate the real-time installation twin.

[0014] This application also provides an adaptive adjustment system for bevel gear pairs incorporating installation distance deviation, comprising: an installation distance deviation adjustment analysis module for acquiring real-time installation characteristic data of the bevel gear pair, and performing installation distance deviation adjustment analysis based on the installation design scheme and the real-time installation characteristic data to obtain an initial installation adjustment scheme; a multi-dimensional implicit installation error trend prediction module for predicting multi-dimensional implicit installation error trends based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data to construct an implicit installation error trend map; an implicit installation error suppression and compensation module for performing implicit installation error suppression and compensation on the initial installation adjustment scheme based on the implicit installation error trend map to obtain an installation adjustment space; a real-time installation twin construction module for constructing a real-time installation twin and a bevel gear pair installation evaluation model; and a bevel gear pair adaptive installation adjustment module for performing iterative optimization of the installation adjustment space under digital twin based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an installation adjustment optimization strategy, and executing adaptive installation adjustment of the bevel gear pair according to the installation adjustment optimization strategy.

[0015] The proposed adaptive adjustment method and system for bevel gear pairs, incorporating installation distance deviation, firstly acquires real-time installation characteristic data of the bevel gear pair. Based on the installation design scheme and the real-time installation characteristic data, it analyzes the installation distance deviation to obtain an initial installation adjustment scheme. Next, based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data, it predicts multi-dimensional implicit installation error trends, constructing an implicit installation error trend map. Then, based on the implicit installation error trend map, it performs implicit installation error suppression compensation on the initial installation adjustment scheme to obtain the installation adjustment space. Furthermore, it constructs a real-time installation twin and a bevel gear pair installation evaluation model. Finally, based on the real-time installation twin and the bevel gear pair installation evaluation model, iterative optimization of the installation adjustment space under digital twin conditions is performed to obtain an installation adjustment optimization strategy. Adaptive installation adjustment of the bevel gear pair is then executed according to this optimization strategy. Through the above process, the proposed method and system achieve the technical effects of dynamically adapting to error evolution, improving installation adjustment accuracy, and ensuring long-term operational stability. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a flowchart illustrating the adaptive adjustment method for bevel gear pairs incorporating installation distance deviation, provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the structure of the adaptive adjustment system for bevel gear pairs that incorporates installation distance deviation, provided in an embodiment of this application.

[0019] Explanation of reference numerals in the attached figures: 10 for installation distance deviation adjustment analysis module, 20 for multi-dimensional implicit installation error trend prediction module, 30 for implicit installation error suppression and compensation module, 40 for real-time installation twin construction module, and 50 for bevel gear pair adaptive installation adjustment module. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] This application provides an embodiment of an adaptive adjustment method for bevel gear pairs that incorporates installation distance deviation, such as... Figure 1 As shown, the method includes: Step S100: Obtain real-time installation characteristic data of the bevel gear pair, and analyze the installation distance deviation adjustment based on the installation design scheme and the real-time installation characteristic data to obtain the initial installation adjustment scheme.

[0022] Specifically, the target installation distance parameters and allowable assembly stability range are extracted from the installation design documents. For example, the target installation distance parameters include the theoretical axial installation distance between the driving and driven wheels, with a stability range of ±0.15mm. Real-time installation characteristic data, including the axial position of the driving and driven wheels and the runout of the bearing housing end face, are collected using measuring equipment such as laser displacement sensors and dial indicators. The difference between the target installation distance and the real-time installation distance is calculated to obtain the installation distance deviation vector. The deviation vector is then optimized with tolerance, truncating deviation components exceeding the allowable assembly stability range to the range boundary, resulting in a tolerance-optimized deviation vector. Based on the tolerance-optimized deviation vector, adjustment parameters such as the thickness of the adjusting shims and the number of turns of the bearing housing adjusting screws are determined through table lookup or linear interpolation to generate an initial installation adjustment scheme.

[0023] In one possible implementation, the installation distance deviation adjustment is analyzed based on the installation design scheme and the real-time installation characteristic data to obtain an initial installation adjustment scheme. Step S100 further includes step S110, analyzing the installation design scheme to determine the target installation distance parameter and the allowable assembly stability range. Specifically, the three-dimensional model parameters and assembly dimension chain in the installation design scheme are read, and the theoretical axial installation distance between the axis of the driving bevel gear and the axis of the driven bevel gear is extracted as the target installation distance parameter. At the same time, the reference end face position, positioning shoulder size, and bearing installation positioning size given in the design are read. Then, based on the gear material, module, tooth width, design load, and the elastic deformation limit of the housing, and the radial and axial stiffness values ​​of the bearing, the gear assembly verification calculation method is used to convert the allowable fluctuation range of backlash, the allowable range of contact area offset, and the allowable range of axial movement into the upper and lower limits of the axial installation distance, thereby forming a continuous allowable assembly stability range, that is, the installation distance range that can maintain stable meshing under machining and assembly errors and transmission loads. For example, if the target installation distance is set to 80 mm, the allowable assembly stability range is calculated to be 79.85 mm to 80.15 mm.

[0024] Step S120: Based on the target installation distance parameter, identify the installation distance deviation of the real-time installation characteristic data to obtain the installation distance deviation vector. Specifically, a laser displacement sensor or a high-precision inductive micrometer is used to collect the measured positions of the axial positioning end face of the driving gear, the axial positioning end face of the driven gear, and the intersection of the axes of the two gears. These real-time installation characteristic data are substituted into a preset installation distance geometric calculation model to calculate the current actual installation distance. The axial installation distance deviation is then obtained by subtracting the target installation distance from the actual installation distance. At the same time, the deviation of the driving wheel skew angle and the deviation of the driven wheel skew angle are calculated by measuring the height difference between the left and right points of the bearing housing. The axial installation distance deviation, the deviation of the driving wheel skew angle, and the deviation of the driven wheel skew angle are combined in a fixed order to form an installation distance deviation vector containing multiple components. For example, if the target installation distance is 80 mm, the measured installation distance is 80.20 mm, the driving wheel skew angle is 0.02 degrees, and the driven wheel skew angle is 0.01 degrees, then the obtained installation distance deviation vector is: axial positive deviation 0.20 mm, driving wheel skew deviation 0.02 degrees, and driven wheel skew deviation 0.01 degrees.

[0025] Step S130: The installation distance deviation vector is optimized to a tolerance level based on the allowable assembly stability range to obtain a tolerance-optimized deviation vector. Specifically, the axial installation distance deviation, drive wheel skew deviation, and driven wheel skew deviation in the installation distance deviation vector are sequentially traversed. Each deviation component is compared with the lower and upper limits of the allowable assembly stability range. If the deviation component is less than the lower limit, it is assigned the lower limit value; if the deviation component is greater than the upper limit, it is assigned the upper limit value; if it is within the range, it remains unchanged. After completing the limit processing for all components, the vector dimension and order remain unchanged to obtain the tolerance-optimized deviation vector. For example, if the allowable assembly stability range is ±0.15 mm and the allowable skew range is ±0.03 degrees, the original installation distance deviation vector is 0.20 mm axially, 0.02 degrees for the drive wheel, and 0.01 degrees for the driven wheel. The axial 0.20 mm exceeds the upper limit of 0.15 mm, so it is corrected to 0.15 mm, while the other components remain unchanged. The final tolerance-optimized deviation vector is 0.15 mm axially, 0.02 degrees for the drive wheel, and 0.01 degrees for the driven wheel.

[0026] Step S140: Based on the tolerance-optimized deviation vector, an installation distance deviation adjustment decision is made for the bevel gear pair to generate the initial installation adjustment scheme. Specifically, a mapping relationship is established between the tolerance-optimized deviation vector and the installation adjustment actuator. The axial installation distance deviation directly corresponds to the increase or decrease in the thickness of the adjusting shim at the rear end of the driving wheel or driven wheel. A positive deviation reduces the shim thickness, while a negative deviation increases the shim thickness. The driving wheel skew deviation corresponds to the number of turns the adjusting screws on the left or right side of the bearing housing are screwed in or out. The driven wheel skew deviation corresponds to the adjustment amount at the corresponding position of the driven bearing housing. The calculation is performed according to a preset unit deviation corresponding to the adjustment amount coefficient. For example, an axial deviation of 0.15 mm corresponds to a shim thickness reduction of 0.15 mm, and a driving wheel skew of 0.02 degrees corresponds to a clockwise rotation of one-quarter turn of the left screw. Finally, all adjustment amounts are integrated into an initial installation adjustment scheme that includes the shim thickness adjustment value, the number of turns of each adjusting screw, and the positioning reference compensation amount.

[0027] Step S200: Based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data, perform multi-dimensional implicit installation error trend prediction and construct an implicit installation error trend map. The basic characteristic data includes the bevel gear pair structural data and the bevel gear pair material data.

[0028] Specifically, basic characteristic data such as the module, tooth width, shaft diameter, housing wall thickness, material elastic modulus, Poisson's ratio, coefficient of thermal expansion, bearing stiffness, and preload decay characteristics of the bevel gear pair are read from the design documents. At the same time, real-time installation characteristic data such as installation position, real-time support reaction force, real-time temperature rise, and real-time preload are collected through displacement sensors, temperature sensors, load sensors, and a preload monitoring module. Three types of physical driving methods are used to predict the installation error paths: housing deformation installation error path, thermal deformation installation error path, and bearing preload decay installation error path. Then, the three paths are uniformly interpolated to the same time step, and all error components are uniformly converted into axial installation distance equivalent error. Finally, the data are superimposed and integrated according to the time series to form a multi-dimensional implicit installation error trend map that includes time, error type, error amplitude, and error direction. For example, by sampling one point per second and continuously predicting for one hour, a complete trend map including axial equivalent error, skew error, and meshing offset error is formed.

[0029] In one possible implementation, multidimensional implicit installation error trend prediction is performed based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data to construct an implicit installation error trend map. Step S200 further includes step S210, which predicts the shell deformation installation error trend based on the basic characteristic data and the real-time installation characteristic data to obtain the shell deformation installation error path. Specifically, using shell material properties, structural dimensions, real-time support load, and real-time installation constraints as inputs, a combination of material mechanics deformation analysis and linear finite element static calculation is used. First, the nodal displacements and rotation angles of the bearing seat area and gear mounting hole area are calculated. Then, the local deformation of the shell is mapped to the axial movement, axis misalignment, and meshing point offset of the driving gear and driven gear. The calculation is iteratively performed step by step according to the load step or time step to obtain a series of equivalent installation distance deviations, axis misalignment deviations, and meshing offset deviations at continuous moments. These deviations are arranged in time sequence to form a continuously changing shell deformation installation error path. For example, under a constant load of 10000 N, the axial equivalent error sequence caused by shell deformation at 0 seconds, 5 seconds, 10 seconds, and up to the stable moment is obtained.

[0030] Step S220: Based on the basic characteristic data and the real-time installation characteristic data, predict the trend of thermal deformation installation error and obtain the thermal deformation installation error path. Specifically, a one-dimensional heat conduction and lumped parameter thermal expansion model is used, with the real-time monitored ambient temperature, bearing friction temperature rise, and gear meshing temperature rise as inputs. Combined with the thermal expansion coefficient, geometric length, and assembly constraints of each component, calculate the axial thermal elongation and angular thermal deformation of the drive shaft, driven shaft, housing, and bearing seat. Then, convert the thermal deformation into equivalent installation distance change and axial misalignment change. Calculate the equivalent installation error of thermal deformation at each moment according to time steps to form a thermal deformation installation error path that changes synchronously with time and temperature. For example, during the process of rising from room temperature of 25 degrees Celsius to working temperature of 80 degrees Celsius, calculate the axial installation distance increment and axial misalignment increment caused by thermal deformation once per second.

[0031] Step S230: Based on the basic characteristic data and the real-time installation characteristic data, predict the bearing preload decay installation error trend to obtain the bearing preload decay installation error path. Specifically, adopt the bearing preload exponential decay model coupled with axial stiffness calculation method, input initial preload, bearing axial stiffness, real-time load, real-time temperature, and cumulative operating time, calculate the remaining preload at different times using the preload decay formula over time, and then convert the preload decay into shaft axial positioning deviation and installation distance deviation according to the correspondence between preload decrease and axial displacement. Calculate the continuous error data sequentially according to a fixed time step to form the bearing preload decay installation error path. For example, with an initial preload of 5000 N, predict the preload decay and equivalent installation distance deviation for each second from 0 to 3600 seconds.

[0032] Step S240 involves aligning the time dimension and unifying the action direction of the shell deformation installation error path, the thermal deformation installation error path, and the bearing preload attenuation installation error path to generate the implicit installation error trend map. Specifically, the three error paths are first uniformly interpolated into discrete sequences with the same time interval to ensure that the same moment corresponds to the same prediction time point. Then, all error components are uniformly converted into equivalent axial errors that directly affect the installation distance, and it is stipulated that the direction of increasing installation distance is positive and the direction of decreasing installation distance is negative to eliminate directional ambiguity. Then, at each moment, the three types of equivalent errors are algebraically superimposed to obtain the total implicit installation error at that moment. Finally, a multidimensional implicit installation error trend map is formed with time as the horizontal axis, equivalent installation error as the vertical axis, and error type as the classification dimension, which includes time, single-type error, total error, and change slope. For example, with a uniform sampling interval of 1 second and a prediction duration of 3600 seconds, continuous error trend data that can be used for subsequent compensation is formed.

[0033] In one possible implementation, the shell deformation installation error trend is predicted based on the basic characteristic data and the real-time installation characteristic data to obtain the shell deformation installation error path. Step S210 further includes step S211, predicting the elastic deformation of the shell under load based on the basic characteristic data and the real-time installation characteristic data to obtain the shell deformation trend characteristics. Specifically, based on the shell geometry, material elastic modulus, Poisson's ratio, and the real-time applied bearing support reaction force and gear meshing force, a shell stress deformation calculation model is established. The linear elastic deformation equation is used to solve for the axial displacement, radial displacement, and end face rotation of key areas such as gear shaft holes, bearing seats, and positioning end faces on the shell. At the same time, the linear change law of deformation with increasing load and the convergence characteristics that tend to stabilize are recorded. Parameters such as key position displacement, rotation angle, deformation gradient, and stabilization time are integrated into the shell deformation trend characteristics. For example, as the load gradually increases from 0 to 10000 N, the axial displacement of the bearing seat increases linearly from 0 to 0.08 mm and tends to stabilize.

[0034] Step S212: Identify the influence of gear axial movement based on the housing deformation trend characteristics to obtain axial movement influence features. Specifically, extract the axial displacement of the driving gear bearing seat and the axial displacement of the driven gear bearing seat from the housing deformation trend characteristics, calculate the relative displacement difference between the two, and directly equate this relative displacement to the change in installation distance. Establish a quantitative mapping relationship between the housing deformation axial displacement and the installation distance deviation to form axial movement influence features. For example, if the housing deformation causes the driving gear bearing seat to move axially inward by 0.05 mm and the driven gear bearing seat to move axially outward by 0.03 mm, then the relative axial movement between the two is 0.08 mm, corresponding to an increase in installation distance of 0.08 mm.

[0035] Step S213: Identify the tilting effect of the bevel gear pair based on the deformation trend characteristics of the housing, and obtain the tilting effect characteristics. Specifically, extract the height difference and end face angle of the left and right bearing seats from the deformation trend characteristics of the housing, convert the angle into the tilt angle of the gear axis, and then calculate the equivalent additional deviation of the installation distance and the change in meshing backlash due to the tilt of the axis based on the geometric relationship between the gear pitch cone angle, shaft intersection angle and installation distance, forming the tilting effect characteristics. For example, if the housing deformation causes the bearing seat end face to tilt by 0.015 degrees, the corresponding gear axis tilt is 0.015 degrees, which is equivalent to an additional deviation of 0.03 mm in the installation distance.

[0036] Step S214: Identify the impact of meshing point offset based on the deformation trend characteristics of the housing, and obtain the influence characteristics of meshing point offset. Specifically, using the axial movement and axis tilt angle caused by housing deformation as inputs, combined with the bevel gear pair cone parameters, module, tooth width, and theoretical meshing center position, the offset distance and direction of the actual meshing point relative to the theoretical meshing point are calculated through spatial meshing geometry analysis. The meshing point offset is converted into the equivalent installation distance deviation and transmission error increment, forming the influence characteristics of meshing point offset. For example, if the meshing point offsets by 0.03 mm along the tooth width, the corresponding equivalent installation distance deviation increases by 0.015 mm.

[0037] Step S215 involves analyzing the coupling relationships among the shell deformation trend characteristics, axial movement influence characteristics, tilt influence characteristics, and engagement point offset influence characteristics to generate the shell deformation installation error path. Specifically, the axial movement influence characteristics, tilt influence characteristics, and engagement point offset influence characteristics are superimposed according to spatial geometric coupling relationships. The total shell deformation equivalent installation error is equal to the sum of the installation distance deviation caused by axial movement, the equivalent installation distance deviation caused by axial tilt, and the equivalent installation distance deviation caused by engagement point offset. Using time or load step as the independent variable, the total equivalent installation error is calculated point by point, forming a continuous error sequence from the initial state to the stable deformation state, which is the shell deformation installation error path. For example, a set of total errors is calculated every 0.1 seconds, and this is continuously recorded until the deformation stabilizes, forming a complete path.

[0038] Step S300: Perform hidden installation error suppression compensation on the initial installation adjustment scheme according to the hidden installation error trend map to obtain the installation adjustment space.

[0039] Specifically, the maximum equivalent axial installation error, the time of error occurrence, and the direction of error change within the prediction period are extracted from the implicit installation error trend map. The required pre-compensation amount and direction are determined so that the initial adjustment amount and the compensation amount can be superimposed to maintain the installation distance within the allowable assembly stability range throughout the entire operation period. Then, with the compensated target adjustment amount as the center, the adjustment is expanded in the positive and negative directions according to the minimum adjustment step size, such as 0.01 mm. At the same time, invalid schemes that exceed the design allowable range and will cause side clearance to exceed the limit or contact area offset to exceed the standard are eliminated. Finally, an installation adjustment space composed of multiple discrete and feasible installation adjustment schemes is formed. For example, if the target shim adjustment amount after compensation is a reduction of 0.15 mm, 11 sets of feasible adjustment schemes ranging from a reduction of 0.10 mm to a reduction of 0.20 mm are generated with a step size of 0.01 mm to form the installation adjustment space.

[0040] Step S400: Construct a real-time installation evaluation model for the twin and bevel gear pair.

[0041] Specifically, a real-time installation twin is constructed based on the geometric structure, material properties, assembly constraints, real-time installation position, and real-time load and temperature data of the bevel gear pair, with synchronized updates of geometric and mechanical parameters. This achieves a one-to-one mapping from the physical installation state to the digital model. Next, historical installation state data and corresponding installation stability indicators such as vibration amplitude, noise level, bearing temperature rise, and transmission error indicators such as rotation angle error and meshing impact are collected. These are used to train dedicated models that can output stability scores and transmission error prediction values. Finally, the two types of models are integrated in parallel to form a unified bevel gear pair installation evaluation model, which is used to perform multi-index quantitative evaluation of the installation state obtained from the twin simulation.

[0042] In one possible implementation, a real-time installation twin and a bevel gear pair installation evaluation model are constructed. Step S400 further includes step S410, which involves performing digital twin modeling based on the basic characteristic data and the real-time installation characteristic data to generate the real-time installation twin. Specifically, a parametric three-dimensional geometric model and a simplified finite element mechanical model are constructed based on basic characteristic data such as gear module, number of teeth, shaft diameter, housing structure, and bearing type. Constraint relationships for shaft axial positioning, bearing preload, housing support, and gear meshing are established. Then, real-time installation characteristic data such as axial displacement, bearing seat tilt angle, preload force, temperature, and load are accessed in real time through a data interface. The positioning dimensions, preload state, temperature field, and load conditions in the model are updated at a fixed refresh cycle, such as 100 milliseconds, so that the digital model and the physical bevel gear pair installation state remain consistent in real time, forming a real-time installation twin that can be directly used for adjustment simulation.

[0043] In one possible implementation, a real-time installation twin and bevel gear pair installation evaluation model are constructed. Step S400 further includes step S420, loading the installation status history set, installation stability evaluation history set, and transmission error evaluation history set of the bevel gear pair. Specifically, the historically stored installation status data, including shim thickness, bearing preload, shaft misalignment, and measured installation distance, are read through an industrial data interface. The installation stability evaluation data includes the effective value of vibration acceleration, noise pressure level, bearing temperature rise, and shaft movement, while the transmission error evaluation data includes unidirectional transmission error, bidirectional backlash error, and peak error. Missing values ​​are filled with linear interpolation, and outliers, such as abrupt changes that significantly exceed the physical range, are removed. All data are normalized to a uniform numerical range, such as 0 to 1, and divided into a model training set and a test set in an 8:2 ratio to complete the preparation of training samples.

[0044] Step S430: Using the historical set of installation states of the bevel gear pair as input data and the historical set of installation stability evaluation data as output data, train the installation stability evaluation model. Specifically, construct a fully connected feedforward neural network containing an input layer, two hidden layers, and an output layer. The input layer dimension is consistent with the installation state feature dimension. The hidden layers are set with 64 and 32 neurons respectively and use the ReLU activation function. The output layer has one neuron and uses Sigmoid activation to map the result to a stability score from 0 to 100. The mean squared error is used as the loss function, the Adam optimizer is used, the initial learning rate is set to 0.001, the batch size is set to 32, and the training epochs are set to 50. Iteratively update the weights on the training set and verify on the test set. When the prediction error is less than a preset threshold, such as 2%, stop training to obtain a usable installation stability evaluation model.

[0045] Step S440: Based on the historical set of installation states of the bevel gear pair and the historical set of transmission error evaluations, train the transmission error evaluation model. Specifically, a fully connected network or a lightweight LSTM network suitable for continuous numerical prediction is used. The input layer includes features such as installation state parameters, real-time load, and real-time temperature. The hidden layer has 64 neurons activated with tanh. The output layer directly predicts the physical value of the transmission error. The loss function is the mean absolute error, the Adam optimizer is used, the learning rate is set to 0.001, and the training epochs are set to 30. End-to-end training is performed using historical installation states and corresponding transmission error labels, enabling the model to output high-precision transmission error prediction values ​​under a given installation state.

[0046] Step S450: The installation stability evaluation model and the transmission error evaluation model are connected as parallel nodes to generate the bevel gear pair installation evaluation model. Specifically, a multi-task learning framework is adopted, and the input layers of the installation stability evaluation model and the transmission error evaluation model share features. After inputting the same set of installation state feature vectors, they enter their respective independent network branches. One branch outputs an installation stability score from 0 to 100, and the other branch outputs the transmission error prediction value. After training, the weights of the two branches are frozen and encapsulated into a unified inference interface. The input is the installation state data, and the output is a two-dimensional evaluation sequence containing the stability score and the transmission error, forming a bevel gear pair installation evaluation model that can be directly used for simulation optimization.

[0047] Step S500: Based on the real-time installation twin and the bevel gear pair installation evaluation model, perform iterative optimization of the installation adjustment space under digital twin to obtain the installation adjustment optimization strategy, and execute adaptive installation adjustment of the bevel gear pair according to the installation adjustment optimization strategy.

[0048] Specifically, the process involves traversing all feasible solutions in the installation adjustment space, sequentially performing parameter adjustments in the real-time installation twin, and simulating the corresponding installation state. The simulation state is then input into the bevel gear pair installation evaluation model to obtain stability scores and transmission errors. Solutions that meet the stability and transmission error constraints are selected and added to the optimization candidate set. The candidate solutions are then ranked using a weighted installation quality calculation model, and the solution with the highest installation quality value is selected as the optimal solution. This forms an installation adjustment optimization strategy that includes shim thickness, screw adjustment amount, and positioning compensation amount. Finally, automatic installation adjustment is completed through a servo adjustment mechanism or an adjustable shim actuator.

[0049] In one possible implementation, the installation adjustment space is iteratively optimized under a digital twin based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an installation adjustment optimization strategy. Step S500 further includes step S510, which involves performing a twin installation evaluation optimization on the installation adjustment space based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an optimized installation adjustment space. Specifically, each set of adjustment parameters in the installation adjustment space is read sequentially, and the corresponding parameters such as shim thickness, axial positioning, and preload are modified in the real-time installation twin, and mechanical and meshing simulations are completed. The simulated installation state data is output, and this data is input into the bevel gear pair installation evaluation model to obtain a stability score and a transmission error prediction value. The two indicators are compared with preset constraint thresholds, such as a stability score ≥ 90 points and a transmission error ≤ 0.03 mm. Schemes that meet all constraints are retained and added to the optimized installation adjustment space, while those that do not meet the constraints are directly eliminated, ultimately forming an optimization space containing only qualified and feasible schemes.

[0050] Step S520 involves assigning weights to the multi-dimensional installation evaluation indicators of the bevel gear pair installation evaluation model to obtain an installation quality calculation model. The multi-dimensional installation evaluation indicators include installation stability and transmission error. Specifically, based on the priority of the bevel gear pair assembly project, the weights for installation stability and transmission error are determined using the analytic hierarchy process (AHP). The transmission error is mapped to a score from 0 to 100 through a linear transformation. Then, the installation quality value is calculated using a weighted summation formula. The installation quality value equals the installation stability weight multiplied by the stability score, plus the transmission error weight multiplied by the transmission error mapping score, forming a quantifiable, comparable, and optimal single-dimensional installation quality calculation model.

[0051] Step S530: Analyze the installation quality of the optimal installation adjustment space according to the installation quality calculation model to obtain the installation quality distribution. Specifically, traverse each grid installation adjustment scheme in the optimal installation adjustment space, sequentially substitute the corresponding installation evaluation sequence into the installation quality calculation model, calculate a unique installation quality value for each, and then match all schemes with their quality values ​​one-to-one and sort them from largest to smallest quality value to form a complete installation quality distribution table containing scheme number, adjustment parameters, stability score, transmission error, and installation quality value.

[0052] Step S540: Iteratively optimize the installation adjustment space based on the installation quality distribution to generate the installation adjustment optimization strategy. Specifically, read the set of adjustment parameters with the largest installation quality value from the installation quality distribution, including axial shim adjustment thickness, number of adjustment turns of the drive wheel bearing seat, adjustment amount of the driven wheel bearing seat, and positioning end face compensation amount, etc., and convert these parameters into control commands recognizable by the actuator, such as servo motor rotation angle, shim replacement specifications, and positioning compensation dimensions, and integrate them into a complete installation adjustment optimization strategy that includes the adjustment object, adjustment direction, adjustment amount, and execution sequence.

[0053] In one possible implementation, the installation adjustment space is optimized using a twin installation evaluation based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain the optimized installation adjustment space. Step S510 further includes step S511, extracting the f-th installation adjustment scheme based on the installation adjustment space, where f is a positive integer. Specifically, all feasible adjustment schemes within the installation adjustment space are sorted by adjustment amount from smallest to largest and a continuous integer index is established, with each index corresponding to f. f starts from 1 and increments sequentially. Based on the index f, all adjustment parameters of the corresponding scheme are directly read, including the axial shim adjustment amount, the driving wheel bearing seat adjustment amount, the driven wheel bearing seat adjustment amount, etc., forming a complete and executable f-th installation adjustment scheme for twin simulation and evaluation.

[0054] Step S512: Perform simulation adjustment on the real-time installation twin according to the f-th installation adjustment scheme to obtain the f-th simulated installation state data. Specifically, map the shim thickness change, bearing housing tilt angle adjustment, axial positioning compensation, etc. in the f-th installation adjustment scheme to the geometric and mechanical parameters of the real-time installation twin, update the gear axial position, axis tilt angle, bearing preload, housing constraint state, etc., run static and assembly geometry simulation, and calculate the equivalent installation distance, axis misalignment angle, meshing offset, support reaction force, predicted deformation, etc., as input to the evaluation model.

[0055] Step S513: Input the f-th simulated installation state data into the bevel gear pair installation evaluation model to obtain the f-th installation evaluation sequence. Specifically, the equivalent installation distance, axis misalignment, preload, temperature, load, and other features in the f-th simulated installation state data are normalized and dimensionally aligned according to the model requirements, and input into the shared input layer of the bevel gear pair installation evaluation model. The stability evaluation branch outputs a stability score from 0 to 100, and the transmission error evaluation branch outputs the corresponding transmission error value. The two outputs are combined in a fixed order to form the f-th installation evaluation sequence containing the stability evaluation value and the transmission error evaluation value.

[0056] Step S514: If the f-th installation evaluation sequence satisfies the installation evaluation constraints, the f-th installation adjustment scheme is added to the optimization installation adjustment space. Specifically, preset installation evaluation constraints include a stability score ≥ 90 points and a transmission error ≤ 0.03 mm. The stability score and transmission error in the f-th installation evaluation sequence are compared with the above thresholds. When the stability score ≥ 90 and the transmission error ≤ 0.03 mm, it is judged as qualified, and the f-th installation adjustment scheme is added to the optimization installation adjustment space. If any one of the constraints is not met, it is judged as unqualified, and the scheme is directly discarded and does not enter the subsequent optimization stage.

[0057] This application embodiment acquires real-time installation characteristic data of the bevel gear pair, analyzes the installation distance deviation adjustment based on the installation design scheme, and obtains an initial installation adjustment scheme. Based on the basic characteristics of the bevel gear pair and the real-time installation characteristics, it predicts the trend of multi-dimensional implicit installation errors and constructs an implicit installation error trend map. Based on this trend map, it performs error suppression compensation on the initial adjustment scheme to form a feasible installation adjustment space. It constructs a real-time installation twin synchronized with the physical installation state and trains the bevel gear pair installation evaluation model. It uses the real-time installation twin and the installation evaluation model to perform digital twin simulation and iterative optimization on the installation adjustment space to obtain the optimal installation adjustment optimization strategy. Based on this strategy, it completes the adaptive installation adjustment of the bevel gear pair and other technical means, which solves the technical problems of existing bevel gear pair installation distance adjustment that cannot adapt to dynamic error evolution and cannot guarantee adjustment accuracy and stability. It achieves the technical effects of dynamically adapting to error evolution, improving installation adjustment accuracy, and ensuring long-term operational stability.

[0058] In the above text, refer to Figure 1 The adaptive adjustment method for bevel gear pairs incorporating installation distance deviation according to embodiments of the present invention is described in detail. Next, reference will be made to... Figure 2 An adaptive adjustment system for bevel gear pairs incorporating installation distance deviation is described according to an embodiment of the present invention.

[0059] The bevel gear pair adaptive adjustment system based on installation distance deviation, according to an embodiment of the present invention, solves the technical problems of existing bevel gear pair installation distance adjustment systems, such as the inability to adapt to dynamic error evolution and the difficulty in guaranteeing adjustment accuracy and stability. It achieves the technical effects of dynamically adapting to error evolution, improving installation adjustment accuracy, and ensuring long-term operational stability. The bevel gear pair adaptive adjustment system based on installation distance deviation includes: an installation distance deviation adjustment analysis module 10, a multi-dimensional implicit installation error trend prediction module 20, an implicit installation error suppression and compensation module 30, a real-time installation twin construction module 40, and a bevel gear pair adaptive installation adjustment module 50.

[0060] The installation distance deviation adjustment analysis module 10 is used to acquire real-time installation characteristic data of the bevel gear pair, and perform installation distance deviation adjustment analysis based on the installation design scheme and the real-time installation characteristic data to obtain an initial installation adjustment scheme; the multi-dimensional implicit installation error trend prediction module 20 is used to predict the multi-dimensional implicit installation error trend based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data to construct an implicit installation error trend map; the implicit installation error suppression compensation module 30 is used to perform implicit installation error suppression compensation on the initial installation adjustment scheme based on the implicit installation error trend map to obtain the installation adjustment space; the real-time installation twin construction module 40 is used to construct a real-time installation twin and a bevel gear pair installation evaluation model; the bevel gear pair adaptive installation adjustment module 50 is used to perform iterative optimization of installation evaluation under digital twin based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an installation adjustment optimization strategy, and execute adaptive installation adjustment of the bevel gear pair according to the installation adjustment optimization strategy.

[0061] The detailed description of the specific configuration of the installation distance deviation adjustment and analysis module 10 is explained as follows: As mentioned above, the installation distance deviation adjustment and analysis is performed based on the installation design scheme and the real-time installation characteristic data to obtain an initial installation adjustment scheme. The installation distance deviation adjustment and analysis module 10 may further include: an installation design scheme analysis unit for analyzing the installation design scheme and determining the target installation distance parameter and the allowable assembly stability range; an installation distance deviation identification unit for identifying the installation distance deviation based on the target installation distance parameter and the real-time installation characteristic data to obtain an installation distance deviation vector; a tolerance optimization unit for performing tolerance optimization on the installation distance deviation vector based on the allowable assembly stability range to obtain a tolerance optimization deviation vector; and an installation distance deviation adjustment decision unit for making an installation distance deviation adjustment decision on the bevel gear pair based on the tolerance optimization deviation vector to generate the initial installation adjustment scheme.

[0062] The detailed configuration of the multidimensional implicit installation error trend prediction module 20 is explained as follows: As mentioned above, multidimensional implicit installation error trend prediction is performed based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data to construct an implicit installation error trend map. The multidimensional implicit installation error trend prediction module 20 may further include: a housing deformation installation error trend prediction unit for predicting the housing deformation installation error trend based on the basic characteristic data and the real-time installation characteristic data to obtain the housing deformation installation error path; a thermal deformation installation error trend prediction unit for predicting the thermal deformation installation error trend based on the basic characteristic data and the real-time installation characteristic data to obtain the thermal deformation installation error path; a bearing preload attenuation installation error trend prediction unit for predicting the bearing preload attenuation installation error trend based on the basic characteristic data and the real-time installation characteristic data to obtain the bearing preload attenuation installation error path; and an action direction unification unit for aligning the housing deformation installation error path, the thermal deformation installation error path, and the bearing preload attenuation installation error path in terms of time dimension and action direction to generate the implicit installation error trend map.

[0063] The shell deformation installation error trend prediction unit, which predicts the shell deformation installation error path based on the basic characteristic data and the real-time installation characteristic data, may further include: a shell elastic deformation prediction subunit for predicting shell elastic deformation under load based on the basic characteristic data and the real-time installation characteristic data, and obtaining shell deformation trend characteristics; a gear axial movement influence identification subunit for identifying gear axial movement influence based on the shell deformation trend characteristics, and obtaining axial movement influence characteristics; a bevel gear pair tilt influence identification subunit for identifying bevel gear pair tilt influence based on the shell deformation trend characteristics, and obtaining tilt influence characteristics; a meshing point offset influence identification subunit for identifying meshing point offset influence based on the shell deformation trend characteristics, and obtaining meshing point offset influence characteristics; and a coupling relationship sorting subunit for sorting the coupling relationship between the shell deformation trend characteristics, the axial movement influence characteristics, the tilt influence characteristics, and the meshing point offset influence characteristics, and generating the shell deformation installation error path.

[0064] The detailed description of the specific configuration of the real-time installation twin construction module 40 is explained as follows: As mentioned above, the real-time installation twin and bevel gear pair installation evaluation model are constructed. The real-time installation twin construction module 40 may further include: a history set loading unit for loading the installation state history set, installation stability evaluation history set, and transmission error evaluation history set of the bevel gear pair; an installation stability evaluation model training unit for training the installation stability evaluation model with the installation state history set of the bevel gear pair as input data and the installation stability evaluation history set as output data; a transmission error evaluation model training unit for training the transmission error evaluation model based on the installation state history set of the bevel gear pair and the transmission error evaluation history set; and a bevel gear pair installation evaluation model generation unit for connecting the installation stability evaluation model and the transmission error evaluation model as parallel nodes to generate the bevel gear pair installation evaluation model.

[0065] The specific configuration of the bevel gear pair adaptive installation adjustment module 50 is described in detail below: As mentioned above, the installation adjustment space is iteratively optimized under a digital twin based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an installation adjustment optimization strategy. The bevel gear pair adaptive installation adjustment module 50 may further include: a twin installation evaluation optimization unit for performing twin installation evaluation optimization on the installation adjustment space based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an optimized installation adjustment space; a weight allocation unit for allocating weights to the multi-dimensional installation evaluation indicators of the bevel gear pair installation evaluation model to obtain an installation quality calculation model, wherein the multi-dimensional installation evaluation indicators include installation stability and transmission error; an installation quality analysis unit for performing installation quality analysis on the optimized installation adjustment space based on the installation quality calculation model to obtain an installation quality distribution; and an iterative optimization unit for iteratively optimizing the optimized installation adjustment space based on the installation quality distribution to generate the installation adjustment optimization strategy.

[0066] Specifically, the installation adjustment space is optimized using a twin installation evaluation based on the real-time installation twin and the bevel gear pair installation evaluation model to obtain an optimized installation adjustment space. The twin installation evaluation optimization unit may further include: an f-th installation adjustment scheme extraction subunit for extracting the f-th installation adjustment scheme from the installation adjustment space, where f is a positive integer; a simulation adjustment subunit for performing simulation adjustment on the real-time installation twin based on the f-th installation adjustment scheme to obtain the f-th simulated installation state data; a bevel gear pair installation evaluation subunit for inputting the f-th simulated installation state data into the bevel gear pair installation evaluation model to obtain the f-th installation evaluation sequence; and an optimized installation adjustment space acquisition subunit for adding the f-th installation adjustment scheme to the optimized installation adjustment space if the f-th installation evaluation sequence satisfies the installation evaluation constraints.

[0067] The multidimensional hidden installation error trend prediction module 20 may further include: the basic characteristic data includes bevel gear pair structure data and bevel gear pair material data.

[0068] The real-time installation twin and bevel gear pair installation evaluation model are constructed. The real-time installation twin construction module 40 may further include: a digital twin modeling unit for performing digital twin modeling based on the basic characteristic data and the real-time installation characteristic data to generate the real-time installation twin.

[0069] The adaptive adjustment system for bevel gear pairs combined with installation distance deviation provided in this embodiment of the invention can execute the adaptive adjustment method for bevel gear pairs combined with installation distance deviation provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0070] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0071] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for adaptive adjustment of bevel gear pairs based on installation distance deviation, characterized in that, The method includes: Obtain real-time installation characteristic data of the bevel gear pair, and analyze the installation distance deviation adjustment based on the installation design scheme and the real-time installation characteristic data to obtain the initial installation adjustment scheme; Based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data, a multi-dimensional implicit installation error trend prediction is performed, and an implicit installation error trend map is constructed. Based on the hidden installation error trend map, the initial installation adjustment scheme is subjected to hidden installation error suppression compensation to obtain the installation adjustment space; Construct a real-time installation evaluation model for twins and bevel gear pairs; Based on the real-time installation twin and the bevel gear pair installation evaluation model, the installation adjustment space is iteratively optimized under digital twin to obtain the installation adjustment optimization strategy, and the bevel gear pair adaptive installation adjustment is executed according to the installation adjustment optimization strategy.

2. The adaptive adjustment method for bevel gear pairs combining installation distance deviation as described in claim 1, characterized in that, Based on the installation design scheme and the real-time installation characteristic data, the installation distance deviation adjustment analysis is performed to obtain the initial installation adjustment scheme, including: Analyze the installation design scheme to determine the target installation distance parameters and the allowable assembly stability range; Based on the target installation distance parameter, the installation distance deviation of the real-time installation characteristic data is identified to obtain the installation distance deviation vector. Based on the allowable assembly stability range, the installation distance deviation vector is optimized to obtain the tolerant optimized deviation vector. Based on the tolerance-optimized deviation vector, the installation distance deviation adjustment decision is made for the bevel gear pair, and the initial installation adjustment scheme is generated.

3. The adaptive adjustment method for bevel gear pairs combined with installation distance deviation as described in claim 1, characterized in that, Based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data, a multi-dimensional implicit installation error trend prediction is performed, and an implicit installation error trend map is constructed, including: Based on the basic characteristic data and the real-time installation characteristic data, the shell deformation installation error trend is predicted, and the shell deformation installation error path is obtained. Based on the basic characteristic data and the real-time installation characteristic data, the trend of thermal deformation installation error is predicted, and the path of thermal deformation installation error is obtained. Based on the basic characteristic data and the real-time installation characteristic data, the bearing preload attenuation installation error trend is predicted, and the bearing preload attenuation installation error path is obtained. The time dimension of the installation error path of the housing deformation, the installation error path of the thermal deformation, and the installation error path of the bearing preload attenuation are aligned and the direction of action is unified to generate the hidden installation error trend map.

4. The adaptive adjustment method for bevel gear pairs combined with installation distance deviation as described in claim 3, characterized in that, Based on the basic characteristic data and the real-time installation characteristic data, the shell deformation installation error trend is predicted, and the shell deformation installation error path is obtained, including: Based on the basic characteristic data and the real-time installation characteristic data, the elastic deformation of the shell after loading is predicted, and the deformation trend characteristics of the shell are obtained. Based on the deformation trend characteristics of the housing, the influence of gear axial movement is identified, and the axial movement influence characteristics are obtained. Based on the deformation trend characteristics of the shell, the tilting effect of the bevel gear pair is identified, and the tilting effect characteristics are obtained. Based on the deformation trend characteristics of the shell, the influence of the meshing point offset is identified, and the influence characteristics of the meshing point offset are obtained. The coupling relationship between the shell deformation trend characteristics, the axial movement influence characteristics, the tilt influence characteristics, and the meshing point offset influence characteristics is analyzed to generate the shell deformation installation error path.

5. The adaptive adjustment method for bevel gear pairs combined with installation distance deviation as described in claim 1, characterized in that, Construct a real-time installation evaluation model for the twin and bevel gear pair, including: Load the installation status history set, installation stability evaluation history set, and transmission error evaluation history set of the bevel gear pair; Using the installation status history set of the bevel gear pair as input data and the installation stability evaluation history set as output data, the installation stability evaluation model is trained. Based on the installation status history set of the bevel gear pair and the transmission error evaluation history set, train the transmission error evaluation model; The installation stability evaluation model and the transmission error evaluation model are connected as parallel nodes to generate the bevel gear pair installation evaluation model.

6. The adaptive adjustment method for bevel gear pairs combining installation distance deviation as described in claim 1, characterized in that, Based on the real-time installation twin and the bevel gear pair installation evaluation model, the installation adjustment space is iteratively optimized under digital twin to obtain an installation adjustment optimization strategy, including: Based on the real-time installation twin and the bevel gear pair installation evaluation model, the installation adjustment space is evaluated and optimized using twin installation to obtain the optimized installation adjustment space; The installation evaluation indexes of the bevel gear pair installation evaluation model are weighted and the installation quality calculation model is obtained. The multi-dimensional installation evaluation indexes include installation stability and transmission error. Based on the installation quality calculation model, the installation quality of the optimal installation adjustment space is analyzed to obtain the installation quality distribution. The installation adjustment space is iteratively optimized based on the installation quality distribution to generate the installation adjustment optimization strategy.

7. The adaptive adjustment method for bevel gear pairs combined with installation distance deviation as described in claim 6, characterized in that, Based on the real-time installation twin and the bevel gear pair installation evaluation model, the installation adjustment space is evaluated and optimized using a twin installation method to obtain the optimized installation adjustment space, including: The f-th installation adjustment scheme is extracted based on the installation adjustment space, where f is a positive integer; The real-time installation twin is simulated and adjusted according to the f-th installation adjustment scheme to obtain the f-th simulated installation state data; Input the f-th simulated installation state data into the bevel gear pair installation evaluation model to obtain the f-th installation evaluation sequence; If the f-th installation evaluation sequence satisfies the installation evaluation constraints, the f-th installation adjustment scheme is added to the optimization installation adjustment space.

8. The adaptive adjustment method for bevel gear pairs combining installation distance deviation as described in claim 1, characterized in that, The basic characteristic data includes bevel gear pair structure data and bevel gear pair material data.

9. The adaptive adjustment method for bevel gear pairs combining installation distance deviation as described in claim 1, characterized in that, Construct a real-time installation evaluation model for the twin and bevel gear pair, including: Digital twin modeling is performed based on the basic characteristic data and the real-time installation characteristic data to generate the real-time installation twin.

10. A bevel gear pair adaptive adjustment system incorporating installation distance deviation, characterized in that, The system is used to implement the adaptive adjustment method for bevel gear pairs in conjunction with installation distance deviation as described in any one of claims 1-9, the system comprising: The installation distance deviation adjustment analysis module is used to obtain real-time installation characteristic data of the bevel gear pair, and to perform installation distance deviation adjustment analysis based on the installation design scheme and the real-time installation characteristic data to obtain the initial installation adjustment scheme. The multidimensional implicit installation error trend prediction module is used to predict the multidimensional implicit installation error trend based on the basic characteristic data of the bevel gear pair and the real-time installation characteristic data, and to construct an implicit installation error trend map. The implicit installation error suppression and compensation module is used to perform implicit installation error suppression and compensation on the initial installation adjustment scheme according to the implicit installation error trend map, and obtain the installation adjustment space; A real-time installation twin building module is used to build real-time installation twins and bevel gear pair installation evaluation models; The bevel gear pair adaptive installation adjustment module is used to perform iterative optimization of the installation evaluation under digital twin based on the real-time installation twin and the bevel gear pair installation evaluation model, obtain the installation adjustment optimization strategy, and execute the bevel gear pair adaptive installation adjustment according to the installation adjustment optimization strategy.

Images